You are on page 1of 121

Maternal and Child Nutrition June, 2013

www.thelancet.com

The Series identies a set of ten proven nutrition-specic interventions, which if scaled up from present population coverage to cover 90% of the need, would eliminate about 900 000 deaths of children younger than 5 years in the 34 high nutrition-burden countries where 90% of the worlds stunted children live.
Maternal and Child Nutrition

The LancetLondon 32 Jamestown Road, London NW1 7BY, UK T +44 (0)20 7424 4910 F +44 (0)20 7424 4911 The LancetNew York 360 Park Avenue South, New York, NY 100101710, USA T +1 212 633 3810 F +1 212 633 3853 The LancetBeijing Unit 16, 7F, Tower W1, Oriental Plaza, Beijing 100738, China T + 86 10 85208872 F + 86 10 85189297 editorial@lancet.com

Maternal and Child Nutrition June, 2013

Comment
1
R Horton, S Lo

Nutrition: a quintessential sustainable development goal Maternal and child nutrition: building momentum for impact Delivery platforms for sustained nutrition in Ethiopia Only collective action will end undernutrition Nutrition-sensitive food systems: from rhetoric to action Global child and maternal nutritionthe SUN rises Early nutrition and adult outcomes: pieces of the puzzle

2
Editor Richard Horton Deputy Editor Astrid James Senior Executive Editors Pam Das Sabine Kleinert William Summerskill Executive Editors Stephanie Clark Justine Davies Joanna Palmer Pia Pini Stuart Spencer Richard Turner Managing Editor Hannah Jones Web Editors Richard Lane Erika Niesner Senior Assistant Web Editor Katherine Rolfe Senior Editors Niall Boyce Audrey Ceschia Lin Guo Selina Lo Udani Samarasekera Onisillos Sekkides Asia Editor Helena Hui Wang (Beijing) North America Editor Rebecca Cooney (New York) Conference Editors Laura Hart Nicolai Humphreys Deputy Managing Editor Laura Benham Senior Assistant Editors Olaya Astudillo Stephanie Bartlett Abi Cantor Sean Cleghorn Tim Dehnel Dara Mohammadi Odhran ODonoghue Helen Penny Frances Whinder Farhat Yaqub Assistant Editors Neil Bennet Hannah Cagney Stephanie Clague Katherine Gourd Natalie Harrison Rebecca Heald Zena Nyakoojo Louise Rishton Media Relations Manager Daisy Barton Editorial Assistants Holly Baker Nawsheen Boodhun Nicolas Dolan Francesca Towey The Lancet is a registered trademark of Reed Elsevier Properties SA, used under licence.

Maternal and Child Nutrition Study Group F Lemma, J Matji

7 9 10 12

A Taylor and others

P Pinstrup-Andersen D Nabarro

Z A Bhutta

Series
15
R E Black and others

Maternal and child undernutrition and overweight in low-income and middle-income countries Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Nutrition-sensitive interventions and programmes: how can they help to accelerate progress in improving maternal and child nutrition? The politics of reducing malnutrition: building commitment and accelerating progress

40

Z A Bhutta and others

66

M T Ruel and others

82

S Gillespie and others

Articles
100 Mortality risk in preterm and small-for-gestational-age infants in low-income and middle-income countries: a pooled country analysis
J Katz and others

109 Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: findings from five birth cohort studies
L S Adair and others

Cover image Betty Press/Panos Previously published online See www.thelancet/com for WebExtra content Version verified by CrossMark
International Advisory Board Karen Antman (Boston) Valerie Beral (Oxford) Robert Beaglehole (Auckland) Anthony Costello (London) Robert Fletcher (Boston) Suzanne Fletcher (Boston)

Karen Gelmon (Vancouver) David Grimes (Durham) Ana Langer (Cambridge, MA) Judith Lumley (Melbourne) Elizabeth Molyneux (Blantyre) Christopher Murray (Seattle)

Alwyn Mwinga (Lusaka) Marie-Louise Newell (Somkhele) Magne Nylenna (Oslo) Peter Piot (London) Stuart Pocock (London) Giuseppe Remuzzi (Bergamo)

Caroline Savage (Birmingham) Ken Schulz (Chapel Hill) Frank Shann (Melbourne) Jan Vandenbroucke (Leiden) Cesar Victora (Pelotas) Nick White (Bangkok)

Comment

Nutrition: a quintessential sustainable development goal


In the final paper of our 2008 Lancet Series on maternal and child undernutrition, Saul Morris and colleagues wrote that, The international nutrition systemmade up of international and donor organisations, academia, civil society, and the private sectoris fragmented and dysfunctional.1 They concluded that, incredibly, no evidence base existed to prioritise actions to improve nutrition. And they argued that the voice of countries must be better heard, felt, and reflected in global decision making. Too often country priorities to strengthen nutrition were ignored by donors and agencies alike. 5 years on, thanks to the work of a consortium of scientists led by Robert E Black from Johns Hopkins Bloomberg School of Public Health (the Maternal and Child Nutrition Study Group), we review the progress made against these findings and recommendations.26 Although some news is better than 5 years ago, there is still a deeply worrying gulf between country needs and global actions. But what is most differentan extraordinary opportunity as well as a severe challengeis the political urgency of nutrition. This latest Lancet Series updates, with extensive new data, the contribution undernutrition in its various forms makes to child mortality and morbidity. Compared with 2008, the result is a radically different picture of the relation between nutritional deficiencies and child health. The overall finding is that 31 million children younger than 5 years die every year from undernutrition; that is a staggering 45% of total child deaths in 2011. To address this enormous and too often hidden cause of child mortality, the Maternal and Child Nutrition Study Group propose a new framework to optimise the delivery of priority evidence-based interventions to prevent and treat undernutrition across the whole life course. Unique to this Series is the systematic approach to both the timing of the interventions and to creating an enabling environment for nutrition. The Maternal and Child Nutrition Study Group emphasises ten interventions targeted to women of reproductive age, during pregnancy, and to infants and children. They calculate the effects of these interventions in 34 countries across Africa, Asia, and the Middle East, where 90% of the global burden of undernutrition resides. In doing so, they reinforce the importance of
www.thelancet.com

the first 1000 days from conception to 2 years. What goes right and what goes wrong for fetal and child nutrition during this period has lasting and irreversible consequences for later life. There are several entirely new findings in this Series. First, the adolescent girl is identified as especially vulnerable to the effects of undernutrition. But that very predicament also makes adolescent girls a group with a special opportunity too. Second, the importance of fetal growth restriction or being born small for gestational age is highlighted. According to new estimates, fetal growth restriction causes more than 800000 neonatal deaths and 20% of stunting in children younger than 5 years worldwide. These findings are presented by Robert E Black and colleagues,2 and Joanne Katz and colleagues7 in the companion Article. Third, the Series is not only concerned with interventions. It also identifies delivery platforms for the implementation of those interventions, most promisingly in the community and in schools. Fourth, the Series costs these interventions and explains why those costsan additional Int$96 billion annually for the 34 countries identifiedare much less prohibitive than they might at first seem. And finally, the Series identifies a further threat to maternal and child nutritional status: overweight and obesity. On June 8, 2013, the Governments of Brazil and the UK will co-host a Nutrition for Growth event. There is therefore an immediate opportunity to foster political

Published Online June 6, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)61100-9 See Online/Series http://dx.doi.org/10.1016/ S0140-6736(13)60937-X, http://dx.doi.org/10.1016/ S0140-6736(13)60996-4, http://dx.doi.org/10.1016/ S0140-6736(13)60843-0, and http://dx.doi.org/10.1016/ S0140-6736(13)60842-9 See Online/Articles http://dx.doi.org/10.1016/ S0140-6736(13)60993-9 and http://dx.doi.org/10.1016/ S0140-6736(13)60103-8

Corbis

Comment

support for the interventions that can be quickly scaled up or linked to nutrition programmessuch as early child development initiatives. It is equally important to take note of the message of Marie Ruel and colleagues4 that in certain sectors, such as agriculture, the evidence of the effect of targeted programmes on maternal and child nutrition is largely inconclusive and requires new approaches to field evaluation. Since 2008, there have been only limited increases in donor aid for nutrition. It is true that nutrition is not so readily attractive to politicians as an international development priority. Undernutrition has a complex set of political, social, and economic causes, none of which are amenable to easy solutions that fit within the timeframe of a single political cycle. For this reason, the outlook today for nutrition is not wholly good. The target endorsed only a year ago at the World Health Assemblyto reduce by 40% the number of children stunted by 2025is already on course to be missed. As the endpoint of the Millennium Development Goals approaches, countries and the international community may agree that nutrition was one of the great missed opportunities of the past 15 years. But this neglect can be turned around quickly. As sustainable development becomes the dominant idea post-2015, nutrition emerges as the quintessential example of a sustainable development objective. If maternal and child nutrition is optimised, the benefits will accrue and extend over several generations. This remarkable opportunity is why Stuart Gillespie and colleagues5 take a very different approach to implementation than in any previous Lancet Series. Instead of exhorting politicians

and policy makers to do somethingor worse, simply hoping that political commitment will appear like a rabbit out of a hatthey set out a practical guide about how to seize the agenda for nutrition, how to create political momentum, and how to turn that momentum into results. This is the prize we have to grasp in the next 18 months. Richard Horton, Selina Lo
The Lancet, London NW1 7BY, UK
We thank the Maternal and Child Nutrition Study Group, led by Robert E Black, for leading the conception and design of this Series. We also acknowledge the generosity of the Bill & Melinda Gates Foundation for providing financial support. 1 2 Morris S, Cogill B, Uauy R, et al. Effective international action against undernutrition: why has it proven so difficult and what can be done to accelerate progress? Lancet 2008; 371: 60821. Black RE, Victora CG, Walker SP, et al, and the Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in lowincome and middle-income countries. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60937-X. Bhutta ZA, Das JK, Rizvi A, et al, The Lancet Nutrition Interventions Review Group, and the Maternal and Child Nutrition Study Group. Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet 2013; published online June 6. http://dx. doi.org/10.1016/S0140-6736(13)60996-4. Ruel MT, Alderman H, and the Maternal and Child Nutrition Study Group. Nutrition-sensitive interventions and programmes: how can they help to accelerate progress in improving maternal and child nutrition? Lancet 2013; published online June 6. http://dx.doi.org/10.1016/ S0140-6736(13)60843-0. Gillespie S, Haddad L, Mannar V, Menon P, Nisbett N, and the Maternal and Child Nutrition Study Group. The politics of reducing malnutrition: building commitment and accelerating progress. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60842-9. Maternal and Child Nutrition Study Group. Maternal and child nutrition: building momentum for impact. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60988-5. Katz J, Lee ACC, Kozuki N, et al, and the CHERG Small-for-Gestational-AgePreterm Birth Working Group. Mortality risk in preterm and small-forgestational-age infants in low-income and middle-income countries: a pooled country analysis. Lancet 2013; published online June 6. http:// dx.doi.org/10.1016/S0140-6736(13)60993-9.

6 7

Maternal and child nutrition: building momentum for impact


Published Online June 6, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)60988-5

In the 5 years since the Maternal and Child Under nutrition Series15 was published in The Lancet there has been a substantial increase in commitment to reduction of malnutrition at global and national levels. Most development agencies have developed or revised their strategies to address undernutrition focused on the first 1000 days of lifethe period from pregnancy to a childs second birthdayas called for in the 2008 Series. One of the main drivers of this new international momentum is the Scaling Up Nutrition movement.6,7 National commit ment in low-income

and middle-income countries (LMICs) is growing, donor funding is rising, and civil society and the private sector are increasingly engaged. Despite this progress, improvements in nutrition still represent a massive unfinished agenda. The 165 million children with stunted growth in 2011 have compromised cognitive development and physical capa bilities, making yet another generation less pro ductive than they would otherwise be.8 Countries will not be able to break out of poverty or sustain economic advances when so much of their population is unable
www.thelancet.com

Comment

to achieve the nutritional security that is needed for a healthy and productive life. Under nutrition is estimated to reduce a nations economic advancement by at least 8% (direct productivity losses, losses via poorer cognition, and losses via reduced schooling).9,10 Although preventable child mortality continues to decrease, undernutrition is responsible for 45% of deaths of children younger than 5 years, amounting to more than 3 million deaths each year.8 Deficiencies of essential vitamins and minerals are widespread and have important adverse effects on child survival and development. Additionally, overweight in adults and increasingly in children constitutes an emerging burden that is quickly establishing itself globally, affecting both poor and rich populations.8 Evidence presented in the accompanying Series on nal and Child Nutrition8,1012 shows the importance Mater of adolescent and maternal nutrition for the health of the mother and for ensuring healthy fetal growth and development. Fetal growth restriction is a cause of 800000 deaths in the first month of life each year, more than a quarter of all neonatal deaths.8 Newborn babies with fetal growth restriction have a substantially increased risk of developing stunting by 24 months of age. Furthermore, these adverse nutritional insults early in life, when coupled with rapid weight gain later in childhood, are important determinants of obesity and non-communicable diseases in adulthood. Thus, it is imperative to act as early as possible in the crucial window of opportunity of pregnancy and the first 2 years of life.8 The emerging platforms for adolescent health and nutrition might offer opportunities for enhanced benefits.10 According to our conservative estimates, we identify a set of ten proven nutrition-specific inter ven tions, which if scaled up from present population coverage to cover 90% of the need, would eliminate about 900 000 deaths of children younger than 5 years in the 34 high nutrition-burden countrieswhere 90% of the worlds stunted children liveand reduce the prevalence of stunting by a fifth, reducing the number of children with stunted growth and development by 33 million.10 The interventions with the largest predicted effects on child mortality are treatment of severe acute malnutrition throughout childhood; promotion of infant and young child feeding, including breastfeeding and appropriate complementary foods;
www.thelancet.com

Panel: Global nutrition targets for 2025, endorsed by the World Health Assembly 40% reduction of the global number of children younger than 5 years who are stunted 50% reduction of anaemia in women of reproductive age 30% reduction of low birthweight No increase in childhood overweight Increase the rate of exclusive breastfeeding in the first 6 months to at least 50% Reduce and maintain childhood wasting to less than 5%

and zinc supplementation. It is, however, important that interventions that have so far contributed to reductions in child mortality, such as vitamin A supplementation, be continued where the need still exists. The cost of scaling up this set of needed nutrition interventions to 90% coverage is estimated at Int$96 billion per year.10 Additionally, nutrition-sensitive activities should be pursued in sectors that address the underlying deter minants of nutrition. Some, but not all, programmes in agriculture, cash transfers, early child development, and schooling have been shown to improve nutrition and broader developmental outcomes for children.11 The studies with the most positive effects had strong designs (including nutrition goals and actions), reached mothers and children early (and for longer durations), and targeted the poorest and most undernourished groups. Many also included actions to empower women and enhance their social status. More evidence is needed from programmes that have good designs, strong implementation, and rigorous evaluation. An enabling environment for nutrition requires empirically sound, timely data about the nature of the problem, evidence for what works and how, good coherence between sectors, good coordination between national and subnational levels, sufficient capacity to build commitment, implementation of programmes at scale, and sustainable public and private means to finance interventions.12 Countries that have managed to improve nutritional status in these contexts have adopted an approach that targets the whole of society.13 This approach requires a good understanding of the political economy of nutri tion. Governments and other stakeholders in success ful countries have built alliances, managed tensions, identified win-win outcomes, established strong
3

Comment

account ability mechanisms, and innovated in the mobilisation of resources for nutrition.11 The private sector is an important force in shaping nutrition outcomes and has the potential to do more. Much more needs to be known about how different forms of public policy, regulation, and financial incen tives can support private organisations to do the right things to improve nutrition. Knowledge in this area is scarce and must be expanded rapidly. The impetus for improving nutrition is stronger today than 5 years ago. The World Health Assembly nutrition targets14 for reduction of stunting, wasting, low birth weight, anaemia, and overweight, and increasing exclu sive breastfeeding in the first 6 months of life (panel), can be achieved by 2025 with sufficient support. The costs of inaction are enormous. As economies grow and the rate of population growth slows, the returns to improved cognitive performance and psychological functioning in the workforce will expand substantially. Benefits will be greater where strategies integrate the promotion of nutrition and child development.15 The new evidence provided in the Maternal and Child Nutrition Series strengthens the case for a continued focus on the first 1000 days. Investments within this window can help meet crucial goals: the prevention of undernutrition, overweight, and poor child development outcomes with longlasting effects on human capital formation. Because many women do not access nutrition-promoting services until month 5 or 6 of pregnancy, we draw attention to the need to ensure women enter pregnancy in a state of optimum nutrition. Nutrition is foundational to both individual and national development. The post-Millennium Develop ment Goals agenda must put the resolution of all forms of malnutrition at the top of its aims. An increase in donor spending is crucial if nutrition targets are to be met or surpassed. Government spending in LMICs needs to match or exceed this rate of increase. Nutrition budget lines need to be established in all high-burden countries. Governments need to be supported to raise public resources for nutrition. The increased mobilisation of private resources from individuals, businesses, and new philanthropies needs to be incentivised towards the most effective ways of improving nutrition. Scaling Up Nutrition is a crucial driver of these needed actions and support for it must remain strong.
4

Many nutrition gains have been made, but they need to be protected in the face of new stressors such as climate change, humanitarian crises, and food price volatility. We need to encourage innovation in design and delivery of nutrition-specific interventions, to make them even more affordable at scale. New incentives need to be established that support innovations in nutritionsensitive pro gramme design and implementationto unleash their potential to achieve their own goals by providing crucial additional support to efforts to reduce malnutrition. This Series strengthens the evidence that good nutrition is a fundamental driver of a wide range of development goals. Investments need to be directed not only to inter ventions, but also to the creation of environments that enable them. This approach requires strategic investment in commitment building, capacity, and leadership; timely data describing the nature of the malnutrition problem and its causes; evidence for what works; accountability mechanisms; resource mobilisation; and building of institutions required for sustainable implementation. A political economy approach to prioritisation of such investments is crucial if viable enabling environments are to be created. More research is needed to develop scalable inter ven tions or improve the effectiveness of existing ones to have greater effects, especially by preventing fetal growth restriction and growth faltering in infancy. Although promising service delivery platforms exist in communities, evidence is needed about how to ensure that nutrition interventions reach the populations with greatest need. More research is needed into the barriers to effective implementation and into the costs and logistics of scaling up: into the crucial elements of capacity at different levels, into the development and assessment of financing mechanisms for nutrition, and into ways to reduce the costs of implementation. Rigorous evidence is needed to show how the private sector can best support optimum nutrition. Research is also needed into the drivers of country success, how to create enabling environments, and into the features of nutrition-sensitive programmes that improve nutrition. This year, 2013, represents the best opportunity yet to make these proposed actions a reality. National and international momentum to address human nutrition and related food security and health needs is at a high level. Nutrition is now more prominent on the agendas
www.thelancet.com

Comment

of the UN, G8, and G20, and supporting civil society, business, and academic organisations. We must work together to seize this opportunity. Maternal and Child Nutrition Study Group
Group members: *Robert E Black (Johns Hopkins Bloomberg School of Public Health, USA), Harold Alderman (International Food Policy Research Institute, USA), Zulfiqar A Bhutta (Aga Khan University, Pakistan), Stuart Gillespie (International Food Policy Research Institute, USA), Lawrence Haddad (Institute of Development Studies, UK), Susan Horton (University of Waterloo, Canada), Anna Lartey (University of Ghana, Ghana), Venkatesh Mannar (The Micronutrient Initiative, Canada), Marie Ruel (International Food Policy Research Institute, USA), Cesar G Victora (Universidade Federal de Pelotas, Brazil), Susan P Walker (The University of the West Indies, Jamaica), Patrick Webb (Tufts University, USA) rblack@jhsph.edu
REB serves on the Boards of the Micronutrient Initiative, Vitamin Angels, the Child Health and Nutrition Research Initiative, and the Nestl Creating Shared Value Advisory Committee. VM serves on the Nestl Creating Shared Value Advisory Committee. HA, ZAB, SG, LH, SH, AL, MR, CGV, SPW, and PW declare that they have no conflicts of interest. 1 Black RE, Allen LH, Bhutta ZA, et al, for the Maternal and Child Undernutrition Study Group. Maternal and child undernutrition: global and regional exposures and health consequences. Lancet 2008; 371: 24360. Bhutta ZA, Ahmed T, Black RE, et al, for the Maternal and Child Undernutrition Study Group. What works? Interventions for maternal and child undernutrition and survival. Lancet 2008; 371: 41740. Victora CG, Adair L, Fall C, et al, for the Maternal and Child Undernutrition Study Group. Maternal and child undernutrition: consequences for adult health and human capital. Lancet 2008; 371: 34057.

4 5 6 7 8

10

11

12

2 3

13 14 15

Bryce J, Coitinho D, Darnton-Hill I, et al, for the Maternal and Child Undernutrition Study Group. Maternal and child undernutrition: effective action at national level. Lancet 2008; 371: 51026. Morris SS, Cogill B, Uauy R. Effective international action against undernutrition: why has it proven so difficult and what can be done to accelerate progress? Lancet 2008; 371: 60821. Scaling Up Nutrition. A framework for action. 2010. http://unscn.org/files/ Activities/SUN/PolicyBriefNutritionScalingUpApril.pdf (accessed April 2, 2013). Bezanson K, Isenman P. Scaling up nutrition: a framework for action. Food Nutr Bull 2010; 31: 17886. Black RE, Victora CG, Walker SP, et al, and the Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60937-X. Horton S, Steckel RH. Global economic losses attributable to malnutrition 19902000 and projections to 2050. In: Lombard B, ed. How much have global problems cost the world? A scorecard from 1900 to 2050. Cambridge: Cambridge University Press, 2013 (in press). Bhutta ZA, Das JK, Rizvi A, et al, The Lancet Nutrition Interventions Review Group, and the Maternal and Child Nutrition Study Group. Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet 2013; published online June 6. http://dx. doi.org/10.1016/S0140-6736(13)60996-4. Ruel MT, Alderman H, and the Maternal and Child Nutrition Study Group. Nutrition-sensitive interventions and programmes: how can they help to accelerate progress in improving maternal and child nutrition? Lancet 2013; published online June 6. http://dx.doi.org/10.1016/ S0140-6736(13)60843-0. Gillespie S, Haddad L, Mannar V, Menon P, Nisbett N, and the Maternal and Child Nutrition Study Group. The politics of reducing malnutrition: building commitment and accelerating progress. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60842-9. Dube L, Pingali P, Webb P. Paths of convergence for agriculture, health, and wealth. Proc Natl Acad Sci USA 2012; 109: 12294301. WHO. Discussion paper. Proposed global targets for maternal, infant and young child nutrition. Geneva: World Health Organization, 2012. Grantham-McGregor S, Cheung YB, Cueto S, et al. Developmental potential in the first 5 years for children in developing countries. Lancet 2007; 369: 6070.

Delivery platforms for sustained nutrition in Ethiopia


The 2013 Lancet Series on Maternal and Child Nutrition emphasises the crucial importance of scale-up of effective nutrition interventions through health and community delivery platforms. The Series acknowledges that strong health systems are central to achievement of this goal, and for progress towards the 2015 Millennium Development Goals (MDGs).1 A broad consensus exists about the need for strengthening of health systems to meet the goals of the healthrelated MDGs by 2015.1 Because disease-control programmes and general health services often share common service-delivery platforms, they are necessary and complementary in countries with a high disease burden, especially in sub-Saharan Africa.2,3 Some findings have suggested that health and nutrition programmes can strengthen health systems and, similarly, that health systems can strengthen programme implementation.47 Furthermore, for such
www.thelancet.com

strengthening to take place, a system or platform for service delivery should be country led and owned to ensure sustainability and effectiveness.

Published Online June 6, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)61054-5

Corbis

Comment

In the past few years, many countries have worked to develop systems and infrastructure at the most decentralised level of services, and these investments have enabled populations to access essential services in sectors such as health, agriculture, education, and social welfare. Ethiopia is exemplary in this regard in view of the countrys progress towards some of the key MDGs, which is mainly attributable to a decentral ised service delivery platformthe Health Extension Programme.8 Launched in 2003, this programme was organised to provide universal access to primary health care, mainly preventive services,6,7 through more than 38000 government-salaried female health extension workers. Two workers were placed in a health post to serve each kebele (the smallest administrative unit) of about 5000 people nationwide. Through this programme new vaccines were introduced and health services expanded, which improved health ments were and nutrition care practices, and invest made in education and social economic development, contributing to a reduction in the number of child deaths by nearly half.9,10 The present estimate (supported by the Central Statis tical Agency) for the mortality rate in children younger than 5 years in Ethopia is 77 per 1000 livebirths (compared with 166 in 2000 and 123 in 2005).11 On the basis of the present trend, Ethiopia is predicted to meet MDG 4, to reduce child mortality, by 2015, by having a mortality rate in children younger than 5 years of 68 per 1000 livebirths.12 Furthermore, a comparison of national levels of malnutrition in the 2000 and 2011 Ethiopia Demographic and Health Surveys (EDHS) shows that stunting has declined from 58% to 44%, underweight from 41% to 29%, and prevalence of wasting from 12% to 10%.10 Globally, the prevalence of stunting in children younger than 5 years has fallen by 36% in the past two decades, from an estimated 40% in 1990, to 26% in 2011.9 To consolidate the gains and enhance the effectiveness of the Health Extension Programme, the Government of Ethiopia has designed a scaling-up strategy, in the form of a so-called health development army, which will scale up documented best practices and use families as role models. Such a strategy is based on social learning theory whereby peer-to-peer modelling can dissemin ate emerging information and instil improved healthseeking behaviours at community level.
6

The Health Extension Programme plays a crucial part in the success of the national nutrition programme and strategy that was introduced in Ethiopia in 2008. The community-based management of acute malnutrition approach of the Health Extension Programme manages more than 300000 children in more than 10000 health posts annually, has provided vitamin A supplementation and deworming tablets to 11 million children and 700000 pregnant and lactating women every 6 months since 200506, and distributes iron-folate supple men tation targeted to reach 80% of pregnant women every year.13 Interventions of the community-based nutrition programme include infant and young child nutri tion, and growth monitoring and promotion via the Triple A cycle (assess the problem, analyse its causes and possible solutions, and take appropriate action). The communitybased nutrition programme is currently being supported in more than 300 food-insecure woredas (districts), reaching 1500185 (80%) children younger than 2 years. Efforts of the community-based nutrition programme have resulted in more than 50% of children in Ethiopia being exclusively breastfed (EDHS, 2011).10 However, the proportion of children receiving a minimum acceptable diet is only 4% in Ethiopia, show ing the urgent need to finalise a national strategy for improvement of quantity and quality of complementary feeding. The revised national nutrition programme spanning 201315 will address these challenges by emphasising the first 1000 days of life, with a focus on children younger than 2 years, pregnant and lactating women, and adoles cent girls, to break the intergenerational cycle of malnutrition. Furthermore, the revised pro gramme will emphasise actions for acceleration of stunting reduction by focusing on nutrition-sensitive interventions in other development sectors such as education, agriculture, social protection and womens affairs, and civil society organisations and the private sector. The role of health extension workers and the health development army will continue to be central to achievement of equitable access of all vulnerable women and children to both curative and preventive services, and to ensure that targets specified in the health sector development plan IV of Ethiopia are met.13 Ethiopias actions have enabled development workers to engage people at risk in an integrated manner using a unified Health Extension Programme, enabling achievement of great gains in child survival and
www.thelancet.com

Comment

nutrition. The Government of Ethiopia, on the basis of experiences in the past 10 years of this programme and the substantial improvements in nutritional status, believes that even greater efforts can be made to reduce stunting. The govern ment will continue to optimise the revised national nutrition programme and the global efforts on nutrition such as Scaling Up Nutrition (SUN) and Renewed Efforts Against Child Hunger (REACH),14 which are mechanisms to catalyse further multisectoral nutrition-sensitive actions beyond the health sector. Nutrition, as one of many crucial indicators of health status, should be used for close programmatic link ages and synergies between targeted social protection interventions. These synergistic actions across social services will contribute towards increased resource allocation for the national nutrition programme, and ensure that sustainable interventions are scaled up to improve food and nutrition security. *Ferew Lemma, Joan Matji
Ethiopian Federal Ministry of Health, PO Box 1234, Addis Ababa, Ethiopia (FL); and UNICEF Ethiopia, Addis Ababa, Ethiopia (JM) lferew@gmail.com
We declare that we have no conflicts of interest. 1 Singh A. Strengthening health systems to meet MDGs. Health Policy Plan 2006; 21: 32628.

2 3 4

5 6 7

8 9 10 11 12 13 14

WHO. Everybodys business. Strengthening health systems to improve health outcomes: WHOs framework for action. Geneva: World Health Organization, 2007. WHO. The Global Fund strategic approach to health systems strengthening: report from WHO to the Global Fund Secretariat. Geneva: World Health Organization, 2007. Atun R, Bennett S, Duran A. When do vertical (stand-alone) programmes have a place in health systems? Copenhagen: World Health Organization Regional Office for Europe and the European Observatory on Health Systems and Policies, 2008. Atun R, de Jongh T, Secci F, Ohiri K, Adeyi O. A systematic review of the evidence on integration of targeted health interventions into health systems. Health Policy Plan 2010; 25: 114. Atun R, de Jongh T, Secci F, Ohiri K, Adeyi O. Integration of targeted health interventions into health systems: a conceptual framework for analysis. Health Policy Plan 2010; 25: 10411. Atun R, Menabde N. Health systems and systems thinking: In: Coker R, Atun R, McKee M, eds. Health systems and the challenge of communicable diseases: experiences from Europe and Latin America. Berkshire, UK: Open University Press, 2009: 12140. Wakabi W. Extension workers drive Ethiopias primary health. Lancet 2008; 372: 880. UN Childrens Fund. Improving child nutrition: the achievable imperative for global progress. New York: UNICEF, 2013. Central Statistical Agency, ICF International. Ethiopia Demographic and Health Survey 2011. March, 2012. http://measuredhs.com/pubs/pdf/ FR255/FR255.pdf (accessed May 22, 2013). WHO. Ethiopia: health profile. May, 2013. http://www.who.int/gho/ countries/eth.pdf (accessed May 22, 2013). Ministry of Finance and Economic Development, UN Ethiopia. Child survival, health and nutrition. In: Nebede A, Pearson R, eds. Investing in boys and girls in Ethiopia: past, present and future, 2012. Ethiopia: UNICEF, 2012: 3553. Federal Democratic Republic of Ethiopia, Ministry of Health. Health Sector Development Program IV, 2010/112014/15. Addis Ababa: Federal Democratic Republic of Ethiopia Ministry of Health, 2010. REACH Partnership. Annual report 2012. 2012. http://www. reachpartnership.org/documents/312104/315126/REACH+Annual+Report+ 2012?version=1.0 (accessed May 22, 2013).

For more on the SUN initiative see http://scalingupnutrition. org/resources-archive

Only collective action will end undernutrition


We are in a race against time to eradicate the global scourge of undernutrition. Undernutrition cripples global economic growth and development, and future global prosperity and security are intimately linked with our ability to respond adequately to this urgent challenge. The new Series in The Lancet shows that undernutrition contributes to the deaths of about 3 million children each year45% of the total.1 Its results stunt the physical growth and life chances of millions of people, and for Africa and Asia estimates suggest that up to 11% of national economic productivity is lost to undernutrition.2 The evidence provided in this Series should act as a turning point to galvanise global action. The solution lies largely in the early years of life, when the foundations for human potential are laidgetting the right nutrients at the right time prevents undernutrition. The result is heightened educational attainment, adult wages, and economic productivity.
www.thelancet.com

Women and girls are at the heart of this message. As the bearers and carers of children, their health and economic potential is entwined with that of future

Published Online June 6, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)61084-3

Corbis

Comment

For the first Lancet Series on maternal and child undernutrition see http://www. thelancet.com/series/maternaland-child-undernutrition

For Scaling Up Nutrition see http://scalingupnutrition.org

generations. Unless girls grow well in early childhood and adolescence and enter into motherhood well nourished, are lent support during pregnancy, protected from heavy physical labour, and empowered to breastfeed and provide good food for their babies and toddlers, the intergenerational cycle of undernutrition will not be broken. This Series shows that poor maternal nutrition at conception and during pregnancy is a major contributor to undernutrition in childhood.1 Empowering women to make the right choices for their health, and that of their children, is crucial to solving this challenge. Why is this such an urgent issue? Important demo graphic changes are occurring in many countries with high levels of undernutrition. The ratio of the work ing age to non-working age population is rising and will peak in the next 20 years, and this increase in the available workforce has substantially boosted economic growth in many parts of the world.3 Any such demographic dividend will be even greater in wellnourished populations. Additionally, rapid urbanisation, increased sedentary behaviour, and a transition in dietary patterns has resulted in a fast rise of obesity in middle-income and even low-income countries. This Series emphasises that undernourished children are at increased risk of becoming overweight and developing non-communicable diseases such as diabetes in later life.1 Acting now brings a triple benefit: it saves lives today, maximises economic opportunity, and helps to reduce obesity and chronic disease in the future. This Series shows that there are simple and proven interventions that can substantially reduce undernutrition and mortality in children. Many of these interventions deliver an excellent return on investment and should be delivered at scale without delay. However, making a lasting effect on the root causes of undernutrition will need more effort. Brazils remarkable experience during the past 20 years shows us that the right programmes need to be matched with strong political leadership and determination. Brazils success resulted from a whole-government response, a clear focus on groups at greatest risk, strong civil society engagement, and investments to track progress and use data to strengthen accountability and inform policy choices.4 In addition to strong national action on under nutrition, we need to take a hard look at our global

agriculture and food system. As the global population rises, our food system needs to keep pace with the demand for both dietary energy and the essential vitamins and minerals needed for human health. Our agriculture and food system needs to be profitable for farmers and the wider food sector, environmentally sustainable, and directly supportive of the health and nutritional needs of populations. Everyone is part of the solution. Governments need to lead; businesses need to identify how to improve nutrition through their business models and employment practice; civil society organisations need to help citizens to drive transparency and accountability; and the scientific community needs to keep us focused on evidence about what works. Policy commitments, capacity strengthening, and targeted financing are all essential. Global efforts on food and nutrition will likewise be substantially boosted by a clear signal of nutrition priorities in the post-2015 development agenda. This agenda will do more than steer aid; it should provide direction on global investment, buy in support from the private sector, and encourage a coherent approach from international institutions. Nutrition should be centrally positioned in that agenda to ensure energy and nutrient needs are met at each stage of life. The first Lancet Series on maternal and child under nutrition, published in 2008, helped to start the race to eradicate poor nutrition. In the past 5 years the governments of 35 countries have committed to do more to tackle undernutrition, and have joined the Scaling Up Nutrition movement. On the last day of the 2012 Olympics, the governments of the UK and Brazil co-hosted an event in London, UK, to generate political momentum in the fight against undernutrition. On June 8, 2013, the Nutrition for Growth high level event in London will help to secure a global response that will include financial, business, scientific, and political commitments matched to the scale of the challenge. Progress will be reviewed annually and again at the Olympics in Rio de Janeiro, Brazil, in 2016. We cannot afford to miss this opportunity to act together to beat undernutrition. *Anna Taylor, Alan D Dangour, K Srinath Reddy
UK Department for International Development, London SW1A 2EG, UK (AT, ADD); Faculty of Epidemiology and Population

www.thelancet.com

Comment

Health and Leverhulme Centre for Integrative Research in Agriculture and Health, London School of Hygiene & Tropical Medicine, London, UK (ADD); and Public Health Foundation of India, New Delhi, India (KSR) a-taylor@dfid.gov.uk
We declare that we have no conflicts of interest. 1 Black RE, Victora CG, Walker SP, et al, and the Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60937-X.

3 4

Horton S, Steckel RH. Malnutrition: Global economic losses attributable to malnutrition 19002000 and projections to 2050. In: Lomborg B, ed. How much have global problems cost the world? A scorecard from 1900 to 2050. Cambridge: Cambridge University Press, 2013 (in press). Eastwood R, Lipton M. Demographic transition in sub-Saharan Africa: how big will the economic dividend be? Popul Stud (Camb) 2011; 65: 935. Monteiro CA, Benicio MH, Conde WL, et al. Narrowing socioeconomic inequality in child stunting: the Brazilian experience, 19742007. Bull World Health Organ 2010; 88: 30511.

Nutrition-sensitive food systems: from rhetoric to action


Action to improve the nutrition sensitivity of food systemsand thereby increase the nutritional value of offers substantial food for people around the world but underused opportunities.1,2 The rhetoric about such opportunities brought about by the global food crisis in 200708 has not resulted in much new action, for at least two reasons. First, goals other than improved nutrition are pursued by strong economic and political interests in both the agricultural sector and the postharvest value chain.3 Farmers and other economic agents in food systems aim to make money subject to reasonable levels of risk, and governments pursue policies that are compatible with the interests of politically powerful stakeholder groups.4 Malnourished populations are rarely among these interests. The very high value of improved nutrition to societies should be supported by alignments to create compatibility between nutrition and economic goals for farmers and processors, and political momentum3 has to be created to foster policy interventions that make food systems nutrition sensitive. Governments could pursue two kinds of policy action: they could either change the behaviour of farmers, consumers, food processors, and other economic agents in the system through incentives, regulations, and knowledge; or they could accept present behaviours and introduce health-specific and nutrition-specific interventions to compensate for any nutritional damage done or improvements forgone. Although changing of behaviour is likely to be more cost-effective and sustainable, the second option is the most common. For example, food-system policies and the private sector promote inexpensive calories and expensive nutrients, resulting in overweight and micronutrient deficiencies. Health and nutrition-specific interventions, such
www.thelancet.com

as treatment of chronic diseases and micronutrient supplementation, are intro duced to remedy problems that could have been avoided. The appropriate policy interventions to change behaviour will be context specific and might include agricultural research to increase productivity of fruit and vegetable cultivation and reduce micronutrient deficiency; taxes on sugar, sweeteners, and fat to reduce the prevalence of obesity; regulations for advertising and promotion; and education about nutrition.5 In highincome and rapidly growing low-income countries, the agricultural sector has become or is rapidly becoming a supplier of raw materials for the food processing industry, rather than a provider of food for direct consumption. As this transition proceeds, the potential for improvements to nutrition through nutritionsensitive food systems moves from agriculture to the post-harvest value chain. The transition amplifies health and nutrition risks by promotion of what Monteiro

Published Online June 6, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)61053-3

Bloomberg via Getty Images

Comment

and colleagues call ultra-processed foods,6,7 resulting in unhealthy dietary patterns. However, policy action to regulate and incentivise the food industry to avoid such negative health and nutrition effects and change consumer preferences is very scarce. A second reason for lack of action to improve nutrition is the fixation of the health and nutrition community on randomised controlled trials (RCTs) as the only legitimate source of evidence.8,9 Unfortunately, RCTs the gold standard in health researchare generally impossible to apply to the food system except in small, usually unimportant, projects. Health and nutrition effects resulting from agricultural and other foodsystem policies and programmes are very difficult to assess with RCTs, partly because treatments cannot be randomised and because the effect pathway is long. Yet the most promising opportunities for improvement of health and nutrition are undoubtedly found in such policies, and not in home gardens and other minor projects which are amenable to study within the framework of randomised trials. Although existing evidence obtained by other approaches is deemed inconclusive1 and does not support policy intervention, the pathways through which food systems can affect nutrition (positively or negatively) are well known. Furthermore, key components making up these pathways, such as incomes, prices, womens time allocation, dietary diversity, advertising and pro motion, and household and individual behaviour have a substantial effect on nutrition. Thus, if pathway analysis shows that changes in the food system improve one or more of these componentseg, dietary diversity or womens time allocationand such improvements reduce micro nutrient deficiencies, is such evidence really acceptable for policy guidance only if it is derived from RCTs? If so, the evidence will be limited to small food-systems

programmes such as kitchen garden projects, whereas the really important changes for nutrition, such as prioritisation of agricultural research to enhance pro duc tivity in fruit and vegetable cultivation so as to reduce prices and improve micronutrient status, and various policies to change womens time allocation or prices of various foods, will be ignored because they cannot be studied in RCTs.1,10 Per Pinstrup-Andersen
Division of Nutritional Sciences, Cornell University, Ithaca, NY 14853, USA pp94@cornell.edu
I declare that I have no conflicts of interest. Ruel MT, Alderman H, and the Maternal and Child Nutrition Study Group. Nutrition-sensitive interventions and programmes: how can they help to accelerate progress in improving maternal and child nutrition? Lancet 2013; published online June 6. http://dx.doi.org/10.1016/ S0140-6736(13)60843-0. 2 Pinstrup-Andersen P. Food systems and human health and nutrition: an economic policy perspective with a focus on Africa. Oct 11, 2012. http:// iis-db.stanford.edu/evnts/6697/Pinstrup-Andersen_10_11_12.pdf (accessed May 2, 2013). 3 Gillespie S, Haddad L, Mannar V, Menon P, Nisbett N, and the Maternal and Child Nutrition Study Group. The politics of reducing malnutrition: building commitment and accelerating progress. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60842-9. 4 Pinstrup-Andersen P, Watson DD II. Food policy for developing countries: the role of government in global, national, and local food systems. Ithaca, NY: Cornell University Press, 2011. 5 Herforth A, Jones A, Pinstrup-Andersen P. Prioritizing nutrition in agriculture and rural development: guiding principles for operational investments. Washington, DC: World Bank, 2012. 6 Monteiro CA, Levy RB, Claro RM, deCastro IRR, Cannon G. Increasing consumption of ultra-processed foods and likely impact on human health: evidence from Brazil. Public Health Nutr 2011; 14: 513. 7 Monteiro CA. Nutrition and health. The issue is not food, nor nutrients, so much as processing. Public Health Nutr 2009; 12: 72931. 8 Bhutta ZA, Ahmed T, Black RE, et al, for the Maternal and Child Undernutrition Study Group. What works? Interventions for maternal and child undernutrition and survival. Lancet 2008; 371: 41740. 9 Bhutta ZA, Das JK, Rizvi A, et al, The Lancet Nutrition Interventions Review Group, and the Maternal and Child Nutrition Study Group. Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet 2013; published online June 6. http://dx. doi.org/10.1016/S0140-6736(13)60996-4. 10 Masset E, Haddad L, Cornelius A, Isaza-Castro J. A systematic review of agricultural interventions that aim to improve nutritional status of children. London: EPPI-Centre, Social Science Research Unit, Institute of Education, University of London, 2011. 1

Global child and maternal nutritionthe SUN rises


Published Online June 6, 2012 http://dx.doi.org/10.1016/ S0140-6736(13)61086-7 For the Scaling Up Nutrition policy brief see http:// scalingupnutrition.org/ wp-content/uploads/pdf/ SUN_Framework.pdf

In April, 2010, a policy briefScaling Up Nutrition: A Framework for Actionwas released at the spring meetings of the World Bank and International Monetary Fund. It was a collective effort stimulated by the publication in January, 2008, of The Lancets Series on undernutrition. The Lancet Series encouraged

an emphasis on the 1000 day window from the start of pregnancy to a childs second birthday, with interventions that are both cost-effective and yield high returns for cognitive development, individual adult earnings, and economic growth. A second Series on nutrition, published in The Lancet,14 now explicitly
www.thelancet.com

10

Comment

shows that the solution to malnutrition relies on a collective effort in which all stakeholdersgovernments, academia, civil society, UN system organisations, foundations, development banks, and businessescarry out specific roles in ensuring that interventions are delivered equitably and at scale. The policy brief, which quickly became known as the SUN Framework, set the stage for the transformation that is now happening in global nutrition. It called for country-owned nutrition strategies and programmes; urgent scaling up of evidence-based and cost-effective interventions; integration of nutrition within national strategies for gender equality, agriculture, food security, social protection, education, water supply, sanitation, and health care; and a substantial increase in domestic support and external assistance for nutrition within the food security, social protection, and health sectors. The SUN Road Map, prepared later in 2010 and revised in 2012, set out ways for a wide range of groups to work together in sharpening, scaling up, and aligning their responses to peoples nutritional needsand achieving results. Alongside the 1000 days advocacy partnership, the SUN Movement was launched at the UN General Assembly in September, 2010. It takes forward the SUN Road Map by encouraging coherence and effectiveness among all groups working for better nutrition; it is not an initiative, project, or programme. By April, 2013, 35 countries had joined the SUN Movement with commitments that are in line with the SUN Framework and Road Map. These countries nutrition solutions show the commitments of political leaders, whole-ofsociety responses, careful tracking of progress, and the benefits of shared experience. The second Lancet nutrition Series provides a range of valuable insights as the SUN Movement moves through 2013, a year dense with events that will move nutrition to the heart of the development agenda. It calls for a substantial increase of political commitment in responding to the complex causes of undernutrition. It recognises that the SUN Movement has the potential to harness such change and yield durable results. The first paper1 of the Series leaves no doubt as to why nutrition is key for sustainable development and the wellbeing of entire populations. The second paper2 strengthens the arguments of the 2008 Lancet
www.thelancet.com

Series by bringing additional evidence to support the focus on ensuring that all women, girls, and young children are able to access specific interventions of good quality; they should be included in mainstream efforts for public health, family planning, and water and sanitation. The third paper3 sets the foundation for evidence-based research into achieving outcomes through nutrition-sensitive strategies in four key areas: agriculture and food, social security, early child development, and class room education. It draws on the experience of countries that have made great progress when stressing that gender and social equality are the cornerstones of nutritional success. By focusing on the political context for effective action, the fourth paper4 recalls that the realisation of human rights, a commitment to equity, and gender equality should be properly prioritised. It includes important proposals for ways in which business can best be engaged (and the challenges of doing so), and encourages increased involvement of civil society at all levels. It likewise underlines the need for governments to increase their own accountability for ensuring that people are able to achieve good nutrition and to ensure the existence of a fair and transparent framework for regulating any entity that mighteven unwittinglyundermine nutritional justice. At the May, 2012, World Health Assembly, government representatives agreed ambitious goals for reduction of all forms of malnutrition, including obesity. This Lancet Series points out that

For the Lancets 2008 Series on maternal and child undernutrition see http://www. thelancet.com/series/maternaland-child-undernutrition For the SUN Movement Revised Roadmap see http:// scalingupnutrition.org/wpcontent/uploads/2012/10/SUNMovement-Road-MapSeptemeber-2012_en.pdf For the 1000 days partnership see http://www.thousanddays. org

11

Bloomberg via Getty Images

Comment

these goals can only be achieved through improved nutrition governance, more human resources, better demonstration of results, and increased investments from domestic and international sources. In the past 3 years the SUN Movement has provided a platform to enable leaders to pledge to intensify efforts for improved nutrition. 2013 provides a once-in-a-lifetime opportunity to strengthen worldwide resolve for improved nutrition, through commitments being made in a series of international and regional events. The publication of the second Lancet Series is timely and reinforces the urgency for transformation of political commitment into actions that lead to improvements in nutrition. The Series brings scientific rigour to the challenge of equitable delivery of effective services at scaleboth now and in the years to come.

David Nabarro
SUN Movement Secretariat, Villa La Pelouse (2nd Floor), Palais Des Nations, 1201 Geneva, Switzerland david.nabarro@undp.org
I declare that I have no conflicts of interest. 1 Black RE, Victora CG, Walker SP, et al, and the Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60937-X. Bhutta ZA, Das JK, Rizvi A, et al, The Lancet Nutrition Interventions Review Group, and the Maternal and Child Nutrition Study Group. Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet 2013; published online June 6. http:// dx.doi.org/10.1016/S0140-6736(13)60996-4. Ruel MT, Alderman H, and the Maternal and Child Nutrition Study Group. Nutrition-sensitive interventions and programmes: how can they help to accelerate progress in improving maternal and child nutrition? Lancet 2013; published online June 6. http://dx.doi.org/10.1016/ S0140-6736(13)60843-0. Gillespie S, Haddad L, Mannar V, Menon P, Nisbett N, and the Maternal and Child Nutrition Study Group. The politics of reducing malnutrition: building commitment and accelerating progress. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60842-9.

Early nutrition and adult outcomes: pieces of the puzzle


Published Online March 28, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)60716-3 See Online/Articles http://dx.doi.org/10.1016/ S0140-6736(13)60103-8

The association between nutrition in early life and long-term health has been of interest for decades. Since the articulation of the fetal origins hypothesis by David Barker and colleagues,1 there has been debate about the implications of fetal undernutrition and early childhood growth on outcomes of importance in adult health and risks of chronic diseases. Both epidemiological and animal studies have shown that the risk of metabolic syndrome is significantly increased

after exposure to suboptimum nutrition during crucial periods of development.1 The importance of these findings greatly increased after reports about the global burden of non-communicable diseases and risk factors were published in December, 2012.2 Evidence for the importance of early nutrition for adult outcomes was derived initially from obser vational cohort studies3 and was reaffirmed by analysis of outcome data from several cohort studies in 2008.4 This analysis4 was focused on a meta-analysis of coefficients from different sites: birthweight, weight and length Z scores, and stunting at age 2 years. In The Lancet, Linda Adair and colleagues5 report findings from a study in which they pooled data from five birth cohorts and investigated how linear growth and relative weight gain in several age ranges affected adult outcomes. They report that higher birthweight was associated with an adult bodymass index of greater than 25 kg/m (odds ratio 128, 95% CI 121135) and a reduced likelihood of short stature (049, 044054) and of not completing secondary school (082, 078087). Faster linear growth was also strongly associated with reduced likelihood of short adult stature (age 2 years: 023, 020052; mid-childhood 039, 036043) and of not completing secondary school (age 2 years: 074,
www.thelancet.com

Andrew Aitchison/In Pictures/Corbis

12

Comment

067078; mid-childhood 087, 083092). Faster relative weight gain was associated with an increased risk of adult overweight (age 2 years: 151, 143160; mid-childhood 176, 169191) and elevated blood pressure (age 2 years: 107, 101113; mid-childhood: 122, 115130). Notwithstanding the key findings, several limitations of this pooled analysis should be recognised. The authors had to make do with disparate information about socio economic status and income, and impute some information that was missing. Although they adjusted for maternal education and socioeconomic status (largely assets rather than income), other potential confounding factors (eg, household and learning environment) could not be assessed in relation to attained schooling. Several additional limitations clusions. Little or no information was preclude firm con available about maternal nutrition and micronutrient status. Additionally, Adair and colleagues do not report any outcomes related to intrauterine growth retardation or gestational age at birth, and merely report association with birthweight, which might be oversimplified. Being small for gestational age at term, and especially preterm, has now been recognised as a major risk factor for excess newborn and infant mortality6 and accounts for a substantial proportion of child stunting.7 Prematurity is associated with increased risks of metabolic syndromes in later life.8 Potential variations in body composition of newborn babies might not be captured by mere measurement of birthweight or size. So-called thin-fat infantsie, small newborn babies that have elevated body fat content have been described9 and could be associated with increased risks of insulin resistance in childhood.10 These limitations aside, Adair and colleagues find ings5 are some of the most important from existing cohorts linking early childhood nutritionespecially birthweight and improved patterns of linear growth with long-term outcomes. They have clear implications for public health policy and nutrition interventions. As shown by an analysis of evidence-based interventions,11 a focus on improvements in nutrition in pregnancy and linear growth in the first 2 years of life could lead to substantial reductions in stunting and improved survival. These improvements form the basis for the emphasis on the first 1000 days of life, which has been used effectively to scale up nutrition activities.12
www.thelancet.com

However, this tenet could be too simplistic, because it focuses on care during pregnancy and ignores the vital contribution of maternal health and nutrition in the periods before and just after conception to intrauterine and postnatal growth. Evidence supports men an association between micronutrient supple tation around the time of conception and DNA methylation13 and increased methylation of the IGF2 genes in childhood,14 indicating that these factors could affect linear growth postnatally. Although Adair and colleagues analysis of birth outcomes in the international cohorts does not shed light on the importance of maternal health and nutrition before conception, these factors might be just as important as postnatal factors and should be investigated. What is the way forward? Although the evidence emerging from observational studies such as Adair and colleagues is important for policy, well designed prospective studies with appropriate interventions and follow-up are clearly needed. The outcomes should include elements of child development, education, employment, and earnings, which would allow improved estimation of effect on human capital. Although expensive and difficult to organise and implement, such cohort studies are a crucial investment for the future and, in view of the interest in human development in the post-2015 era, should be prioritised for funding. Zulfiqar A Bhutta
Division of Women and Child Health, Aga Khan University, Karachi 74800, Pakistan zulfiqar.bhutta@aku.edu
I declare that I have no conflicts of interest. 1 2 Barker DJ. Sir Richard Doll lecture: developmental origins of chronic disease. Public Health 2012; 126: 18589. Lim SS, Vos T, Flaxman AD, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 19902010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012; 380: 222460. Barker DJP, Godfrey KM, Gluckman PD, Harding JE, Owens JA, Robinson JS. Fetal nutrition and cardiovascular disease in adult life. Lancet 1993; 341: 93841. Victora CG, Adair L, Fall C, et al, for the Maternal and Child Undernutrition Study Group. Maternal and child undernutrition: consequences for adult health and human capital. Lancet 2008; 371: 34057. Adair LS, Fall CHD, Osmond C, et al, for the COHORTS group. Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: findings from five birth cohort studies. Lancet 2013; published online March 28. http:// dx.doi.org/10.1016/S0140-6736(13)60103-8. Marchant T, Willey B, Katz J, et al. Neonatal mortality risk associated with preterm birth in East Africa, adjusted by weight for gestational age: individual participant level meta-analysis. PLoS Med 2012; 9: e1001292.

3 4 5

13

Comment

Black RE, Victora CG, Walker SP, et al, and the Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60937-X. 8 Parkinson JRC, Hyde M J, Gale C, Santhakumaran S, Modi N. Preterm birth and the metabolic syndrome in adult life: a systematic review and meta-analysis. Pediatrics 2013; published online March 18. DOI:10.1542/ peds.2012-2177. 9 Yajnik CS, Fall CH, Coyaji KJ, et al. Neonatal anthropometry: the thin-fat Indian babythe Pune Maternal Nutrition Study. Int J Obes Relat Metab Disord 2003; 27: 17380. 10 Lakshmi S, Metcalf B, Joglekar C, Yajnik CS, Fall CH, Wilkin TJ. Differences in body composition and metabolic status between white UK and Asian Indian children (EarlyBird 24 and the Pune Maternal Nutrition Study). Pediatr Obes 2012; 7: 34754.

11 Bhutta ZA, Ahmed T, Black RE, et al, for the Maternal and Child Undernutrition Study Group. What works? Interventions for maternal and child undernutrition and survival. Lancet 2008; 371: 41740. 12 Save The Children. Nutrition in the first 1000 days: state of the worlds mothers 2012. May, 2012. http://www.savethechildren.ca/document. doc?id=195 (accessed March 21, 2013). 13 Cooper WN, Khulan B, Owens S, et al. DNA methylation profiling at imprinted loci after periconceptional micronutrient supplementation in humans: results of a pilot randomized controlled trial. FASEB J 2012; 26: 178290. 14 Steegers-Theunissen RP, Obermann-Borst SA, Kremer D, et al. Periconceptional maternal folic acid use of 400 microg per day is related to increased methylation of the IGF2 gene in the very young child. PLoS One 2009; 4: e7845.

14

www.thelancet.com

Series

Maternal and Child Nutrition 1 Maternal and child undernutrition and overweight in low-income and middle-income countries
Robert E Black, Cesar G Victora, Susan P Walker, Zulfiqar A Bhutta*, Parul Christian*, Mercedes de Onis*, Majid Ezzati*, Sally Grantham-McGregor*, Joanne Katz*, Reynaldo Martorell*, Ricardo Uauy*, and the Maternal and Child Nutrition Study Group

Maternal and child malnutrition in low-income and middle-income countries encompasses both under nutrition and a growing problem with overweight and obesity. Low body-mass index, indicative of maternal undernutrition, has declined somewhat in the past two decades but continues to be prevalent in Asia and Africa. Prevalence of maternal overweight has had a steady increase since 1980 and exceeds that of underweight in all regions. Prevalence of stunting of linear growth of children younger than 5 years has decreased during the past two decades, but is higher in south Asia and sub-Saharan Africa than elsewhere and globally affected at least 165 million children in 2011; wasting affected at least 52 million children. Deficiencies of vitamin A and zinc result in deaths; deficiencies of iodine and iron, together with stunting, can contribute to children not reaching their developmental potential. Maternal undernutrition contributes to fetal growth restriction, which increases the risk of neonatal deaths and, for survivors, of stunting by 2 years of age. Suboptimum breastfeeding results in an increased risk for mortality in the first 2 years of life. We estimate that undernutrition in the aggregateincluding fetal growth restriction, stunting, wasting, and deficiencies of vitamin A and zinc along with suboptimum breastfeedingis a cause of 31 million child deaths annually or 45% of all child deaths in 2011. Maternal overweight and obesity result in increased maternal morbidity and infant mortality. Childhood overweight is becoming an in creasingly important contributor to adult obesity, diabetes, and non-communicable diseases. The high present and future disease burden caused by mal nutrition in women of reproductive age, pregnancy, and children in the first 2 years of life should lead to interventions focused on these groups.

Published Online June 6, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)60937-X This is the first in a Series of four papers about maternal and child nutrition *Members listed alphabetically Members listed at end of paper Johns Hopkins University, Bloomberg School of Public Health, Baltimore, MD, USA (Prof R E Black MD, Prof P Christian DrPH, Prof J Katz ScD); Universidade Federal de Pelotas, Pelotas, Rio Grande do Sol, Brazil (Prof C G Victora MD); The University of the West Indies, Tropical Medicine Research Institute, Mona Campus, Kingston, Jamaica (Prof S P Walker PhD); The Aga Khan University and Medical Center, Department of Pediatrics, Karachi, Pakistan (Prof Z A Bhutta PhD); World

Introduction
Maternal and child malnutrition, encompassing both undernutrition and overweight, are global problems with important consequences for survival, incidence of acute and chronic diseases, healthy development, and the economic productivity of individuals and societies. Maternal and child undernutrition, including stunting, wasting, and deficiencies of essential vitamins and minerals, was the subject of a Series15 in The Lancet in 2008, which quantified their prevalence, short-term and long-term consequences, and potential for reduction through high and equitable coverage of proven nutrition interventions. The Series identified the need to focus on the crucial period of pregnancy and the first 2 years of lifethe 1000 days from conception to a childs second birthday during which good nutrition and healthy growth have lasting benefits throughout life. The 2008 Series also called for greater national priority for nutrition pro grammes, more in tegration with health programmes, enhanced inter sectoral approaches, and more focus and coordination in the global nutrition system of inter national agencies, donors, academia, civil society, and the private sector. 5 years after that series, we intend not only to reassess the problems of maternal and child lems undernutrition, but also to examine the growing prob of overweight and obesity for women and children and their consequences in low-income and middle-income countries (LMICs). Many of these countries are said to suffer the so-called double burden of malnutrition, with
www.thelancet.com

continuing stunting of growth and deficiencies of essential nutrients along with obesity in national popu lations and within families. We also want to assess national progress in nutrition programmes and inter national actions con sistent with our previous recommendations.
Key messages

Iron and calcium deficiencies contribute substantially to maternal deaths Maternal iron deficiency is associated with babies with low weight (<2500 g) at birth Maternal and child undernutrition, and unstimulating household environments, contribute to deficits in childrens development and health and productivity in adulthood Maternal overweight and obesity are associated with maternal morbidity, preterm birth, and increased infant mortality Fetal growth restriction is associated with maternal short stature and underweight and causes 12% of neonatal deaths Stunting prevalence is slowly decreasing globally, but affected at least 165 million children younger than 5 years in 2011; wasting affected at least 52 million children Suboptimum breastfeeding results in more than 800000 child deaths annually Undernutrition, including fetal growth restriction, suboptimum breastfeeding, stunting, wasting, and deficiencies of vitamin A and zinc, cause 45% of child deaths, resulting in 31 million deaths annually Prevalence of overweight and obesity is increasing in children younger than 5 years globally and is an important contributor to diabetes and other chronic diseases in adulthood Undernutrition during pregnancy, affecting fetal growth, and the first 2 years of life is a major determinant of both stunting of linear growth and subsequent obesity and non-communicable diseases in adulthood

15

Series

Health Organization, Department of Nutrition for Health and Development, Geneva, Switzerland (M de Onis MD); Imperial College of London, St Marys Campus, School of Public Health, MRC-HPA Centre for Environment and Health, Department of Epidemiology and Biostatistics, London, UK (Prof M Ezzati PhD); Institute of Child Health, University College London, London, UK (Prof S Grantham-McGregor FRCP); The University of the West Indies, Mona, Jamaica (Prof S Grantham-McGregor); Emory University, Atlanta, GA, USA (Prof R Martorell PhD); and London School of Hygiene and Tropical Medicine, London, UK (Prof R Uauy PhD) Correspondence to: Prof Robert Black, Johns Hopkins University, Bloomberg School of Public Health, Baltimore, MD 21205, USA rblack@jhsph.edu

See Online for appendix

The present Series is guided by a framework (figure 1) that shows the means to optimum fetal and child growth and development, rather than the determinants of undernutrition as shown in the conceptual model developed by UNICEF and used in the 2008 Series.1 This new framework shows the dietary, behavioural, and health determinants of optimum nutrition, growth, and development and how they are affected by underlying food security, caregiving resources, and environmental con ditions, which are in turn shaped by economic and social conditions, national and global contexts, resources, and governance. This Series examines how these determinants can be changed to enhance growth and development. These changes include nutrition-specific interventions that address the immediate causes of suboptimum growth and development. The framework shows the potential effects of nutrition-sensitive inter ventions that address the underlying determinants of malnutrition and incorporate specific nutrition goals and actions. It also shows the ways that an enabling environment can be built to support interventions and programmes to enhance growth and development and their health consequences. In the first paper we assess the prevalence of nutritional conditions and their health and development consequences. We deem a life-course perspective to be essential to conceptualise the nutritional effects and benefits of interventions. The nutritional status of women at the time of conception and during pregnancy is important for fetal growth and development, and these factors, along with nutritional status in the first 2 years of life, are important determinants of both

undernutrition in childhood and obesity and related diseases in adulthood. Thus, we organise this paper to sequences of nutritional consider prevalence and con conditions during the life course from adolescence to pregnancy to childhood and discuss the implications for adult health. In the second paper, we describe evidence supporting nutrition-specific interventions and the health effects and costs of increasing their population coverage. In the third paper we examine nutrition-sensitive inter ventions and approaches and their potential to improve nutrition. In the fourth paper we examine the features of an enabling environment that are needed to provide support for nutrition programmes and how they can be favourably changed. Finally, in a Comment6 we will examine the desired national and global response to address nutritional and developmental needs of women and children in LMICs.

Prevalence and consequences of nutritional conditions


Adolescent nutrition
Adolescent nutrition is important to the health of girls and is relevant to maternal nutrition. There are 12 billion adolescents (aged 1019 years) in the world, 90% of whom live in LMICs. Adolescents make up 12% of the population in industrialised countries, com pared with 19% in LMICs (appendix p 2 shows values for ten countries studied in depth).7 Adolescence is a period of rapid growth and maturation from childhood to adulthood. Indeed, some researchers have argued that adolescence is a period with some potential for height catch-up in children with

Benets during the life course Morbidity and mortality in childhood Cognitive, motor, socioemotional development School performance and learning capacity Adult stature Obesity and NCDs Work capacity and productivity

Nutrition specic interventions and programmes Adolescent health and preconception nutrition Maternal dietary supplementation Micronutrient supplementation or fortication Breastfeeding and complementary feeding Dietary supplementation for children Dietary diversication Feeding behaviours and stimulation Treatment of severe acute malnutrition Disease prevention and management Nutrition interventions in emergencies

Optimum fetal and child nutrition and development

Breastfeeding, nutrientrich foods, and eating routine

Feeding and caregiving practices, parenting, stimulation

Low burden of infectious diseases

Food security, including availability, economic access, and use of food

Feeding and caregiving resources (maternal, household, and community levels)

Access to and use of health services, a safe and hygienic environment

Nutrition sensitive programmes and approaches Agriculture and food security Social safety nets Early child development Maternal mental health Womens empowerment Child protection Classroom education Water and sanitation Health and family planning services

Knowledge and evidence Politics and governance Leadership, capacity, and nancial resources Social, economic, political, and environmental context (national and global)

Building an enabling environment Rigorous evaluations Advocacy strategies Horizontal and vertical coordination Accountability, incentives regulation, legislation Leadership programmes Capacity investments Domestic resource mobilisation

Figure 1: Framework for actions to achieve optimum fetal and child nutrition and development

16

www.thelancet.com

Series

stunting from early childhood.8 Adolescent fertility is three times higher in LMICs than in high-income coun tries. Preg nancies in adolescents have a higher risk of complications and mortality in mothers9 and children10 and poorer birth outcomes than pregnancies in older women.10,11 Further more, pregnancy in adolescence will slow and stunt a girls growth.12,13 In some countries, as many as half of adolescents are stunted (height-for-age Z score [HAZ] <2), increasing the risk of poor perinatal outcomes in their offspring (appendix p 2). We used age-specific, low body-mass index (BMI) cutoffs (BMI Z score [BMIZ] <2) from the WHO reference for children aged 519 years to examine ten selected countries; in these locations as many as 11% (India) of adolescent girls are thin. In these countries, prevalence of high BMI for age, defined as BMIZ >2 (obesity), is as high as 5% (Brazil; appendix p 2). Adolescents have as high a prevalence of anaemia as women aged 2024 years.
Africa 600 500 Prevalence (%) 400 300 200 100 0 Asia 600 500 Prevalence (%) 400 300 200 100 0 Oceania 600 500 Prevalence (%) 400 300 200 100 0 BMI 185 BMI 25 BMI 30 1980 1985 1990 1995 2000 2005 2008

In India, for example, 558% of adolescents aged 1519 years and 567% of women aged 2024 years were anaemic;14 corresponding values for Guatemala were 210% and 204 %, respectively.15

Maternal nutrition
Prevalence of low BMI (<185 kg/m) in adult women has decreased in Africa and Asia since 1980, but remains higher than 10% in these two large developing regions (figure 2). During the same period, prevalence of overweight (BMI 25 kg/m) and obesity (BMI 30 kg/m) has been rising in all regions, together reaching more than 70% in the Americas and the Caribbean and more than 40% in Africa by 2008.16,17 Few studies have examined the risk of maternal mortality in relation to maternal anthropometry with a prospective design. In one study in Nepal18 of about 22 000 women, mid upper arm circumference during
Americas and the Caribbean

Europe

Global

BMI 185

BMI 25

BMI 30

Figure 2: Trends in thinness (BMI <185 kg/m), overweight (BMI 25 kg/m), and obesity (BMI 30 kg/m), using population weighted average prevalences for women aged 2049 years UN regions and globally, 19802008 Error bars are 95% CIs. BMI=body-mass index.

www.thelancet.com

17

Series

pregnancy was inversely associated with all-cause mater nal mortality up to 42 days post-partum after adjust ing for numerous factors. An inverse association exists between maternal height and the risk of dystocia (difficult labour), as measured by cephalopelvic dispro portion or assisted or caesarean deliveries.1923 Figure 2 also shows trends for overweight and obesity in women aged 2049 years in different UN regions. Oceania, Europe, and the Americas had the highest proportion of overweight and obese women; however northern and southern Africa, and central and west Asia also had high prevalences (appendix p 3). Maternal obesity leads to several adverse maternal and fetal complications during pregnancy, delivery, and post-partum.24 Obese pregnant women (pre-pregnant BMI 30 kg/m) are four times more likely to develop gestational diabetes mellitus and two times more likely to develop pre-eclampsia compared with women with a BMI 185249 kg/m2).2528 During labour and delivery, maternal obesity is associated with maternal death, haemor rhage, caesarean delivery, or infection;2931 and a higher risk of neonatal and infant death,32 birth trauma, and macrosomic infants.3337 In the post-partum period, obese women are more likely to delay or fail to lactate and to have more weight retention than women of normal weight.38 Obese women with a history of gestational diabetes have an increased risk of subsequent type two diabetes, metabolic syndrome, and cardio vascular disease.39 The early intra uterine environment has a role in programming pheno type, affecting health in later life. Maternal overweight and obesity at the time of pregnancy increases the risk for childhood obesity that continues into adoles cence and early adulthood, potentiating the trans generational trans mission of obesity.40,41
Vitamin A deficiency45

Maternal vitamin deficiencies


Anaemia and iron
Anaemia (haemoglobin <110 g/L), which might be attributable to low consumption or absorption in the diet or to blood loss, such as from intestinal worms, is highly prevalent during pregnancy. This Series focuses on anaemia amenable to correction with iron supple men tation.42 To establish the importance of iron deficiency as a cause of maternal anaemia we used the results of trials of iron supplementation to work out the shift in the lation haemoglobin distribution. A meta-analysis of popu the effects of iron supple mentation trials showed that, among pregnant women with anaemia at baseline, iron supple mentation led to a 102 g/L increase in haemo globin.43 The corresponding figure for children was 80 g/L.44 We applied these shifts to the present dis tri butions of haemoglobin estimated by Stevens and colleagues42 and calculated the proportion of pregnant women with anaemia whose blood haemoglobin would increase to at least 110 g/L. We likewise calculated the proportion of severe anaemia that would increase to at least 70 g/L. These results constitute the prevalence of iron amenable or iron deficiency anaemia (IDA) or severe IDA, defined as the proportion of anaemia or severe anaemia that would be reduced if only iron was provided, holding other determinants of anaemia unchanged. With this approach in 2011, Africa had the highest proportion of IDA for pregnant women followed closely by Asia (table 1,4547 appendix p 4). Likewise, Africa had the highest prevalence of severe IDA, but in all regions prevalence was less than 1%. Our previous analyses48 showed that anaemia in preg nancy increased the risk of maternal mortality. An updated analysis49 with ten studies (four more than the
Zinc deficiency47 Iron deficiency anaemia (haemoglobin Iodine <110 g/L) deficiency46 (weighted (UIC <100 g/L) average of country means)

Children <5 years Night blindess Global Africa Americas and the Carribean Asia Europe Oceania 09% (0118) 21% (1031) 06% (0013) 05% (0013) 07% (0015) 05% (0110) Serum retinol <070 mol/L 333% (294371) 416% (344449) 156% (66245) 335% (307363) 149% (01297) 126% (60192)

Pregnant women Night blindness 78% (6591) 94% (81107) 44% (2762) 78% (6690) 29% (1146) 92% (03182) Serum retinol <070 mol/L 153% (60246) 143% (97190) 20% (0436) 184% (54314) 22% (0043) 14% (0040) 285% (282289) 400% (394406) 137% (125148) 316% (307325) 442% (435450) 173% (166181) 173% (159188) 239% (211268) 96% (68124) 194% (169220) 76% (6291) 57% (10103)

Children <5 years

Pregnant women

181% (156208) 202% (186217) 127% (98160) 190% (145234) 121% (78162) 154% (70252)

192% (171215) 203% (183224) 152% (117186) 198% (158235) 162% (126197) 172% (97256)

Data are % (95% CI). UIC=urine iodine concentration.

Table 1: Prevalence of vitamin A deficiency (19952005), iodine deficiency (2013), inadequate zinc intake (2005), and iron deficiency anaemia (2011)

18

www.thelancet.com

Series

previous analysis) showed that the odds ratio (OR) for maternal deaths was 071 (95% CI 060085) for a 10 g/L greater mean haemoglobin in late pregnancy. Only two of the ten studies adjusted for socioeconomic con founding variables; one showed no attenuation and the other a 20% attenuation of the effect. There is strong biological plausibility for a causal link between maternal IDA and adverse birth outcomes includ ing low birthweight and increased perinatal mortality.5052 A meta-analysis53 that included 11 trials indenti fied a significant 20% reduction in the risk of low birthweight associated with antenatal supplementation with iron alone or combined with folic acid (relative risk [RR] 080, 95% CI 071090). A previous Cochrane review54 had much the same findings. Dibley and colleagues55 pooled data from demographic and health surveys from Indonesia for 1994, 1997, 200203, and 2007 and showed that risk of death of children younger than 5 years was reduced by 34% when the mother consumed any iron-folic acid supplements (hazard ratio [HR] 066; 95% CI 053081). Dibley and colleagues further showed that the protective effect was greatest for deaths on the first day of life (040; 021077), but the protective effect was also shown for neonatal deaths (069; 95% CI 049097) and post-neonatal deaths (074; 056099). A randomised controlled trial56 from China showed a significant 54% (RR 046, 95% CI 021098) reduction in neonatal mortality with antenatal iron and folic acid supplementation compared with folic acid alone as control. In Nepal, mortality from birth to 7 years was reduced by 31% (HR 069, 95% CI 049099) in the offspring of mothers who had received iron and folic acid during pregnancy compared with controls who received vitamin A only.57 Randomised controlled trials from high-income coun tries have shown benefits of iron supplementation for improved maternal mental health and reduced fatigue.58 Evidence from LMICs for the effect of mater nal IDA on mothers mental health and mother-child interactions is scarce. In a small South African trial,59 iron supplemen tation of women with IDA from 10 weeks post-partum to 9 months led to lower maternal depression and perceived stress at 9 months compared with placebo. At 9 months, iron supplemented mothers had better maternal-child interactions.60 By contrast, in Bangladesh higher levels of maternal iron supple mentation decreased the quality of maternal-child interaction at age 34 months and had no effect on maternal distress (anxiety and depression).61 There is some evidence for whether maternal IDA affects child development. Infants of mothers identified as having IDA at 68 weeks post-partum had lower developmental levels at 10 weeks and 9 months compared with infants of control mothers without IDA.60 In Nepal, children whose mothers received iron and folate supple mentation during pregnancy had better general intelli gence and cognitive functioning at age 79 years compared with children of mothers receiving placebo,
www.thelancet.com

suggesting that benefits can be detected in later childhood when more complex tasks can be measured.62

Vitamin A
Maternal vitamin A deficiency can cause visual impair ment and possibly other health consequences. Deficiency is assessed in pregnant women as either a history of night blindness or serum or plasma retinol concentrations of less than 070 mol/L (subclinical vitamin A deficiency). WHO provides prevalence estimates for 19952005 from 64 countries, which we used for estimates for the UN world regions (table 1).45 Globally, the prevalence of night blindness in pregnant women is estimated to be 78% (95% CI 7087), affecting 97 million women. An estimated 153% (74232) of pregnant women globally (191 million women) have deficient serum retinol con centrations. The degree to which night blindness and low serum retinol overlap is not accounted for in this esti mation, but night blindness is known to be associated with a four-times higher odds of low serum retinol (OR 402, 95% CI 2274).63 Night blindness is reduced by vitamin A supplementation in pregnancy.63,64 Maternal night blindness has been associated with increased low birthweight64 and infant mortality,65 yet trials of vitamin A in pregnancy have not showed significant effects on these outcomes.6669

Zinc
Zinc is a key micronutrient with a ubiquitous role in biological functions, including protein synthesis, cellular division, and nucleic acid metabolism. Esti mates revised in 2012 suggest that 17% of the worlds population is at risk of zinc deficiency, on the basis of an analysis of national diets.47 Excess losses of zinc during diarrhoea also contribute to zinc deficiency. The effect of subclinical zinc deficiency (defined as low plasma zinc concentration without obvious signs of zinc deficiency) in women of reproductive age and during pregnancy on health and development outcomes is poorly understood, although zinc deficiency has been suggested as a risk factor with adverse long-term effects on growth, immunity, and metabolic status of surviving offspring.70 Zinc deficiency due to a rare genetic abnormalityacrodermatitis entero pathicain pregnancy results in a high risk of preterm and prolonged labour, post-partum haemor rhage, and fetal growth restriction.70,71 A review of supple mentation trials with zinc in pregnancy showed a significant 14% reduction in preterm births in women in low-income settings, but no significant effect on low birthweight.72

Iodine
Maternal iodine deficiency is of concern in regard to adverse effects on fetal development, yet few countries have nationally representative data from large-scale surveys of urinary iodine concentration in pregnant women. Because of the correlation between urinary iodine concentration in pregnant women and children
19

Series

aged 612 years (r 069),73 status assessment in schoolage children is used to estimate country, regional, and global prevalence of iodine deficiency. Global estimates of iodine deficiency suggest that 285% of the worlds population or 19 billion individuals are iodinedeficient (table 1).46,74 This figure represents largely mild deficiency (defined as urinary iodine concentration of 5099 ug/L). Severe iodine deficiency in pregnancy causes cretin ism, which can be eliminated with iodine supple men tation before conception or in the first trimester of pregnancy.75 Furthermore, two meta-analyses showed average deficits of 125135 intelligence quotient (IQ) points in children associated with iodine deficiency of their mothers in pregnancy; however, they controlled for only a limited number of socioeconomic con founders.76,77 A review of the effects of iodine supple mentation in deficient populations showed a small increase in birth weight.78 Effects of mild or moderate iodine deficiency on brain development are not well established.78,79 The index of iodine deficiency (urinary iodine concentration) is a population measure and not an individual one,80 therefore some individuals in regions of mild to moderate deficiency might have more severe deficiency.

Folate
The global prevalence of folate deficiency has not been estimated because of the scarcity of suitable populationbased data.81 A substantial proportion of neural tube defects (congenital malformations of the spinal cord and brain) are related to inadequate consumption of folic acid around the time of conception, in some populations associated with genetic factors that increase the need for dietary folic acid. A Cochrane review82 in 2010 included five trials of folic acid (a synthetic form of folate) supplementation and identified a 72% (RR 028; 95% CI 015052) reduction in the risk of neural tube defects. A more recent systematic review had much the same findings and estimated that in 2005 56000 deaths were attributable to insufficient dietary folic acid.83 These deaths were not added to the total deaths associated with undernutrition in the present analysis, because of the uncertainty about this estimate.

Calcium and vitamin D


Calcium is an essential nutrient for several body func tions, including enzymatic and hormonal homoeo stasis. Evidence for the association between maternal dietary calcium intake and maternal bone density and fetal mineralisation is inconsistent.84 Epidemiological evi dence does show an inverse association between calcium intake and development of hypertension in pregnancy.85,86 Gestational hypertensive disorders are the second leading cause of maternal morbidity and mortality and are associated with increased risk of preterm birth and fetal growth restriction.87,88
20

Substantial evidence suggests that calcium supple mentation in pregnancy is associated with a reduction in gestational hypertensive disorders and preterm birth.89,90 However, the effect varies according to the baseline calcium intake of the population and preexisting risk factors. A review of 1591 randomised con trolled trials suggested that calcium supplementation during preg nancy was associated with a reduction in the risk of gestational hypertension and a 52% reduction in the incidence of pre-eclampsia, along with a 24% reduction in preterm birth and an increase in birth weight of 85 g. There was no effect on low birthweight or perinatal or neonatal mortality. The effect was mainly noted in populations with low calcium intake.92 The effects of calcium supplementation interventions are described in the accompanying report by Zulfiqar A Bhutta and colleagues.93 The US Institute of Medicine has defined adequate vitamin D status as having serum 25-hydroxyvitamin D ([OH]D) concentrations greater than 50 nmol/L in both the general population and pregnant women;94 serum concentrations of less than 25 nmol/L denote vita min D deficiency whereas concentrations of less than 50 nmol/L denote vitamin D insufficiency.95 Although few nationally representative surveys exist for vitamin D status, an estimated 1 billion people globally residing in diverse geographies, many in LMICs, might be vitamin D insufficient or deficient.96103 Vitamin D has an essential role in fetal development, ensuring fetal supply of calcium for bone development, enabling immunological adaptation required to main tain normal pregnancy, preventing miscarriage, and promot ing normal brain development.99,104108 Poor maternal vitamin D status has been associated with severe pre-eclampsia (new-onset gestational hyper tension and protein uria after 20 weeks of gestation) in turn leading to an increased risk of perinatal morbidity and mortality.99,109,110 Maternal vitamin D deficiency, especially in early pregnancy, has been associated with risk of pre-eclampsia (OR 209, 95% CI 150290), preterm birth (158, 108231), and smallfor-gestational age (SGA; 152, 108225).109,111 A systematic review of three trials of vitamin D in pregnancy showed an overall reduction of low birthweight of border line significance (relative risk [RR] 048; 95% CI 023101).111 These asso ciations need to be better quantified before they can be included in the global disease burden related to undernutrition.

Effect of maternal stature or BMI on fetal growth restriction or postnatal growth


Maternal characteristics
The 2008 Maternal and Child Undernutrition Series examined the association of maternal nutritional status (BMI and short stature) and fetal growth restriction, defined as low birthweight at term. Here, we use data from nine (height) and seven (BMI) population-based
www.thelancet.com

Series

cohort studies and WHO perinatal facility-based data for 24 countries to examine associations separately for term and preterm SGA in LMICs.112121 Maternal stunting (height <145 cm) put infants at risk of term and preterm SGA (appendix p 5). Low maternal BMI in early pregnancy also put infants at higher risk of SGA (appendix p 5). BMI of 25 or greater was somewhat protective against term and preterm SGA (appendix p 5). Notably, most women in the BMI category of greater than 25 kg/m were very mildly overweight with very few obese women, which might explain the protective effect. Maternal stature is a composite indicator representing genetic and environmental effects on the growing period of childhood. In a study122 involving 109 Demographic Health Surveys, analyses adjusted for wealth, education, and urban or rural residence showed that the absolute risk of dying among children younger than 5 years born to the tallest mothers (160 cm) was 0073 (95% CI 00720074) and to those born to the shortest mothers (<145 cm) was 0128 (01260130). The correspond ing absolute risk for a child being stunted was 0194 (01920196) for the tallest mothers and 0682 (06730690) for the shortest. The association with wasting was significant but much weaker.122

studies that use different reference populations for SGA. Previous studies have also not separated SGA into term and preterm. In a pooled-analysis130 of 22 populationbased cohort studies in LMICs in Asia, sub-Saharan Africa, and Latin America, the RR for neonatal (128 day) mortality associated with SGA (<10th centile) was 183 (95% CI 134250) and for post-neonatal (29365 day) mortality was 190 (132273), compared with appropriate-for-gestational-age (AGA) infants. The RR for term SGA was 306 (95% CI 221423) for neonatal mortality and 198 (139281) for post-neonatal
350 300 250 Proportion (%) 200 150 100 50 0 Prevalence preterm SGA Prevalence term SGA

Fetal growth restriction


Previous global and regional estimates of fetal growth restriction used the term low birthweight as a proxy for being SGA in the absence of population-based birthweight and gestational age data at that time.1 Through recent analy ses, we now have estimates of SGA prevalence from 22 population-based cohort studies and 23 countries with facility-based data,114 which were used to model SGA as a function of low birthweight and other covariates (neonatal mortality rate, representativeness of facility delivery) to obtain country-specific SGA prevalence for 2010.123 The numbers on which the model is based include all livebirths, but exclude babies who died so soon after birth that they were not weighed. Imputing birthweight for these infants did not change the estimation of SGA prevalence, although it did increase the mortality risk associated with being born SGA. Figure 3 shows preva lence of SGA separated into term and preterm SGA. By our estimate, in 2010 324 million babies were born SGA, 27% of all births in LMICs regions. When com paring the estimated numbers of children with SGA with those affected by stunting or wasting, it is important to note that the cutoff for SGA is the 10th centile of a reference population, whereas the cutoffs for wasting or stunting are two Z scores below the median, or the 23rd centile. Appendix p 8 shows prevalence with 95% CIs of term and preterm SGA by UN subregions. Notably, only 20% of preterm births in LMICs were also SGA. Appendix p 8 show the ten countries with the highest number of SGA births in 2010. Associations between fetal growth restriction and infant survival have been previously reported,124129 although it is difficult to compare associations across
www.thelancet.com

Africa

Asia

Latin America and the Caribbean

Oceania

Figure 3: Prevalence of small-for-gestational-age births, by UN regions Attributable deaths with UN prevalences* Proportion of total deaths of children younger than 5 years 118% 147% 144% 126% 74% 17% 23% 116% 194% Attributable deaths with NIMS prevalences Proportion of total deaths of children younger than 5 years 118% 170% 170% 115% 78% 17% 23% 116% 194%

Fetal growth restriction (<1 month) Stunting (159 months) Underweight (159 months) Wasting (159 months) Severe wasting (159 months) Zinc deficiency (1259 months) Vitamin A deficiency (659 months) Suboptimum breastfeeding (023 months) Joint effects of fetal growth restriction and suboptimum breastfeeding in neonates Joint effects of fetal growth restriction, suboptimum breastfeeding, stunting, wasting, and vitamin A and zinc deficiencies (<5 years)

817000 1017000* 999000* 875000* 516000* 116000 157000 804000 1348000

817000 1179000 1180000 800000 540000 116000 157000 804000 1348000

3097000

447%

3149000

454%

Data are to the nearest thousand. *Prevalence estimates from the UN. Prevalence estimates from Nutrition Impact Model Study (NIMS).

Table 2: Global deaths in children younger than 5 years attributed to nutritional disorders

21

Series

mortality, relative to term AGA infants. Infants born preterm and SGA were at highest risk with RR of 1542 (9112612) for neo natal mortality and 522 (283964) for post-neonatal mortality, relative to term AGA. By applying the attributable fractions of deaths to the total neonatal and post-neonatal deaths for 2011 obtained from the UN Interagency Group on Mortality Esti mation,131 we estimated that the number of deaths 000 in neonates and attributed to SGA in 2011 was 817 418 000 in infants aged 111 months using standard methods.1 The largest number of attributed deaths were in Asia (appendix p 22). For the calculation in the group who were both SGA and preterm, the counterfactual was AGA and preterm so that the SGA effect was separated from that of being preterm. In the estimation of the deaths attributed to several nutritional conditions, we attributed only the neonatal deaths to SGA; for children aged 159 months, we attributed deaths to wasting and stunting, to which SGA contributed (table 2).
Panel 1: Determinants of childhood stunting and overweight The determinants of optimum growth and development (figure 1) consist of factors operating at different levels of causation, ranging from the most distal socioeconomic and political determinants to the proximate level where food, disease, and care have a crucial role. A mirror image of figure 1 would show the determinants of linear growth failurethe process leading to child stuntingand overweight. The large socioeconomic inequalities in stunting prevalence in almost all low-income and middle-income countries (LMICs) show the great importance of distal determinants. In particular, maternal education is associated with improved child-care practices related to health and nutrition and reduced odds of stunting, and better ability to access and benefit from interventions. Almost all stunting takes place in the first 1000 days after conception. The few randomised controlled trials of breastfeeding promotion145 that included nutritional status outcomes did not show any effects on the weight or length of infants. By contrast, there is strong evidence that the promotion of appropriate complementary feeding practices reduces the incidence of stunting.93 A meta-analysis of zinc supplementation trials146 has shown a significant protective effect against stunting. Severe infectious diseases in early childhoodsuch as measles, diarrhoea, pneumonia, meningitis, and malaria can cause acute wasting and have long-term effects on linear growth. However, studies have consistently shown that diarrhoea is the most important infectious disease determinant of stunting of linear growth. In a pooled analysis147 of nine community-based studies in low-income countries with daily diarrhoea household morbidity and longitudinal anthropometry, the odds of stunting at 24 months of age increased multiplicatively with each diarrhoea episode or day of diarrhoea before that age. The

Causes of childhood growth faltering are multifactorial, but fetal growth restriction might be an important contributor to stunting and wasting in children. To quantify the association between fetal growth restriction and child undernutrition, we did a systematic scientific literature search to identify longitudinal studies that had taken measurements of birthweight, gestational age, and child anthropometry. Data could be obtained from 19 birth cohort studies which were submitted to a meta-analysis to examine the odds of stunting, wasting, and underweight in children 1260 months of age associated with SGA and preterm birth.132 Four mutually exclusive exposure cate gories were created: AGA and preterm; SGA and term; SGA and preterm; and term AGA (as the reference group). The meta-analysis showed that SGA alone and preterm alone were associated with increased overall ORs of 243 (95% CI 222266) and 193 (171218), respectively, whereas both SGA and preterm increased the OR to 451 (342593) for stunting (appendix p 23). These raised

proportion of stunting attributed to five previous episodes of diarrhoea was 25% (95% CI 838). Environmental (or tropical) enteropathy is an acquired disorder, characterised by reduced intestinal absorptive capacity, altered barrier integrity, and mucosal inflammation, occurring in young children living in unsanitary settings.148 These children also have high rates of symptomatic and asymptomatic infections with enteric pathogens, but the exact association of these infections or of other possible toxic enteric exposures with enteropathy is unclear. Some researchers have suggested that these functional changes and the associated inflammation have significant adverse effects on the growth of children. Alternatively, these changes might be a consequence of nutritional deficits very early in life, including in utero, that lead to intestinal microbial colonisation. Optimum growth in the first 1000 days of life is also essential for prevention of overweight. Whereas attained weight at any age in early life is positively associated with adult body-mass index in LMIC cohorts,2,149,150 rapid weight gains in the first 1000 days are strongly associated with adult lean mass, whereas weight gains later in childhood lead mainly to adult fat mass. In particular, evidence suggests that infants whose growth faltered in early life, and who gained weight rapidly later in childhood, might be at particular risk of adult obesity and non-communicable diseases.2 Child overweight is also related to growing up in an obesogenic environment, in which population changes in physical activity and diet are the main drivers. Modifiable risk factors for childhood obesity are maternal gestational diabetes; high levels of television viewing; low levels of physical activity; parents inactivity; and high consumption of dietary fat, carbohydrate, and sweetened drinks, yet few interventions have been rigorously tested.151153

22

www.thelancet.com

Series

ORs were also noted in the Asian, African, and Latin American UN regions (data not shown). We estimated population attributable risk for childhood stunting for the risk categories of SGA and preterm birth. Because risk estimates were derived as ORs using logistic regression analysis, we used a method to approximate the risk ratio proposed by Zhang and Yu133 for estimation of the population attributable risk. Using the approximated RR estimates across all 19 cohorts, population attributable risk for SGA-term for stunting was 016 (012019), that for SGA-preterm was 004 (002005), and that for AGA-preterm was 004 (002006). The combined population attributable risk related to SGA for stunting was 020 and that for preterm birth was 008. Thus, overall we estimate that about a fifth of childhood stunting could have its origins in the fetal period, as shown by being born SGA. Most studies of fetal growth restriction and childhood cognitive and motor development in LMICs involve term infants of low birthweight or examine birthweight adjusted for gestational age. Consistent evidence exists for associations of fetal growth restriction with lower psychomotor development levels in early childhood (up to age 36 months) with small to moderate effect sizes compared with infants of normal birthweight.134,135 A study136 from Bangladesh showed much the same associations for both mental and motor development. Evidence suggests that there is an effect on development
Stunting (HAZ <2) UN155 Pro Number portion (millions) Africa 356% (333 380) 268% (232 305) 134% (94 177) 563 (525 600) 958 (828 1088) 71 (4894) NIMS158 Pro Number portion (millions) 355% (344 366) 295% (264 313) 146% (136 155) 566 (543 578) 1035 (925 1098) 78 (7382) Wasting (WHZ <2) UN Pro Number portion (millions) 85% (74 96) 101% (79 123) 14% (09 19) 134 (116 152) 361 (282 440) 07 (0510)

that is attributable to birth size, independent of that attributable to poor postnatal growth.137 Evidence for effects of fetal growth restriction on cognition and behaviour after early childhood is less consistent. Birthweight was associated with attained schooling in the COHORTS analyses;138 however, this was unadjusted for gestational age. Size at birth was not related to womens educational achievement in Guatemala139 and term low birthweight was not asso ciated dren in Brazil140 with IQ and behaviour in school-aged chil and Jamaica,141 or behaviour in South Africa.142 In Taiwan, term infants of low birthweight had lower academic achievement at age 15 years than did infants of normal birthweight,143 and in Thailand birth length was associated with IQ at age 9 years independent of postnatal growth to age 1 year;144 however, in both cases effect sizes were small.

Childhood nutrition

Stunting, underweight, and wasting


Panel 12,145153 describes the determinants of stunting and overweight in children. Estimates of the prevalence of stunting, underweight, and wasting worldwide and for UN subregions are based on analyses jointly done by UNICEF, WHO, and the World Bank154 of 639 national surveys from 142 countries in the WHO database, using standard methods.1,155,156 In 2011, globally, 165 million children younger than 5 years had a height-for-age Z score (HAZ) of 2 or lower (stunted) on the basis of
Severe wasting (WHZ <3) NIMS Pro Number portion (millions) 79% (73 86) 100% (82 114) 15% (13 18) 125 (115 136) 52 (4359) 08 (0710) UN Pro Number portion (millions) 35% (29 41) 36% (24 48) 03% (02 04) 55 (4564) 129 (84 173) 02 (0102) NIMS Pro Number portion (millions) 28% (25 33) 36% (27 47) 04% (04 06) 44 (4052) 127 (94 164) 02 (0203) Underweight (WAZ <2) UN Pro portion 177% (157 197) 193% (146 241) 34% (2345) Number (millions) 279 (247 311) 691 (521 861) 18 (1224) NIMS Pro Number portion (millions) 184% (174 191) 219% (188 240) 37% (35 41) 290 (275 301) 766 (659 841) 20 (1822)

Asia

Latin America and the Carib bean Oceania

355% (160 614) 28% (256 304) 72% (41 126) 257% (235 279)

05 (0208) 1597 (1459 1734) 51 (2989) 1648 (1508 1788)

347% (278 395) 299% (279 310)

04 (0305) 1683 (1573 1746)

43% (30 62) 88% (74 103) 17% (08 35) 80% (68 93)

01 (0001) 503 (421 584) 12 (0625) 515 (433 596)

51% (33 68) 93% (84 104)

01 (0001) 526 (474 585)

07% (05 11) 33% (25 40) 03% (00 13) 29% (22 36)

00 (0000) 185 (140 231) 02 (0009) 187 (142 232)

15% (09 24) 31% (26 38)

00 (0000) 173 (144 215)

140% (80 232) 174% (143 204) 24% (1734) 157% (130 184)

02 (0103) 990 (817 1163) 17 (1224) 1007 (833 1180)

139% (107 168) 194% (173 205)

02 (0102) 1091 (972 1154)

LMICs

Highincome countries Global

Data are % (95% CI). HAZ=height-for-age Z score. WHZ=weight-for-height Z score. WAZ=weight-for-age Z score. LMICs=low-income and middle-income countries.

Table 3: Prevalence and numbers of children younger than 5 years with stunting, wasting, severe wasting, and underweight using estimates from UN and NIMS, by UN regions for 2011

www.thelancet.com

23

Series

the WHO Child Growth Standards (table 3)a 35% decline from an estimated 253 million in 1990. The prevalence decreased from an estimated 40% in 1990, to an estimated 26% in 2011an average annual rate of reduction of 21% per year (figure 4154). East and west Africa, and south-central Asia have the highest preva lence estimates in UN subregions with 42% (east Africa) and 36% (west Africa and south-central Asia); the largest number of children affected by stunting, 69 million, live in south-central Asia (appendix p 9). The surveys in the WHO database and other populationrepresentative data were also analysed with a Bayesian hierarchical mixture model to estimate Z-score distri butions of anthropometric indices by the Nutrition Impact Model Study (NIMS).157 These distributions were used to assess trends in stunting and underweight in children, the present prevalence of these measures, and the present prevalence of wasting. These methods have the advantage of esti mating the full distribution of anthropometric variables and therefore measure the full extent of mild-to-severe undernutrition without restrictive assumptions. They also allow for non-linear time trends. NIMS analyses resulted in much the same estimates of the prevalence of stunting, underweight, and wasting as those of the UN in 2011 (table 3, appendix p 10). The complete trend analysis showed that the largest reductions in stunting since 1985 have been in Asia, whereas Africa had an increase until the mid-1990s and subsequently a modest reduction in the prevalence.157
Global 300 270 240 Numbers (millions) 210 180 150 120 90 60 30 0 Africa 2531 2263 2010 1812 1671 1535 Global target =100 million 1402 Numbers Prevalence 60 55 50 45 40 35 30 25 20 15 10 5 0

However, with the increase in population in Africa, this is the only major world region with an increase in the number of stunted children in the past decade. The complex interplay of social, economic, and political determinants of undernutrition (figure 1) results in substantial inequalities between population subgroups. In our analysis, using previously described methods,158 of 79 countries with population-based surveys since the year 2000 (figure 5), stunting prevalence among children younger than 5 years was 247 times (range 100764) higher in the poorest quintile of households than in the richest quintile. Sex inequalities in child nutrition tend to be substantially smaller than economic inequalities (appendix p 24). In 81 countries with data, stunting prevalence is slightly higher (114 times, range 083153) in boys than in girls. This finding is consistent with the higher mortality in children younger than 5 years in boys than in girls in most countries in the world. Place of residence is also an important correlate of the risk of stunting (appendix p 24). In 81 countries with data, stunting was 145 times higher (range 094 to 294) in rural than in urban areas. According to UN estimates, globally in 2011, more than 100 million children younger than 5 years, or 16%, were underweight (weight-for-age Z score [WAZ] <2 on the basis of the WHO Child Growth Standards), a 36% decrease from an estimated 159 million in 1990.154 Estimated prevalences in NIMS were slightly higher at
Asia 300 270 240 Numbers (millions) 210 180 150 120 90 60 30 0 Latin America and the Caribbean 1887 1615 1356 1139 984 833 692 60 55 50 45 40 35 30 25 20 15 10 5 0

Prevalence (%)

Prevalence (%)

1274

565

300 270 240 Numbers (millions) 210 180 150 120 90 60 30 0

457

479

502

533

558

581

599

606

1990

1995

2000

2005 2010 Year

2015

2020

2025

60 55 50 45 40 35 30 25 20 15 10 5 0

300 270 240 Numbers (millions) 210 180 150 120 90 60 30 0


Prevalence (%)

137 1990

121 1995

104 2000

90

74

62 2015

51 2020

42 2025

2005 2010 Year

60 55 50 45 40 35 30 25 20 15 10 5 0

Prevalence (%)

Figure 4: Trends in prevalence and numbers of children with stunted growth (HAZ <2), by selected UN regions and globally, 19902010, and projected to 2025 on the basis of UN prevalence estimates HAZ=height-for-age Z score. Data from UNICEF, WHO, World Bank.154

24

www.thelancet.com

Series

110 million (194%).155 Prevalences were highest in southcentral Asia and western Africa where 30% and 22%, respectively, were underweight (appendix p 9). The UN estimate for wasting (weight-for-height Z score [WHZ] <2 on the basis of WHO Child Growth
Stunting (HAZ <2)
Guatemala (DHS 1998) Timor-Leste (DHS 2009) India (DHS 2005) Madagascar (DHS 2005) Laos (MICS 2006) Niger (DHS 2006) Nepal (DHS 2011) Malawi (DHS 2010) Peru (DHS 2004) Bangladesh (DHS 2007) Rwanda (DHS 2010) Chad (DHS 2004) Nigeria (DHS 2008) Somalia (MICS 2006) Cameroon (MICS 2006) Mozambique (MICS 2006) Honduras (DHS 2005) Benin (DHS 2006) Cambodia (DHS 2010) Central African Republic (MICS 2006) Ethiopia (DHS 2011) Guinea Bissau (MICS 2006) Tanzania (DHS 2010) Zambia (DHS 2007) Burkina Faso (MICS 2006) Cte dIvoire (MICS 2006) Democratic Republic of the Congo (DHS 2007) Bolivia (DHS 2008) Lesotho (DHS 2009) Guinea (DHS 2005) Mali (DHS 2006) Kenya (DHS 2008) Liberia (DHS 2007) Uganda (DHS 2006) Nicaragua (DHS 2001) Gabon (DHS 2000) Haiti (DHS 2005) Tajikistan (MICS 2005) Namibia (DHS 2006) Belize (MICS 2006) So Tom and Prncipe (DHS 2008) Togo (MICS 2006) Congo (Brazzaville) (DHS 2005) Swaziland (DHS 2006) The Gambia (MICS 2005) Syria (MICS 2006) Morocco (DHS 2003) Zimbabwe (DHS 2010) Senegal (DHS 2010) Sierra Leone (DHS 2008) Mauritania (MICS 2007) Ghana (DHS 2008) Azerbaijan (DHS 2006) Mongolia (MICS 2005) Guyana (DHS 2009) Turkey (DHS 2003) Egypt (DHS 2008) Vanuatu (MICS 2007) Palestinians in Lebanon (MICS 2006) Albania (DHS 2008) Armenia (DHS 2010) Georgia (MICS 2005) Maldives (DHS 2009) Uzbekistan (MICS 2006) Kazakhstan (MICS 2006) Thailand (MICS 2005) Kyrgyzstan (MICS 2005) Colombia (DHS 2010) Bosnia and Herzegovina (MICS 2006) Jordan (DHS 2007) Suriname (MICS 2006) Dominican Republic (DHS 2007) Macedonia (MICS 2005) Moldova (DHS 2005) Montenegro (MICS 2005) Serbia (MICS 2005) Belarus (MICS 2005) Brazil (DHS 2006)

Standards) was 8% (52 million) globally in 2011, an 11% decrease from an estimated 58 million in 1990.154 70% of the worlds children with wasting live in Asia, most in south-central Asia, where an estimated 15% (28 million) are affected. Much the same regional pattern occurs for
Overweight (HAZ >2)

10

20

30

40

50

60

70

80

90 100

10

20

30

40

50

60

70

80

90 100

Prevalence (%)

Figure 5: Prevalence of stunting (HAZ <2 Z scores below median) and overweight (BMI for age >2 Z scores above median) for highest and lowest wealth quintiles in selected countries Blue circles are lowest wealth quintiles, red circles are highest wealth quintiles. HAZ=height-for-age Z score. BMI=body-mass index. DHS=Demographic and Health Survey. MICS=Multiple Indicator Cluster Survey.

www.thelancet.com

25

Series

severe wasting (WHZ <3), with a global prevalence in 2011 of 3% or 19 million children. The highest percen tages of children with severe wasting are in south-central Asia (51%) and central Africa (56%). Suboptimum growth, according to anthropometric measures indicative of stunting, wasting, and under weight, has been shown to increase the risk of death from infectious diseases in childhood.159,160 This association has been recently re-examined with the pooled analysis of individual-level data from ten longi tudinal studies 000 child-years of follow-up and involving more than 55 1315 deaths in children younger than 5 years.161 As with previous analyses, all degrees of stunting, wasting, and under weight had higher mortality and the risk increased dix p 11). Undernutrition as Z scores decreased (appen can be deemed the cause of death in a synergistic association with infectious diseases; if the undernutrition did not exist, the deaths would not have occurred.1 All anthropo metric measures of under nutrition were asso ciated with increased hazards of death from diarrhoea, pneumonia, and measles; the association was also noted for other infectious diseases, but not malaria. We calculated the population attributable fractions for stunt ing, under weight, wasting, and its subset of severe wasting using the UN and NIMS prevalence data with standard methods.1 These fractions were multiplied by the corresponding age-specific and cause-specific deaths162 to estimate the number of deaths attributable to each anthropometric measure (table 2).162 For the percentage of total deaths the denominator was 6934 million.131 Details of these estimates for UN subregions and causes of death are in appendix pp 1213. Stunting and underweight have the highest proportions of attributed child deaths, about 14% for both; wasting accounts for 126% (severe wasting 74%) of child deaths. Table 2 and appendix pp 1415 show estimations using the NIMS prevalence data. In these estimates stunting and underweight each account for 17% of child deaths and wasting for 115% (severe wasting 78%). Stunting is a well established risk factor for poor child development with numerous cross-sectional studies showing associations between stunting and motor and cognitive development. Several longitudinal studies show stunting before age 23 years predicts poorer cog nitive and educational outcomes in later childhood and adolescence.135,163 Effect sizes for the longitudinal studies comparing children with HAZ of 2 or lower with nonstunted children (HAZ 1) are moderate to large.163 Length-for-age Z score (LAZ) at age 2 years was con sistently associated with higher cognitive Z scores in children aged 49 years (017019 per unit change LAZ) across four cohorts with moderate (2432%) or high (6786%) stunting prevalence.164 Associations with underweight have also been reported.163 Stunted children show behavioural differences in early childhood including apathy, more negative affect, and reduced activity, play, and exploration.165,166 The first
26

2 years of life are a crucial period linking growth and development; growth from birth to 24 months but not from 24 to 36 months was associated with child develop ment in Guatemala,167 and weight gain in the first 2 years predicted school outcomes in five cohorts.138 Analyses from the COHORTS group that are presented here suggest that growth in the first 2 years of life, but not at later ages, is associated with achieved school grades in adults.168 However, some evidence suggests that growth after 24 months of age might also be associated with lower cognitive ability, but with a smaller effect size than for early growth.169 In a Malawi cohort, height gain from 18 to 60 months predicted mathematics ability at 12 years. Height gain at 1 month and change from 1 to 6 months and 6 to 18 months were not significant predictors.170 We analysed changes in stunting prevalence between 1996 and 2008 in Bangladesh, Brazil, and Nigeria, according to wealth and urban or rural status (panel 2, figure 6).

Overweight and obesity


The prevalence of overweight worldwide and for UN regions is based on the joint analyses done by UNICEF, WHO, and the World Bank.154 In 2011, globally, an estimated 43 million children younger than 5 years, or 7%, were overweight (ie, WHZ greater than two Z scores above the median WHO standard), on the basis of the WHO Child Growth Standards (appendix p 9)a 54% increase from an estimated 28 million in 1990. This trend is expected to continue and reach a prevalence of 99% in 2025 or 64 million children (figure 7). Increasing trends in child overweight are taking place in most world regions, not only in high-income countries, where preva lence is the highest (15% in 2011). However, most overweight children younger than 5 years (32 million in 2011) live in LMICs. In Africa, the estimated prevalence increased from 4% in 1990 to 7% in 2011, and is expected to reach 11% in 2025 (figure 7). Prevalence of over weight is lower in Asia (5% in 2011), but the number of affected children is higher compared with Africa (17 and 12 million, respectively). Differences in childhood overweight prevalence between the richest and poorest quintiles are small in most countries (figure 5), and in general prevalence tends to be higher in the richest quintile than in the poorest. In 78 countries with data, prevalence in the richest quintile was on average 131 (range 055360) times higher than in the poorest quintile. Overweight is much the same between the sexes (appendix p 24) and slightly more prevalent in urban than in rural areas dix p 25). In 81 countries with data, urban preva (appen lence was 108 times higher on average (range 044146) than rural prevalence. Childhood overweight results in both immediate and longer-term risks to health. Among the immediate risks are metabolic abnormalities including raised cholesterol,
www.thelancet.com

Series

triglycerides, and glucose, type 2 diabetes, and high blood pressure.171 Childhood overweight is also a strong risk factor for adult obesity and its consequences.2,172

Panel 2: How do inequalities in stunting evolve with time? Increased availability of survey data, including several surveys from different years, allowed the analyses of time trends in nutritional indicators according to population subgroups. Figure 6A compares trends in stunting by wealth quintile in three countries. In Nigeria, there was almost no change in stunting prevalence from 2003 to 2008, and the degree of inequality remained almost unchanged. In Bangladesh, stunting prevalence decreased in all subgroups, but inequality also remained at the same magnitude. In Brazil, where prevalence of stunting is much lower, equity improved because of a substantial decrease in stunting the poorest populations. Figure 6B shows the corresponding results for urban and rural differences over time. In Nigeria, rural prevalence was higher than urban prevalence in 2003, and both remained at similar levels by 2008. In Bangladesh, both urban and rural rates decreased, but the gap was reduced over time. In Brazil, where there was a two-times rural-urban gap in 1996, full equality had been reached by 2006. Increased data availability has led to the ability to study trends over time for subnational groups. Such data should be used for advocacy purposes, showing which population groups require closer attention, and also as a means to assess the effect of nutrition-specific and nutrition-sensitive interventions, as well as of broader developmental programmes and initiatives.

Childhood vitamin deficiencies


Anaemia and iron
The percentages of children with anaemia (haemoglobin <110 g/L) and severe anaemia (haemoglobin <70 g/L) due to inadequate iron, (ie, anaemia that is correctable by oral iron supplements, calculated as described earlier) are 181% and 15%, respectively. The prevalence is highest in Africa and Asia for all IDA and in Africa for severe IDA (table 1). However, the proportion of all childhood anaemia corrected by iron supplementation ranges from 63% in Europe to 34% in Africa where there are other major causes of anaemia; the proportion of severe anaemia corrected by iron supplementation in Africa is 57% (appendix p 4). Iron supplementation in children aged 5 years and older with IDA generally benefits their cognition, but studies of children younger than 3 years have had mostly negative findings.135,173175 Most cohort and crosssectional studies of children younger than 3 years with IDA find developmental deficits and studies from the past 15 years provide evidence of neurophysiological changes suggestive of delayed brain maturation.135 However, IDA is associated with many social dis advantages that also affect child development and, thus, randomised controlled trials are necessary to establish a causal association. A previous systematic review showed that iron supplementation resulted in a small improvement in mental development scores in children with IDA aged older than 7 years, but had no effect in children younger than 27 months.173 To further assess this scientific literature, we identified seven doubleblind randomised controlled trials176182 of iron lasting at least 8 weeks in children younger than 4 years. Five trials177180,181,182 showed benefits in motor develop ment and two did not.176,180 Only one showed benefits in language,178 and a small study showed benefits in mental development.177 In an eighth randomised controlled trial,183 children given iron in infancy showed no cognitive benefit when followed up at age 9 years. Four additional ran domised controlled trials184187 examined the combined effects of iron and folate supplementation in children younger than 36 months of age. One showed benefits to motor milestones,184 others showed no benefits to motor185,186 or language186 milestones, and one showed no benefit to cognitive function.187 Thus, there is some evidence that iron deficiency affects motor development in children younger than 4 years, but no consistent evidence for an effect on mental development. However, many of the supplementation trials produced only small differences in iron status between the treated and control groups, possibly limiting their ability to affect development. It is also possible that mental develop ment takes longer to
www.thelancet.com

A
100 90 80 Stunting prevalence (%) 70 60 50 40 30 20 10 0 Wealth quintiles Q1: poorest 20% Q2 Q3 Q4 Q5: richest 20%

B
100 90 80 Stunting prevalence (%) 70 60 50 40 30 20 10 0 Area of residence Urban Rural

96 19 99 20 04 20 07

20 06

03 20

19 9

19

Bangladesh

Brazil

Nigeria

Figure 6: Changes in stunting over time, in Bangladesh, Brazil, and Nigeria (A) Stunting prevalence by wealth quintile. The longer the line between two groups, the greater the inequality. (B) Stunting prevalence by urban or rural location.

20

08

27

Series

65 60 55 50 45 Numbers (millions) 40 35 30 25 20 15 10 5 0

Global Numbers

Prevalence 552 478

637

Africa 14 12 10 Numbers (millions)


Prevalence (%)

65 60 55 50 45 40 35 30 25 20 15 10 5 0 Latin America and the Caribbean 69 87 110 138 173 214

14 12 10
Prevalence (%)

412 353 284 296 317

8 6 4 2 0

8 6 4 2 0

46

56

Asia 65 60 55 50 45 Numbers (millions) 40 35 30 25 20 15 10 5 0 1990 1995 2000 2005 2010 Year 2015 2020 2025 0 2 144 141 143 147 165 184 205 231 10 Numbers (millions)
Prevalence (%)

14 12

65 60 55 50 45 40 35 30 25 20 15 10 5 0 36 1990 37 1995 38 2000 39 38 38 2015 38 2020 38 2025

14 12 10
Prevalence (%)

8 6 4

8 6 4 2 0

2005 2010 Year

Figure 7: Trends in prevalence and numbers of overweight (WHZ >2) children, by selected UN regions and globally, 19902010, and projected to 2025, on the basis of UN prevalence estimates WHZ=weight-for-height Z score.

improve than the duration of the trials or that the effects of iron deficiency early in life are irreversible.

Vitamin A
Clinical assessment of ocular symptoms and signs of xerophthalmia and biochemical assessment of serum concentration of retinol are the two common methods in population surveys for estimation of prevalence of vitamin A deficiency. WHO provides prevalence esti mates of vitamin A deficiency in preschool children (<5 years) for 19952005 from 99 countries.45 Globally, 09% (95% CI 0315) or 517 million preschool age children are estimated to have night blindness and 333% (311354) or 90 million to have subclinical vitamin A deficiency, defined as serum retinol concen tration of less than 070 mol/L. Vita min A deficiency using night blindness prevalence can be defined as a global problem of mild public health importance,45 although the prevalence in Africa (2%) is higher than elsewhere. Although prevalence of clinical symptoms
28

has declined, probably because of large-scale vitamin A supplementation programmes in many countries, sub clinical vitamin A deficiency affects high proportions of children in Africa and southeast Asia (table 2). Many randomised controlled trials have been done to examine the effect of supplementation every 46 months and fortification on survival of children aged 6 months and older; these studies provide the best evidence for deaths attributable to vitamin A deficiency.188191 Meta-analyses of these trials show a mortality reduction of 23%,188 30%,189 and 24%190 in children aged 659 months. With publication of a large programme effectiveness study from India, a revised meta-analysis shows a mortality effect of 11%, still a statistically signifi cant benefit.192 In our calcu lations we use only the effects in the trials on particular causes of death, not the effects on overall deaths because the causes of death at the time of the trials and nowadays are probably different (eg, diarrhoea and measles account for a much smaller proportion of child deaths now than
www.thelancet.com

Series

1020 years ago).162 To derive the risk of vitamin A deficiency the inverse (1/risk reduction) of the causespecific mortality reduction identified in the trials was deemed to be the risk of deficiency and adjusted with the assumption that all the effect was in the subset of the trial population with low serum retinol (appendix p 16). Much the same adjust ment was done for the effect of vitamin A deficiency on diarrhoea incidence dix p 17). This allows the adjusted RR to be applied (appen to the present prevalence of low serum retinol to estimate the attributable deaths or disease episodes. Although results of trials193195 in south Asia show a newborn supplementation effect on mortality in the first 6 months of life, the evidence from Africa is less clear, and further studies are underway. Therefore, we do not estimate deaths attributable to vitamin A deficiency in the first 6 months of life. Child deaths attributable to vitamin A deficiency for 2011 are estimated to be 157000 (table 2).

intake in countries to estimate the attributable deaths. The population attributable fractions for diarrhoea and pneumonia were multiplied by the number of these deaths in 2011.162 The number of child deaths attributed to zinc deficiency in 2011 is 116000 (table 2). Zinc deficiency also has a small negative effect on growth. A meta-analysis of randomised controlled trials of zinc supplementation showed a significant benefit for linear growth in children aged 05 years.201 The effect was a gain of 037 cm in zinc-supplemented children. Trials that used a dose of zinc of 10 mg per day for 24 weeks, rather than lower doses, showed a larger benefit of 046 cm.

Breastfeeding practices
Present recommendations are that babies should be put on the breast within 1 h after birth, be exclusively breastfed for the first 6 months, and for an additional 18 months or longer, be breastfed along with comple mentary foods. There are no recently published system atic com pilations of data for breastfeeding patterns so we analysed data from 78 countries with surveys done in LMICs during 200010 (appen dix p 20). Early initiation of breastfeeding (within 1 h) is highest in Latin America (mean 58%, 95% CI 5067), intermediate in Africa (50%, 4555) and Asia (50%, 4258), and lowest in eastern Europe (36%, 2350). Except for eastern Europe, where the lowest rates of breastfeeding are recorded, globally about half of children younger than 1 month, and three in every ten children aged 15 months are exclusively breastfed. Breastfeeding in 623 month olds is most frequent in Africa (mean 77%, 95% CI 7381) followed by Asia (62%, 5471) and Latin America (60%, 5069), with lower occurrence in eastern Europe (33%, 2442). The risks of increased mortality and morbidity due to deviation from present breastfeeding recommendations are well documented.1 Updated systematic reviews of these risks have results that are much the same as our previous estimates (appendix p 21).1,202,203 The number of child deaths attributed to suboptimum breastfeeding in 2011 is 804000 or 116% of all deaths (table 2). Three prospective case-cohort studies provide data for the association of early breastfeeding initiation (within 24 h) with neonatal mortality.204 Although early initiation was associated with lower neonatal mortality (RR 056, 95% CI 046079), in babies who were exclusively breastfed the mortality risk was not significantly reduced (069, 027175). The possible benefit of early initiation of breastfeeding was therefore not deemed to be additive to the effects of exclusive breastfeeding in our analyses. A systematic review shows that breastfeeding is con sistently associated with an increase in IQ of about three points,205 even after adjustment for several con founding factors including maternal IQ. Evidence for the pro tection afforded by breastfeeding against risk factors for non-communicable diseases is less consistent. A series
29

Zinc
Zinc deficiency in populations could be assessed by a shift of the population distribution of serum zinc con cen trations to lower values as recommended by the Inter national Zinc Nutrition Consultative Group;196 however, insufficient data exist to classify countries or sub national populations. Instead the proportion of the national population estimated to have an inadequate zinc intake on the basis of national food availability and dietary requirements is used.70 According to this method, an estimated 17% of the worlds population has an inadequate zinc intake; substantial regional variation exists, with Asia and Africa having the highest preva lences (table 1). A systematic review197 showed that zinc supplementation resulted in a 9% reduction (RR 091, 95% CI 082101) of borderline significance in all-cause child mortality. A separate analysis of available trials showed a significant 18% reduction (RR 082, 070096) in all-cause mortality in children aged 14 years.198 There were suggestive benefits on diarrhoea-specific (RR 082, 95% CI 064105) and pneumonia-specific (085, 065111) mortality.197 This and previous analyses have included cause-specific mortality effects even when they were not statistically significant when the effect on all-cause mortality was statistically significant.199,200 These trials were not powered for cause-specific mortality effects. Supporting evidence for the cause-specific mortality effects comes from randomised controlled trials that showed significant reductions in diarrhoea incidence (RR 087, 95% CI 081094) and pneumonia incidence (081, 073090).197 To derive the risk of zinc deficiency, we adjusted the inverse of the cause-specific mortality (appendix p 18) and incidence (appen dix p 19) reductions noted in the trials with the assumption that all the effect was in the subset of the trial population at risk of zinc deficiency, as estimated from the availability of food in national diets.70 This method allows the adjusted RRs to be applied to the present prevalence of inadequate zinc
www.thelancet.com

Series

of meta-analyses,205 based mainly on studies in adults from high-income settings, showed no evidence of protection against total cholesterol levels or diastolic blood pressure. When the meta-analysis was restricted to high-quality studies, breastfeeding was associated on average with a 1 mm reduction in systolic blood pressure, and with a 12% reduction in the risk of overweight or obesity. Studies of diabetes or glucose levels were too few to allow a firm conclusion.

attributed jointly to fetal growth restriction, suboptimum breastfeeding, stunting, wasting, and deficiencies of vitamin A and zinc. The overall results were the same using UN or NIMS prevalence estimates. As part of this total, the joint distribution of suboptimum breastfeeding and fetal growth restriction in the neonatal period contributes 13 million deaths or 19% of all deaths of children younger than 5 years.

Joint effects of nutritional conditions on child mortality


To estimate the population attributable fraction and the number of deaths attributable to several risk factors, we used Comparative Risk Assessment methods.206 We calculated attributable deaths from four specific causes of mortality (diarrhoea; measles; pneumonia; and other infections, excluding malaria) associated with different nutritional status measures. In the neonatal period all deaths were deemed to be associated only with sub optimum breastfeeding and fetal growth restriction. The Comparative Risk Assessment methods allow the estimation of the reduction in death that would take place if the risk factors were reduced to a minimum theoretical level or counterfactual level of exposure.206 For the assessment the following formula was used:

Effects of fetal and early childhood undernutrition on adult health


In The Lancets 2008 Series on maternal and child undernutrition,2 consequences of early childhood nutrition on adult health and body composition were assessed by reviewing the scientific literature and doing meta-analyses of five birth cohorts from LMICs (India, the Philippines, South Africa, Guatemala, Brazil), an effort that gave rise to the COHORTS collaboration.208 On the one hand, the conclusions were that small size at birth and at 2 years of age (particularly height) were associated with reduced human capital: shorter adult height, less schooling, reduced economic productivity, and for women, lower offspring birthweight. On the other hand, larger child size at 24 months of age was a risk factor for high glucose concentrations, blood pressure, and harmful lipid levels once adjustment for adult BMI was made, suggesting that rapid weight gain, especially after infancy, is linked to these conditions. The COHORTS group later did pooled analyses that use conditional growth variables that remove the correlation between growth measures across ages, allowing inferences to be made about the relative asso ciation between growth during specific age intervals and outcomes.149 Also, because gains in height and weight are correlated, these analyses use conditionals that separate linear growth from weight gain. The conclusions were that heavier birthweight and faster linear growth from 0 to 2 years lead to large gains in human capital, but have little association with adult cardiovascular risk factors. Also, faster weight gain independent of linear growth has little benefit for human capital. After the age of 2 years, and particularly after the age of 4 years, rapid weight gains show adverse effects on adult cardiovascular risk. The COHORTS analyses control for confounding and the findings are broadly much the same across the five cohorts. Additional longitudinal studies report effects on later mental health with higher levels of depression and anxiety and lower self-esteem in adolescents who were stunted by age 2 years compared with non-stunted;209 increased depression in adolescents who had severe acute malnutrition in infancy;210 increased risk of suicidal ideation associated with lower HAZ at 24 months,211 and higher levels of hyperactivity in late adolescence and attention deficit in adults.209,212 Although observational, the findings are persuasive and consistent with the little quasi-experimental evidence available. Follow-up studies of a community randomised nutrition trial in Guatemala
www.thelancet.com

PAF =

P ( RR
i i =1

1)

P (RR
i i =1

1) + 1

For the purpose of this analysis, RRi equals the RR of mortality for the ith exposure category, associated with specific UN regions. Pi equals the proportion of children in the ith exposure in theoretical or counterfactual category. P equals the population weighted mean of countries prevalence estimates for different levels of fetal growth restriction, stunting, wasting, deficiencies of vitamin A and zinc, and suboptimum breastfeeding practices in their relevant age groups. Because specific causes of deaths could potentially be caused by more than one factor, the population attributable fractions for multiple risk factors that affect the same disease outcome overlap and cannot be combined by simple addition. For this reason, the joint population attributable fraction was estimated with the formula: Joint PAFi=1product (1PAFi), where PAFi equals the population attributable fraction of the different risk factors.207 All analyses were done for relevant age groups of that risk factor and then the results from the age groups were aggregated. For stunting and wasting we did calculations separately with UN and NIMS prevalence estimates. The resulting 31 million deaths constitute 45% of global deaths in children younger than 5 years in 2011,
30

Series

have shown long-term effects of exposure to improved nutrition during the first 23 years, but not after 3 years, on education,213 and wages.214 Effects of improved nutri tion in the first several years of life on risk factors for chronic diseases were minor but in some cases bene ficial.215 Exposure to improved nutrition during child hood affected the growth of the next generation in girls and their future children.2 Famines are another source of information about longterm effects of poor nutrition in early life. The Dutch famine of 194445 (brief, intense, but affecting a pre viously well-nourished population) suggests effects of prenatal exposure on schizophrenia, no effects on human capital (height, cognitive function), and weak and in consistent effects on cardiovascular risk factors.216 The 195961 Chinese famine (prolonged, severe, and affecting an already malnourished population) suggests effects of exposure during pregnancy and the first 2 years of life on height, wealth, income, mental health, and inter generational effects on birthweight.217220 Surprisingly, some findings suggest protective effects of famine expo sure, which might be explained by high mortality and intense selection of the hardiest.219221

Discussion
Nutrition has profound effects on health throughout the human life course and is inextricably linked with cog nitive and social development, especially in early child hood. In settings with insufficient material and social resources, children are not able to achieve their full growth and developmental potential. Consequences range broadly from raised rates of death from infectious diseases and decreased learning capacity in childhood to increased non-communicable diseases in adulthood. Nutrition and growth in adolescence is important for a girls health and adult stature. Women with short stature are at risk of complications in delivery, such as obstructed labour. Nutritional status at the time of conception and during pregnancy is crucial for fetal growth. Babies with fetal growth restriction, as shown by being SGA, are at increased risk of death throughout infancy. We estimate that 32 million babies are born SGA, 27% of births in LMICs, and about 800 000 neonatal deaths and 400 000 post-neonatal infant deaths can be attributed to the increased risk associated with having fetal growth restriction. The use of more appropriate methods than those in our previous review1 to understand the preva lence and risk of SGA presented here resulted in estimates that are more than double our previous estimate of attributable neonatal deaths related to term low birthweight. This new finding contradicts the widespread assumption that SGA infants, by contrast with preterm babies, are not at a substantially increased risk of mortality. Additionally, babies who are SGA have an increased risk of growth faltering in the first 2 years of life; our estimates suggest that 20% of stunting might be attributable to fetal growth restriction. The links between
www.thelancet.com

fetal growth restriction and maternal nutritional con ditions, short stature, and low BMI, should lead to more emphasis on nutritional interventions before and during pregnancy. These interventions would have benefits for health of adolescents and women, could reduce compli cations of pregnancy and delivery for the mother, and enhance fetal growth and development. Stunted linear growth has become the main indicator of childhood undernutrition, because it is highly prevalent in all developing regions of the world, and has important consequences for health and development. It should replace underweight as the main anthropometric indicator for children. Underweight indices include chil dren who are short, but who can have an increased WHZ, being therefore at increased risk of long-term adverse health outcomes. Linear growth assessment in primary care is an essential component of country efforts to reduce childhood stunting. More experience is needed in the operational aspects of the assessment and inter pretation of linear growth by health workers and in the effective intervention responses. Prevalence of stunting in children younger than 5 years in developing countries in 2011 was about 28%, a decrease from 40% or more in 1990 and the 32% estimate in our 2008 nutrition Series for 2004.1 The number of stunted children globally has decreased from 253 million in 1990 to 165 million in 2011. Another reported estimate was 167 million in developing countries for 2010.157 The 13-year Comprehensive Implementation Plan (201225) on Maternal, Infant and Young Child Nutrition, endorsed at the 2012 World Health Assembly, includes six global nutrition targets, the first of which calls for a 40% reduction of the global number of children younger than 5 years who are stunted by 2025 (compared with the baseline of 2010).222 This goal would translate into a 39% relative reduction per year and imply reducing the number of stunted children from the 171 million in 2010 to about 100 million. However, at the present rate of decline, prevalence of stunting is expected to reach 20%, or 127 million, in 2025. In Africa, only small reductions in prevalence are anticipated on the basis of present trends. However, in view of the rising number of births, the actual number of stunted children will increase from 56 to 61 million. By contrast, Asia is projected to show a substantial decrease in stunting prevalence and in the absolute number of children affected. Stunted, underweight, and wasted children have an increased risk of death from diarrhoea, pneumonia, measles, and other infectious diseases. Recalculated risks of cause-specific mortality161 confirmed previous estimates1,159,160 and were used in estimation of the deaths attributable to these conditions. The attributable fraction of deaths for these nutritional status measures has remained nearly the same as in previous calculations, but the number of attributed deaths has declined because of the decrease in the prevalence of these measures and a decline in the affected causes of death. By our estimates
31

Series

more than 1 million deaths can be attributed to stunting 000 to wasting, about 60% of which are and about 800 attributable to severe wasting. These attributable deaths cannot be added because of the overlap of these and other nutritional conditions, but are instead included in the calculation of the deaths attributed to the joint effects of all nutritional conditions. Our estimate of about 1 million child deaths due to underweight is higher than the recently published 860000 deaths in the Global Burden of Disease (GBD).200 The GBD Study did not estimate deaths attributed to stunting or wasting. Deficiencies of vitamins and minerals have important health consequences, both through their direct effects, such as iron deficiency anaemia, xerophthalmia due to vitamin A deficiency, and iodine deficiency disorders, and because they increase the risk of serious infectious diseases. In the latter category, vitamin A and zinc deficiencies have been shown to have the greatest effects among the micronutrients. Our estimate of the nearly 157000 child deaths attributed to vitamin A deficiency is smaller than our previous estimate (668 000).1 Notably, we did not include the risk of mortality in the first 6 months of life, as we did last time for Asia, because of more uncertainty about this risk; several trials of neonatal vitamin A supplementation are underway in Asia and Africa to assess the benefits. Likewise, our estimate of 116000 child deaths attributed to zinc deficiency is much smaller than the 450000 in our previous estimates, partly because we now estimate the risk of pneumonia and diarrheal mortality beginning at 12 rather than 6 months of age. In both cases there have also been changes in our methods of estimating the prevalence of the conditions and the risk associations, and combining these factors to get attributed fractions of deaths. These methods are much the same as those reported by the GBD Study.200 Our estimates are 20% higher for zinc and 30% higher for vitamin A, than the GBD estimates. The reduction from our estimates for the year 2004 is partly explained by decreases in the prevalences of these deficienciesfor vitamin A mostly because of large-scale implementation of high-dose supplementation programmes, and a reduction in deaths from diarrhoea and measles affected by vitamin A deficiency and in deaths from diarrhoea and pneumonia affected by zinc deficiency. Some of this mortality reduction (eg, for diarrhoea) could be attributable to the vitamin A supplementation, but might be also related to improvement in nutritional status measures (eg, stunt ing), water and sanitation, and illness treatment, and for measles is largely attributable to improved coverage with measles vaccine. Although the number of deaths attributed to the present prevalence of vitamin A deficiency is relatively small, importantly, a reduction in present coverage of vitamin A interventions would probably result in an increase in mortality, because in most LMICs dietary intake is still inadequate. Breastfeeding exclusively for the first 6 months and then along with complementary food to 24 months of age
32

is highly beneficial. Data from developing regions show that present practices are far from optimum, despite improvements in some countries. Our previous estimates suggested that 14 million deaths were attributed to suboptimum breastfeeding, especially related to nonexclusive breastfeeding in the first 6 months of life. Our new estimates are 804000 deaths, a substantial reduction since our estimate for the year 2004. Although present estimates of the risks associated with suboptimum breast feeding practices are almost unchanged from what was used previously, the reduction in infectious disease deaths in the past 10 years results in a lower number of preventable deaths.173 Our number is higher than the 000 deaths attributed to suboptimum estimated 545 breastfeeding in the GBD Study.200 Insufficient methods are provided by GBD to understand the reasons for differences, but one reason is probably the lower estimates of affected diarrhoea and pneumonia deaths than the UN estimates we used.223 To work out the total deaths attributed to nutritional conditions, we calculated the joint distribution of stunting, wasting, fetal growth restriction, deficiencies of vitamin A and zinc, and suboptimum breastfeeding. The resulting 31 million deaths constitute 45% of the 69 million global child deaths in 2011. The total number of attributed deaths is reduced from the 35 million we estimated for 2004, despite the population attributable fraction increasing from 35% to 45%. In this period the mortality in children younger than 5 years decreased from 10 million to 69 million deaths and there were even larger decreases in the causes of deathsuch as measles, diarrhoea, and pneumonia162that are most affected by nutritional conditions. The increase in the percentage of attributed deaths compared with the previous series is largely because of the more appropriate inclusion of the effects of SGA (about 800 000 deaths) instead of term low birthweight (337000 deaths) and the inclusion of all wasting (WHZ <2) instead of severe wasting (WHZ <3) in the joint calculations. The number of children affected by excessive body weight relative to length or height is increasing globally. Although the prevalence of overweight in high-income countries is more than double that in LMICs, three quarters of the global total live in LMICs. The recorded trends in the prevalence of early childhood overweight are probably a consequence of changes in dietary and physical activity patterns over time. These behavioural changes are affected by many social and environmental factors, including interpersonal (family, peers, and social net works), com munity (school, work place, and institu tions), and govern mental (local, state, and national policies).224 Studies show that the trend toward childhood obesity can start as early as age 6 months.225,226 In LMICs, rapid weight gain after 2 years of age is particularly associated with adult fat mass.2,168 Intrauterine, infant, and preschool periods are deemed to be possible crucial periods for programming long-term regulation of energy balance.227229
www.thelancet.com

Series

If trends are not reversed, increasing rates of childhood cations not overweight and obesity will have vast impli only for future health-care expenditures but also for the ings confirm overall development of nations. These find the need for effective interventions and programmes to reverse anticipated trends. The early recognition of exces sive weight gain relative to linear growth is essential. Routine assessment of both weight and length in all children needs to become standard clinical practice from very early childhood. We have reported previously and confirmed in a new analysis that anaemia that is reduced by supplemental iron in pregnancy is a risk factor for more than a quarter of maternal deaths. Anaemia is probably a particularly important risk factor for haemorrhage, the leading cause of maternal deaths (23% of total deaths).223 Additionally, there is now sound evidence that calcium deficiency increases the risk of pre-eclampsia, now the second most important cause of maternal death (19% of total deaths).223 Thus, addressing deficiencies of these two minerals could result in substantial reduction of maternal deaths. Nutrition is crucial for optimum child development throughout the first 1000 days of life and beyond. Maternal nutrition affects fetal growth and brain develop ment with clear evidence for the need to continue and strengthen programmes to prevent maternal iodine deficiency. Fetal growth restriction and postnatal growth affect motor and cognitive development, with the largest effects from stunting before age 23 years. Growth later in childhood might also affect development though with a smaller effect size. Iron deficiency anaemia affects motor development; effects on cognitive development have been more clearly shown in children older than 5 years than in younger children. Promotion of good early nutrition is essential for children to attain their developmental potential; however, poor nutrition occurs with other risks for development, in particular inadequate stimulation during early child hood. Interventions to promote home stimulation and learning opportunities in addition to good nutrition will be needed to ensure good early development and longerterm gains in human capital, with integration and coord ination of programmes and policies across sectors. Panel 3 lists research priorities. Additionally, there is a general need for better data on micronutrient deficiencies and other nutritional conditions at national and sub national levels. This work should involve the development and use of improved biomarkers that could be used to describe nutritional conditions and increase knowledge about how they affect health and development. Such information is needed to guide intervention programmes in countries and priorities for support globally. In the 5 years since our last series there have been some improvements in nutritional conditions, especially for child growth. Nonetheless, the extent of these conditions remains high with serious detrimental
www.thelancet.com

health con sequences, with a high child mortality burden related to both stunting and wasting. New evidence in this paper supports the focus on pregnancy and the first 2 years of life, the crucial 1000 days, which we called for in the previous series, but adds more emphasis to the nutritional conditions in adolescence, at the time of conception, and during pregnancy, as nal health and survival, fetal growth important for mater sequent early childhood survival, growth, and and sub ment. Fetal growth restric tion and poor growth develop early in infancy are now recognised as important determinants of neonatal and infant mortality, stunting, and overweight and obesity in older children and adults. Preventive efforts should continue to focus on the 1000 days, while therapeutic efforts continue to target severe wasting.
Panel 3: Research priorities How much of the effect of the underlying economic and social determinants of optimum growth and development is mediated through known, measurable proximate determinants? What are the consequences of calcium or vitamin D deficiencies for maternal health, fetal growth, birth outcomes, and infant health? What is the importance of zinc deficiency or other micronutrient deficiencies in the risk of preterm delivery? What is the role of nutrition in adolescence, at the time of conception, and in pregnancy in healthy fetal growth and development? What factors are associated with regional variation in fetal growth restriction and do its consequences on growth and mortality vary by setting? What is the interaction of fetal growth restriction and post-partum infections and diet in contributing to stunting? What is the role of vitamin A deficiency in newborn babies or the first 6 months of life in neonatal and infant mortality? Are there benefits from iron supplementation for mental development in children younger than 5 years with iron deficiency anaemia? Do deficiencies of any micronutrients other than vitamin A and zinc increase the risk of mortality from infectious diseases? Can the risk of stunting be explained by specific dietary factors, micronutrient deficiencies, and clinical or subclinical infections? How can the rapid decreases in stunting noted in selected countries be explained by changes in intervention coverage and in distal determinants? Are there benefits of early initiation of breastfeeding that are independent of the practice of exclusive breastfeeding? How can developmental differences associated with nutritional deficiencies at various ages of childhood be best quantified and related to functioning and productivity in adulthood? What are the optimum physical growth rates in the first year of life to avoid both undernutrition and obesity? What are the effects of physical growth rates and stunting on cognitive abilities and psychological functioning? Can health workers effectively and efficiently assess and interpret height-for-age measures of linear growth in health and nutrition programmes? How important is optimum growth and nutritional status in the fetal and early childhood period (first 1000 days of life) in the development of adult obesity and non-communicable diseases?

33

Series

Contributors REB conceptualised and coordinated the analyses, did the first draft of the paper, responded to reviewer comments and incorporated all revisions until publication. CGV contributed with analyses of inequalities, breastfeeding patterns, and long-term consequences of early nutrition, and served as a coordinator for this paper. SPW contributed sections on the effect on child development of maternal IDA, fetal growth restriction, and childhood undernutrition, maternal IDA, and mental health and served as a coordinator for this paper. ZAB assisted with conceptualising the paper and contributed sections on micronutrients and adolescent nutrition. PC contributed to writing the sections on maternal and childhood micronutrient deficiencies (vitamin A and iodine), mortality and morbidity effects of maternal low BMI and short stature, and micronutrient deficiencies and effects of fetal growth restriction on childhood stunting. MdO contributed global and regional analyses on childhood stunting, wasting, underweight, and overweight. ME contributed to exposures for various child and maternal nutritional indicators, to selected aetiological effect sizes for childhood exposures, and provided advice on methods for calculation of attributable burden. SG-MG reviewed the scientific literature on iron and iodine deficiency and contributed to these sections. JK contributed analyses and text on maternal risk factors for, prevalence of, and mortality consequences of fetal growth restriction and the mortality effect of vitamin A supplementation. RM contributed the sections on adolescent nutrition, the consequences of short maternal stature, and long-term consequences of early nutrition. RU contributed to sections on maternal obesity, considering causes and short-term and long-term consequences to mothers, neonates, and infants. All authors contributed to the final paper. Maternal and Child Nutrition Study Group Robert E Black (Johns Hopkins Bloomberg School of Public Health, USA), Harold Alderman (International Food Policy Research Institute, USA), Zulfiqar A Bhutta (Aga Khan University, Pakistan), Stuart Gillespie (International Food Policy Research Institute, USA), Lawrence Haddad (Institute of Development Studies, UK), Susan Horton (University of Waterloo, Canada), Anna Lartey (University of Ghana, Ghana), Venkatesh Mannar (The Micronutrient Initiative, Canada), Marie Ruel (International Food Policy Research Institute, USA), Cesar Victora (Universidade Federal de Pelotas, Brazil), Susan Walker (The University of the West Indies, Jamaica), Patrick Webb (Tufts University, USA) Series Advisory Committee Marc Van Ameringen (Gain Health Organization, Switzerland), Mandana Arabi (New York Academy of Sciences, USA), Shawn Baker (Helen Keller International, USA), Martin Bloem (United Nations World Food Programme, Italy), Francesco Branca (WHO, Switzerland), Leslie Elder (The World Bank, USA), Erin McLean (Canadian International Development Agency, Canada), Carlos Monteiro (University of So Paulo, Brazil), Robert Mwadime (Makerere School of Public Health, Uganda), Ellen Piwoz (Bill & Melinda Gates Foundation, USA), Werner Schultink (UNICEF, USA), Lucy Sullivan (1000 Days, USA), Anna Taylor (Department for International Development, UK), Derek Yach (The Vitality Group, USA). The Advisory Committee provided advice in a meeting with Series Coordinators for each paper at the beginning of the process to prepare the Series and in a meeting to review and critique the draft reports. Other contributors Tanya Malpica-Llanos, Li Liu, and Jamie Perin assisted with analyses of attributable deaths and Rachel White provided administrative support for this paper and the series. Aluisio Barros, Giovanny Frana, and Maria Clara Restrepo contributed to the analyses of inequalities and of breastfeeding patterns. Anne CC Lee and Naoko Kozuki contributed to analyses on fetal growth restriction. Monika Blssner and Elaine Borghi contributed to the analyses of global and regional estimates of stunting, wasting, underweight, and overweight. Gretchen Stevens and Yuan Lu conducted analysis of maternal BMI; Yuan Lu analysed the studies on the effects of vitamin A and zinc deficiencies. Conflicts of interest REB serves on the Boards of the Micronutrient Initiative, Vitamin Angels, the Child Health and Nutrition Research Initiative, and the

Nestle Creating Shared Value Advisory Committee. VM serves on the Nestle Creating Shared Value Advisory Committee. MdO is a staff member of the World Health Organization. MdO alone is responsible for the views expressed in this publication; they do not necessarily represent the decisions or policies of the World Health Organization. The other authors declare that they have no conflicts of interest. As corresponding author Robert E Black states that he had full access to all data and final responsibility for the decision to submit for publication. Acknowledgments Funding for the preparation of the Series was provided to the Johns Hopkins Bloomberg School of Public Health through a grant from the Bill & Melinda Gates Foundation. The sponsor had no role in the analysis and interpretation of the evidence or in writing the paper and the decision to submit for publication. References 1 Black RE, Allen LH, Bhutta ZA, et al, for the Maternal and Child Undernutrition Group. Maternal and child undernutrition: global and regional exposures and health consequences. Lancet 2008; 371: 24360. 2 Victora CG, Adair L, Fall C, et al, for the Maternal and Child Undernutrition Group. Maternal and child undernutrition: consequences for adult health and human capital. Lancet 2008; 371: 34057. 3 Bhutta ZA, Ahmed T, Black RE, et al, for the Maternal and Child Undernutrition Group. What works? Interventions for maternal and child undernutrition and survival. Lancet 2008; 371: 41740. 4 Bryce J, Coitinho D, Darnton-Hill I, et al, for the Maternal and Child Undernutrition Group. Maternal and child undernutrition: effective action at national level. Lancet 2008; 371: 51026. 5 Morris SS, Cogill B, Uauy R. Effective international action against undernutrition: why has it proven so difficult and what can be done to accelerate progress? Lancet 2008; 371: 60821. 6 Maternal and Child Nutrition Study Group. Maternal and child nutrition: building momentum for impact. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60988-5. 7 Cappa C, Wardlaw T, Langevin-Falcon C, et al. Progress for children: a report card on adolescents. Lancet 2012; 379: 232325. 8 Prentice AM, Ward KA, Goldberg GR, et al. Critical windows for nutritional interventions against stunting. Am J Clin Nutr 2013; 93: 91118. 9 UNICEF. Progress for children: a report card on maternal mortality. New York: United Nations Childrens Fund, 2008. 10 Kozuki N, Lee AC, Vogel J, et al. The association of parity and maternal age with small-for-gestational age, preterm and neonatal and infant mortality: a meta-analysis. BMC Public Health (in press). 11 Gibbs CM, Wendt A, Peters S, et al. The impact of early age at first childbirth on maternal and infant health. Paediatr Perinat Epidemiol 2012; 26 (suppl 1): 25984. 12 Gigante DP, Rasmussen KM, Victora CG. Pregnancy increases BMI in adolescents of a population-based birth cohort. J Nutr 2005; 135: 7480. 13 Rah JH, Christian P, Shamim AA, et al. Pregnancy and lactation hinder growth and nutritional status of adolescent girls in rural Bangladesh. J Nutr 2008; 138: 150511. 14 Ministy of Health and Family Welfare Government of India. International Institute for Population Sciences, Macro International. National family health survey (NFHS-3), 200506: India: Volume I. Mumbai: IIPS, 2007. http://www.measuredhs.com/pubs/pdf/ FRIND3/FRIND3-Vol1AndVol2.pdf (accessed May 20, 2013). 15 Ministerio de Salud Pblica y Asistencia Social, Instituto Nacional de Estadstica, Centros de Control y Prevencin de Enfermedades., Encuesta Nacional de Salud Materno-infantil 2008 (ENSMI-2008/2009). Guatemala, 2010. 16 Finucane MM, Stevens GA, Cowan MJ, et al, on behalf of the Global Burden of Metabolic Risk Factors of Chronic Diseases Collaborating Group (Blood Glucose). National, regional, and global trends in body-mass index since 1980: systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 91 million participants. Lancet 2011; 377: 55767. 17 Stevens GA, Singh GM, Lu Y, et al. National, regional, and global trends in adult overweight and obesity prevalences. Popul Health Metr 2012; 10: 22.

34

www.thelancet.com

Series

18 Christian P, Katz J, Wu L, et al. Risk factors for pregnancy-related mortality: a prospective study in rural Nepal. Public Health 2008; 122: 16172. 19 WHO. Maternal anthropometry and pregnancy outcomes. A WHO collaborative study. Bull World Health Organ 1995; 73 (suppl): S198. 20 Dujardin B, van Cutsem R, Lambrechts T. The value of maternal height as a risk factor of dystocia: a meta-analysis. Trop Med Int Health 1996; 1: 51021. 21 Mazouni C, Porcu G, Cohen-Solal E, et al. Maternal and anthropomorphic risk factors for shoulder dystocia. Acta Obstet Gynecol Scand 2006; 85: 56770. 22 Sheiner E, Levy A, Katz M, et al. Short staturean independent risk factor for cesarean delivery. Eur J Obstet Gynecol Reprod Biol 2005; 120: 17578. 23 Chan BCP, Lao TTH. The impact of maternal height on intrapartum operative delivery: a reappraisal. J Obstet Gynaecol Res 2009; 35: 30714. 24 Van Lieshout RJ, Taylor VH, Boyle MH. Pre-pregnancy and pregnancy obesity and neurodevelopmental outcomes in offspring: a systematic review. Obes Rev 2011; 12: e54859. 25 Torloni MR, Betran AP, Horta BL, et al. Prepregnancy BMI and the risk of gestational diabetes: a systematic review of the literature with meta-analysis. Obes Rev 2009; 10: 194203. 26 Tsoi E, Shaikh H, Robinson S, et al. Obesity in pregnancy: a major healthcare issue. Postgrad Med J 2010; 86: 61723. 27 Sohlberg S, Stephansson O, Cnattingius S, et al. Maternal body mass index, height, and risks of preeclampsia. Am J Hypertens 2012; 25: 12025. 28 OBrien TE, Ray JG, Chan WS. Maternal body mass index and the risk of preeclampsia: a systematic overview. Epidemiology 2003; 14: 36874. 29 Aviram A, Hod M, Yogev Y. Maternal obesity: implications for pregnancy outcome and long-term risks-a link to maternal nutrition. Int J Gynaecol Obstet 2011; 115 (suppl 1): S610. 30 Norman JE, Reynolds RM. The consequences of obesity and excess weight gain in pregnancy. Proc Nutr Soc 2011; 70: 45056. 31 Denison FC, Roberts KA, Barr SM, et al. Obesity, pregnancy, inflammation, and vascular function. Reproduction 2010; 140: 37385. 32 Chen A, Feresu SA, Fernandez C, et al. Maternal obesity and the risk of infant death in the United States. Epidemiology 2009; 20: 7481. 33 Stothard KJ, Tennant PW, Bell R, et al. Maternal overweight and obesity and the risk of congenital anomalies: a systematic review and meta-analysis. JAMA 2009; 301: 63650. 34 Koyanagi A, Zhang J, Dagvadorj A, et al. Macrosomia in 23 developing countries: an analysis of a multicountry, facility-based, cross-sectional survey. Lancet 2013; 381: 47683. 35 Gunatilake RP, Perlow JH. Obesity and pregnancy: clinical management of the obese gravida. Am J Obstet Gynecol 2011; 204: 10619. 36 Gilboa SM, Correa A, Botto LD, et al. Association between prepregnancy body mass index and congenital heart defects. Am J Obstet Gynecol 2010; 202: 51, e110. 37 Cresswell JA, Campbell OM, De Silva MJ, et al. Effect of maternal obesity on neonatal death in sub-Saharan Africa: multivariable analysis of 27 national datasets. Lancet 2012; 380: 132530. 38 Temple R, Murphy H. Type 2 diabetes in pregnancyan increasing problem. Best Pract Res Clin Endocrinol Metwab 2010; 24: 591603. 39 OSullivan JB. Body weight and subsequent diabetes mellitus. JAMA 1982; 248: 94952. 40 McGuire W, Dyson L, Renfrew M. Maternal obesity: consequences for children, challenges for clinicians and carers. Semin Fetal Neonatal Med 2010; 15: 10812. 41 Catalano PM. Obesity and pregnancythe propagation of a viscous cycle? J Clin Endocrinol Metab 2003; 88: 350506. 42 Stevens GA, Finucane MM, De-Regil LM, et al, on behalf of Nutrition Impact Model Study Group (Anaemia). Global, regional, and national trends in haemaglobin concentration and prevalence of total and severe anaemia in children and pregnant and non-pregnant women for 19952011: a systematic analysis of population-representative data. Lancet Glob Health (in press).

43 Haider BA, Olofin I, Wang M, et al. Anaemia, prenatal iron use, and risk of adverse pregnancy outcomes: systematic review and meta-analysis. BMJ (in press). 44 De-Regil LM, Jefferds ME, Sylvetsky AC, et al. Intermittent iron supplementation for improving nutrition and development in children under 12 years of age. Cochrane Database Syst Rev 2011; 12: CD009085. 45 WHO. Global prevalence of vitamin A deficiency in populations at risk 19952005: WHO global database on vitamin A deficiency. Geneva: World Health Organization, 2009. 46 Andersson M, Karumbunathan V, Zimmermann MB. Global iodine status in 2011 and trends over the past decade. J Nutr 2012; 142: 74450. 47 Wessells KR, Brown KH. Estimating the global prevalence of zinc deficiency: results based on zinc availability in national food supplies and the prevalence of stunting. PloS One 2012; 7: e50568. 48 Stoltzfus RJ, Mullany L, Black RE. Iron deficiency anaemia. In: Ezzati M, Lopez AD, Rodgers A, Murray CJ, eds. Comparative quantification of health risks: global and regional burden of disease attributable to selected major risk factors. Geneva: World Health Organization, 2004: 163209. 49 Murray-Kolb LE, Chen L, Chen P, et al. CHERG iron report: maternal mortality, child mortality, perinatal mortality, child cognition, and estimates of prevalence of anemia due to iron deficiency. cherg.org/ publications/iron-report.pdf (accessed March 11, 2013). 50 Rasmussen KM, Stoltzfus RJ. New evidence that iron supplementation during pregnancy improves birth weight: new scientific questions. Am J Clin Nutr 2003; 78: 67374. 51 Allen LH. Biological mechanisms that might underlie irons effects on fetal growth and preterm birth. J Nutr 2001; 131: S58189. 52 Christian P. Micronutrients, birth weight, and survival. Annu Rev Nutr 2010; 30: 83104. 53 Imdad A, Bhutta ZA. Routine iron/folate supplementation during pregnancy: effect on maternal anaemia and birth outcomes. Paediatr Perinat Epidemiol 2012; 26 (S1): 16877. 54 Pena-Rosas JP, De-Regil LM, Dowswell T, et al. Intermittent oral iron supplementation during pregnancy. Cochrane Database Syst Rev 2012; 7: CD009997. 55 Dibley MJ, Titaley CR, dEste C, et al. Iron and folic acid supplements in pregnancy improve child survival in Indonesia. Am J Clin Nutr 2012; 95: 22030. 56 Zeng L, Dibley MJ, Cheng Y, et al. Impact of micronutrient supplementation during pregnancy on birth weight, duration of gestation, and perinatal mortality in rural western China: double blind cluster randomised controlled trial. BMJ 2008; 337: a2001. 57 Christian P, Stewart CP, LeClerq SC, et al. Antenatal and postnatal iron supplementation and childhood mortality in rural Nepal: a prospective follow-up in a randomized, controlled community trial. Am J Epidemiol 2009; 170: 112736. 58 Murray-Kolb LE. Iron status and neuropsychological consequences in women of reproductive age: what do we know and where are we headed? J Nutr 2011; 141: S74755. 59 Beard JL, Hendricks MK, Perez EM, et al. Maternal iron deficiency anemia affects postpartum emotions and cognition. J Nutr 2005; 135: 26772. 60 Perez EM, Hendricks MK, Beard JL, et al. Mother-infant interactions and infant development are altered by maternal iron deficiency anemia. J Nutr 2005; 135: 85055. 61 Frith AL, Naved RT, Ekstrom EC, et al. Micronutrient supplementation affects maternal-infant feeding interactions and maternal distress in Bangladesh. Am J Clin Nutr 2009; 90: 14148. 62 Christian P, Murray-Kolb LE, Khatry SK, et al. Prenatal micronutrient supplementation and intellectual and motor function in early school-aged children in Nepal. JAMA 2010; 304: 271623. 63 Christian P, West KP Jr, Khatry SK, et al. Vitamin A or -carotene supplementation reduces but does not eliminate maternal night blindness in Nepal. J Nutr 1998; 128: 145863. 64 Tielsch JM, Rahmathullah L, Katz J, et al. Maternal night blindness during pregnancy is associated with low birthweight, morbidity, and poor growth in South India. J Nutr 2008; 138: 78792. 65 Christian P, West KP Jr, Khatry SK, et al. Maternal night blindness increases risk of mortality in the first 6 months of life among infants in Nepal. J Nutr 2001; 131: 151012.

www.thelancet.com

35

Series

66 West KP, Christian P, Labrique AB, et al. Effects of vitamin A or beta carotene supplementation on pregnancy-related mortality and infant mortality in rural Bangladesh: a cluster randomized trial. JAMA 2011; 305: 198695. 67 Edmond K, Hurt L, Fenty J, et al. Effect of vitamin A supplementation in women of reproductive age on cause-specific early and late infant mortality in rural Ghana: ObaapaVitA double-blind, cluster-randomised, placebo-controlled trial. BMJ Open 2012; 2: e000658. 68 Christian P, Klemm R, Shamim AA, et al. Effects of vitamin A and beta-carotene supplementation on birth size and length of gestation in rural Bangladesh: a cluster-randomized trial. Am J Clin Nutr 2013; 97: 18894. 69 Katz J, West KP Jr, Khatry SK, et al. Maternal low-dose vitamin A or beta-carotene supplementation has no effect on fetal loss and early infant mortality: a randomized cluster trial in Nepal. Am J Clin Nutr 2000; 71: 157076. 70 King JC. Determinants of maternal zinc status during pregnancy. Am J Clin Nutr 2000; 71: S133443. 71 Simmer K, Thompson RP. Maternal zinc and intrauterine growth retardation. Clin Sci 1985; 68: 39599. 72 Mori R, Ota E, Middleton P, et al. Zinc supplementation for improving pregnancy and infant outcome. Cochrane Database Syst Rev 2012; 7: CD000230. 73 Wong EM, Sullivan KM, Perrine CG, et al. Comparison of median urinary iodine concentration as an indicator of iodine status among pregnant women, school-age children, and nonpregnant women. Food Nutr Bull 2011; 32: 20612. 74 Zimmermann MB, Andersson M. Assessment of iodine nutrition in populations: past, present, and future. Nutr Rev 2012; 70: 55370. 75 Pharoah P, Buttfield I, Hetzel B. Neurological damage to the fetus resulting from severe iodine deficiency during pregnancy. Lancet 1971; 297: 30810. 76 Bleichrodt N, Born MP. A meta-analysis of research on iodine and its relationship to cognitive development. In: Stanbury JB, ed. The damaged brain of iodine deficiency. New York, NY: Cognizant Communication, 1994: 195200. 77 Qian M, Wang D, Watkins WE, et al. The effects of iodine on intelligence in children: a meta-analysis of studies conducted in China. Asia Pacific J Clin Nutr 2005; 14: 3242. 78 Zimmermann MB. The effects of iodine deficiency in pregnancy and infancy. Paediatr Perinat Epidemiol 2012; 26 (S1): 10817. 79 Skeaff SA. Iodine deficiency in pregnancy: the effect on neurodevelopment in the child. Nutrients 2011; 3: 26573. 80 WHO, UNICEF, International Council for the Control of Iodine Deficiency Disorders. Assessment of iodine deficiency disorders and monitoring their elimination: a guide for programme managers, 3rd edn. Geneva: World Health Organization, 2007. 81 McLean E, de Benoist B, Allen LH. Review of the magnitude of folate and vitamin B12 deficiencies worldwide. Food Nutr Bull 2008; 29 (suppl 2): S3851. 82 De-Regil LM, Fernandez-Gaxiola AC, Dowswell T, et al. Effects and safety of periconceptional folate supplementation for preventing birth defects. Cochrane Database Syst Rev 2010; 10: CD007950. 83 Blencowe H, Cousens S, Modell B, et al. Folic acid to reduce neonatal mortality from neural tube disorders. Int J Epidemiol 2010; 39 (suppl 1): i11021. 84 Prentice A. Maternal calcium requirements during pregnancy and lactation. Am J Clin Nutr 1994; 59 (suppl 2): S47782. 85 Villar J, Belizan JM, Fischer PJ. Epidemiologic observations on the relationship between calcium intake and eclampsia. Int J Gynecol Obstet 1983; 21: 27178. 86 Ortega RM, Martinez RM, Lopez-Sobaler AM, et al. Influence of calcium intake on gestational hypertension. Ann Nutr Metabol 1999; 43: 3746. 87 Srinivas SK, Edlow AG, Neff PM, et al. Rethinking IUGR in preeclampsia: dependent or independent of maternal hypertension? J Perinatol 2009; 29: 68084. 88 Duley L. The global impact of pre-eclampsia and eclampsia. Semin Perinatol 2009; 33: 13037. 89 Hofmeyr GJ, Lawrie TA, Atallah A, et al. Calcium supplementation during pregnancy for preventing hypertensive disorders and related problems. Cochrane Database Syst Rev 2010; 8: CD001059.

90 Bucher HC, Cook RJ, Guyatt GH, et al. Effects of dietary calcium supplementation on blood pressure. A meta-analysis of randomized controlled trials. JAMA 1996; 275: 101622. 91 Imdad A, Bhutta ZA. Effects of calcium supplementation during pregnancy on maternal, fetal and birth outcomes. Paediatr Perinat Epidemiol 2012; 26 (suppl 1): 13852. 92 WHO. World Health Organization recommendations for prevention and treatment of pre-eclampsia and eclampsia. Geneva: World Health Organization, 2011. 93 Bhutta ZA, Das JK, Rizvi A, et al, The Lancet Nutrition Interventions Review Group and the Maternal and Child Nutrition Study Group. Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60996-4. 94 Ross AC, Taylor CL, Yaktine AL, et al. Dietary reference intakes for calcium and vitamin D. Washington, DC; National Academies Press, 2010. 95 Ross AC, Manson JE, Abrams SA, et al. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. J Clin Endocrinol Metab 2011; 96: 5358. 96 Alfaham M, Woodhead S, Pask G, et al. Vitamin D deficiency: a concern in pregnant Asian women. Br J Nutr 1995; 73: 88187. 97 Bodnar LM, Catov JM, Simhan HN, et al. Maternal vitamin D deficiency increases the risk of preeclampsia. J Clin Endocrinol Metabol 2007; 92: 351722. 98 Davis L, Chang SC, Mancini J, et al. Vitamin D insufficiency is prevalent among pregnant African American adolescents. J Pediatr Adoles Gynecol 2010; 23: 4552. 99 Hypponen E, Boucher BJ. Avoidance of vitamin D deficiency in pregnancy in the United Kingdom: the case for a unified approach in national policy. Br J Nutr 2010; 104: 30914. 100 Kazemi A, Sharifi F, Jafari N, et al. High prevalence of vitamin D deficiency among pregnant women and their newborns in an Iranian population. J Women Health 2009; 18: 83539. 101 Madar AA, Stene LC, Meyer HE. Vitamin D status among immigrant mothers from Pakistan, Turkey and Somalia and their infants attending child health clinics in Norway. Br J Nutr 2009; 101: 105258. 102 Sachan A, Gupta R, Das V, et al. High prevalence of vitamin D deficiency among pregnant women and their newborns in northern India. Am J Clin Nutr 2005; 81: 106064. 103 van der Meer IM, Karamali NS, Boeke AJP, et al. High prevalence of vitamin D deficiency in pregnant non-Western women in The Hague, Netherlands. Am J Clin Nutr 2006; 84: 35053. 104 Brooke OG, Brown IR, Bone CD, et al. Vitamin D supplements in pregnant Asian women: effects on calcium status and fetal growth. BMJ 1980; 280: 75154. 105 Dawodu A, Wagner CL. Mother-child vitamin D deficiency: an international perspective. Arch Dis Child 2007; 92: 73740. 106 Javaid MK, Crozier SR, Harvey NC, et al, and the Princess Anne Hospital Study Group. Maternal vitamin D status during pregnancy and childhood bone mass at age 9 years: a longitudinal study. Lancet 2006; 367: 3643. 107 Mackay-Sim A, Feron F, Eyles D, et al. Schizophrenia, vitamin D, and brain development. Int Rev Neurobiol 2004; 59: 35180. 108 Shenoy S, Swift P, Cody D, et al. Maternal vitamin D deficiency, refractory neonatal hypocalcaemia, and nutritional rickets. Arch Dis Child 2005; 90: 43738. 109 Dror DK. Vitamin D status during pregnancy: maternal, fetal, and postnatal outcomes. Curr Opin Obstet Gynecol 2011; 23: 42226. 110 MacKay AP, Berg CJ, Atrash HK. Pregnancy-related mortality from preeclampsia and eclampsia. Obstet Gynecol 2001; 97: 53338. 111 De-Regil LM, Palacios C, Ansary A, et al. Vitamin D supplementation for women during pregnancy. Cochrane Database Syst Rev 2012; 2: CD008873. 112 Victora CG, Hallal PC, Araujo CL, et al. Cohort profile: the 1993 Pelotas (Brazil) birth cohort study. Int J Epidemiol 2008; 37: 70409. 113 Victora CG, Barros FC. Cohort profile: the 1982 Pelotas (Brazil) birth cohort study. Int J Epidemiol 2006; 35: 23742. 114 Shah A, Faundes A, Machoki M, et al. Methodological considerations in implementing the WHO Global Survey for Monitoring Maternal and Perinatal Health. Bull World Health Organ 2008; 86: 12631.

36

www.thelancet.com

Series

115 Santos IS, Barros AJ, Matijasevich A, et al. Cohort profile: the 2004 Pelotas (Brazil) birth cohort study. Int J Epidemiol 2011; 40: 146168. 116 Roberfroid D, Huybregts L, Lanou H, et al. Effects of maternal multiple micronutrient supplementation on fetal growth: a double-blind randomized controlled trial in rural Burkina Faso. Am J Clin Nutr 2008; 88: 133040. 117 Osrin D, Vaidya A, Shrestha Y, et al. Effects of antenatal multiple micronutrient supplementation on birthweight and gestational duration in Nepal: double-blind, randomised controlled trial. Lancet 2005; 365: 95562. 118 Isaranurug S, Mo-suwan L, Choprapawon C. A population-based cohort study of effect of maternal risk factors on low birthweight in Thailand. J Med Assoc Thai 2007; 90: 255964. 119 Fawzi WW, Msamanga GI, Urassa W, et al. Vitamins and perinatal outcomes among HIV-negative women in Tanzania. N Engl J Med 2007; 356: 142331. 120 Christian P, West KP, Khatry SK, et al. Effects of maternal micronutrient supplementation on fetal loss and infant mortality: a cluster-randomized trial in Nepal. Am J Clin Nutr 2003; 78: 1194202. 121 Caulfield LE, Zavaleta N, Figueroa A, et al. Maternal zinc supplementation does not affect size at birth or pregnancy duration in Peru. J Nutr 1999; 129: 156368. 122 zaltin E, Hill K. Subramanian S. V. Association of maternal stature with offspring mortality, underweight, and stunting in low- to middle-income countries. JAMA 2010; 303: 150716. 123 Lee AC, Katz J, Blencowe H, et al, for the CHERG SGA-Preterm Birth Working Group. National and regional estimates of term and preterm babies born small-for-gestational age in 138 low-income and middle-income countries in 2010. Lancet Glob Health (in press). 124 Regev RH, Reichman B. Prematurity and intrauterine growth retardationdouble jeopardy? Clin Perinatol 2004; 31: 45373. 125 Pulver LS, Guest-Warnick G, Stoddard GJ, et al. Weight for gestational age affects the mortality of late preterm infants. Pediatrics 2009; 123: e107277. 126 Narchi H, Skinner A, Williams B. Small for gestational age neonates are we missing some by only using standard population growth standards and does it matter? J Matern Fetal Neonatal Med 2010; 23: 4854. 127 Kristensen S, Salihu HM, Keith LG, et al. SGA subtypes and mortality risk among singleton births. Early Hum Dev 2007; 83: 99105. 128 Grisaru-Granovsky S, Reichman B, Lerner-Geva L, et al. Mortality and morbidity in preterm small-for-gestational-age infants: a population-based study. Am J Obstet Gynecol 2012; 206: 150 e17. 129 Clausson B, Cnattingius S, Axelsson O. Preterm and term births of small for gestational age infants: a population-based study of risk factors among nulliparous women. Br J Obstet Gynaecol 1998; 105: 101117. 130 Katz J, Lee AC, Kozuki N, et al, and the CHERG Small-forGestational-Age-Preterm Birth Working Group. Mortality risk in preterm and small-for-gestational-age infants in low-income and middle-income countries: a pooled country analysis. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S01406736(13)60993-9. 131 UNICEF, WHO, World Bank, UNPD. Levels and trends in child mortality, report 2011. New York, NY: United Nations International Childrens Fund, 2011. 132 Christian P, Lee S, Donahue M and the CHERG Working Group. Risk of childhood undernutrition related to small-for-gestational age and preterm birth in developing countries. Int J Epidemiol (in press). 133 Zhang J, Yu KF. Whats the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA 1998; 280: 169091. 134 Walker SP, Wachs TD, Gardner JM, et al. Child development: risk factors for adverse outcomes in developing countries. Lancet 2007; 369: 14557. 135 Walker SP, Wachs TD, Grantham-McGregor S, et al. Inequality in early childhood: risk and protective factors for early child development. Lancet 2011; 378: 132538.

136 Tofail F, Hamadani J, Ahmed A, et al. The mental development and behavior of low-birth-weight Bangladeshi infants from an urban low-income community. Eur J Clin Nutr 2011; 66: 23743. 137 Kuklina EV, Ramakrishnan U, Stein AD, et al. Early childhood growth and development in rural Guatemala. Early Hum Dev 2006; 82: 42533. 138 Martorell R, Horta BL, Adair LS, et al. Weight gain in the first two years of life is an important predictor of schooling outcomes in pooled analyses from five birth cohorts from low- and middle-income countries. J Nutr 2010; 140: 34854. 139 Li H, DiGirolamo AM, Barnhart HX, et al. Relative importance of birth size and postnatal growth for womens educational achievement. Early Hum Dev 2004; 76: 116. 140 Emond AM, Lira PIC, Lima MC, et al. Development and behaviour of low-birthweight term infants at 8 years in northeast Brazil: a longitudinal study. Acta Paediatr 2007; 95: 124957. 141 Walker SP, Chang SM, Younger N, et al. The effect of psychosocial stimulation on cognition and behaviour at 6 years in a cohort of term, low-birthweight Jamaican children. Dev Med Child Neurol 2010; 52: e14854. 142 Sabet F, Richter LM, Ramchandani PG, et al. Low birthweight and subsequent emotional and behavioural outcomes in 12-year-old children in Soweto, South Africa: findings from birth to twenty. Int J Epidemiol 2009; 38: 94454. 143 Wang WL, Sung YT, Sung FC, et al. Low birth weight, prematurity, and paternal social status: impact on the basic competence test in Taiwanese adolescents. J Pediatr 2008; 153: 33338. 144 Pongcharoen T, Ramakrishnan U, DiGirolamo AM, et al. Influence of prenatal and postnatal growth on intellectual functioning in school-aged children. Arch Pediatr Adol Med 2012; 166: 41116. 145 Haroon S, Das JK, Salam RA, et al. Breastfeeding promotion interventions and breastfeeding practices: a systematic review. BMC Public Health (in press). 146 Imdad A, Bhutta ZA. Effect of preventive zinc supplementation on linear growth in children under 5 years of age in developing countries: a meta-analysis of studies for input to the lives saved tool. BMC Public Health 2011; 11 (suppl 3): S22. 147 Checkley W, Buckley G, Gilman RH, et al. Multi-country analysis of the effects of diarrhoea on childhood stunting. Int J Epidemiol 2008; 37: 81630. 148 Humphrey JH. Child undernutrition, tropical enteropathy, toilets, and handwashing. Lancet 2009; 374: 103235. 149 Adair LS, Fall C, Osmond C, et al. Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: findings from five birth cohort studies. Lancet 2013; published online March 28. http:// dx.doi.org/10.1016/S1040-6736(13)60103-8. 150 Kuzawa CW, Hallal PC, Adair L, et al. Birth weight, postnatal weight gain, and adult body composition in five low and middle income countries. Am J Hum Biol 2012; 24: 513. 151 Kipping RR, Jago R, Lawlor DA. Obesity in children. Part 1: Epidemiology, measurement, risk factors, and screening. BMJ 2008; 337: a1824. 152 Kipping RR, Jago R, Lawlor DA. Obesity in children. Part 2: Prevention and management. BMJ 2008; 337: a1848. 153 Han JC, Lawlor DA, Kimm SY. Childhood obesity. Lancet 2010; 375: 173748. 154 UNICEF, WHO, World Bank. Levels and trends in child malnutrition. Joint child malnutrition estimates. New York, NY: United Nations International Childrens Fund; Geneva: World Health Organization; Washington, DC: World Bank, 2012. 155 de Onis M, Blssner M, Borghi E, et al. Methodology for estimating regional and global trends of child malnutrition. Int J Epidemiol 2004; 33: 126070. 156 de Onis M, Blssner M, Borghi E. Prevalence and trends of stunting among pre-school children, 19902020. Pub Health Nutr 2011; 15: 14248. 157 Stevens GA, Finucane MM, Paciorek CJ, et al. Trends in mild, moderate, and severe stunting and underweight, and progress towards MDG 1 in 141 developing countries: a systematic analysis of population representative data. Lancet 2012; 380: 82434.

www.thelancet.com

37

Series

158 Barros AJ, Victora C. Measuring coverage in MNCH: determining and interpreting inequalities in coverage of maternal, newborn, and child health interventions. PLoS Med 2013; 10: e1001390. 159 Pelletier DL, Frongillo EA Jr, Schroeder DG, et al. The effects of malnutrition on child mortality in developing countries. Bull World Health Organ 1995; 73: 44348. 160 Caulfield LE, de Onis M, Blossner M, et al. Undernutrition as an underlying cause of child deaths associated with diarrhea, pneumonia, malaria, and measles. Am J Clin Nutr 2004; 80: 19398. 161 Olofin I, McDonald CM, Ezzati M, et al, for the Nutrition Impact Model Study (anthropometry cohort pooling). Associations of suboptimal growth with all-cause and cause-specific mortality in children under five years: pooled analysis of ten prospective studies. PLoS One (in press). 162 Liu L, Johnson HL, Cousens S, et al. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. Lancet 2012; 379: 215161. 163 Grantham-McGregor S, Cheung YB, Cueto S, et al. Developmental potential in the first 5 years for children in developing countries. Lancet 2007; 369: 6070. 164 Walker CLF, Lamberti L, Adair L, et al. Does childhood diarrhea influence cognition beyond the diarrhea-stunting pathway? PloS One 2012; 7: e47908. 165 Aburto NJ, Ramirez-Zea M, Neufeld LM, et al. Some indicators of nutritional status are associated with activity and exploration in infants at risk for vitamin and mineral deficiencies. J Nutr 2009; 139: 175157. 166 Gardner JM, Grantham-McGregor SM, Himes J, et al. Behaviour and development of stunted and nonstunted Jamaican children. J Child Psychol Psychiatry 1999; 40: 81927. 167 Kuklina EV, Ramakrishnan U, Stein AD, et al. Early childhood growth and development in rural Guatemala. Early Hum Dev 2006; 82: 42533. 168 Adair LS, Fall CH, Osmond C, et al, for the COHORTS group. Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: findings from five birth cohort studies. Lancet 2013; published online March 28. http://dx.doi.org.10.1016/S01406736(13)60103-8. 169 Cheung YB, Ashorn P. Continuation of linear growth failure and its association with cognitive ability are not dependent on initial length-for-age: a longitudinal study from 6 months to 11 years of age. Acta Paediatr 2010; 99: 171923. 170 Gandhi M, Ashorn P, Maleta K, et al. Height gain during early childhood is an important predictor of schooling and mathematics ability outcomes. Acta Paediatr 2011; 100: 111318. 171 Koplan JP, Liverman CT, Kraak VI, Committee on Prevention of Obesity in Children and Youth. Preventing childhood obesity: health in the balance: executive summary. J Am Diet Assoc 2005; 105: 13138. 172 Lloyd LJ, Langley-Evans SC, McMullen S. Childhood obesity and risk of the adult metabolic syndrome: a systematic review. Int J Obes (Lond) 2012; 36: 111. 173 Sachdev H, Gera T, Nestel P. Effect of iron supplementation on mental and motor development in children: systematic review of randomised controlled trials. Public Health Nutr 2005; 8: 11732. 174 Szajewska H, Ruszczynski M, Chmielewska A. Effects of iron supplementation in nonanemic pregnant women, infants, and young children on the mental performance and psychomotor development of children: a systematic review of randomized controlled trials. Am J Clin Nutr 2010; 91: 168490. 175 Grantham-McGregor S, Baker-Henningham H. Iron deficiency in childhood: causes and consequences for child development. Ann Nestl 2010; 68: 10519. 176 Aukett M, Parks Y, Scott P, et al. Treatment with iron increases weight gain and psychomotor development. Arch Dis Child 1986; 61: 84957. 177 Idjradinata P, Pollitt E. Reversal of developmental delays in iron-deficient anaemic infants treated with iron. Lancet 1993; 341: 14. 178 Stoltzfus RJ, Kvalsvig JD, Chwaya HM, et al. Effects of iron supplementation and anthelmintic treatment on motor and language development of preschool children in Zanzibar: double blind, placebo controlled study. BMJ 2001; 323: 138993.

179 Lind T, Lnnerdal B, Stenlund H, et al. A community-based randomized controlled trial of iron and zinc supplementation in Indonesian infants: interactions between iron and zinc. Am J Clin Nutr 2003; 77: 88390. 180 Black MM, Baqui AH, Zaman K, et al. Iron and zinc supplementation promote motor development and exploratory behavior among Bangladeshi infants. Am J Clin Nutr 2004; 80: 90310. 181 Friel JK, Aziz K, Andrews WL, et al. A double-masked, randomized control trial of iron supplementation in early infancy in healthy term breast-fed infants. J Pediatr 2003; 143: 58286. 182 Moffatt M, Longstaffe S, Besant J, et al. Prevention of iron deficiency and psychomotor decline in high-risk infants through use of iron-fortified infant formula: a randomized clinical trial. J Pediatr 1994; 125: 52734. 183 Pongcharoen T, DiGirolamo AM, Ramakrishnan U, et al. Long-term effects of iron and zinc supplementation during infancy on cognitive function at 9 y of age in northeast Thai children: a follow-up study. Am J Clin Nutr 2011; 93: 63643. 184 Olney DK, Pollitt E, Kariger PK, et al. Combined iron and folic acid supplementation with or without zinc reduces time to walking unassisted among Zanzibari infants 5- to 11-mo old. J Nutr 2006; 136: 242734. 185 Katz J, Khatry SK, Leclerq SC, et al. Daily supplementation with iron plus folic acid, zinc, and their combination is not associated with younger age at first walking unassisted in malnourished preschool children from a deficient population in rural Nepal. J Nutr 2010; 140: 131721. 186 Surkan PJ, Siegel EH, Patel SA, et al. Effects of zinc and iron supplementation fail to improve motor and language milestone scores of infants and toddlers. Nutrition 2013; 29: 54248. 187 Siegel EH, Kordas K, Stoltzfus RJ, et al. Inconsistent effects of iron-folic acid and/or zinc supplementation on the cognitive development of infants. J Health Popul Nutr 2011; 29: 593604. 188 Beaton GH, Martorell R, Aronson K, et al. Effectiveness of vitamin A supplementation in the control of young child morbidity and mortality in developing countriesnutrition policy discussion paper no 13. Toronto, ON: International Nutrition Program, 1993. 189 Fawzi WW, Chalmers TC, Herrera MG, et al. Vitamin A supplementation and child mortality. JAMA 1993; 269: 898903. 190 Mayo-Wilson E, Imdad A, Herzer K, et al. Vitamin A supplements for preventing mortality, illness, and blindness in children aged under 5: systematic review and meta-analysis. BMJ 2011; 343: d5094. 191 Imdad A, Yakoob MY, Sudfeld C, et al. Impact of vitamin A supplementation on infant and childhood mortality. BMC Public Health 2011; 11 (suppl 3): S20 192 Awasthi S, Peto R, Read S, et al. Population deworming every 6 months with albendazole in 1 million pre-school children in north India: DEVTA, a cluster-randomised trial. Lancet 2013; 381: 147886. 193 Rahmathullah L, Tielsch JM, Thulasiraj RD, et al. Impact of supplementing newborn infants with vitamin A on early infant mortality: community based randomised trial in southern India. BMJ 2003; 327: 254. 194 Klemm RD, Labrique AB, Christian P, et al. Newborn vitamin A supplementation reduced infant mortality in rural Bangladesh. Pediatrics 2008; 122: e24250. 195 Humphrey JH, Agoestina T, Wu L, et al. Impact of neonatal vitamin A supplementation on infant morbidity and mortality. J Pediatr 1996; 128: 48996. 196 International Zinc Nutrition Consultative Group, Brown KH, Rivera JA, et al. International Zinc Nutrition Consultative Group (IZiNCG) technical document #1. Assessment of the risk of zinc deficiency in populations and options for its control. Food Nutr Bull 2004; 25 (1 suppl 2): S99203. 197 Yakoob MY, Theodoratou E, Jabeen A, et al. Preventive zinc supplementation in developing countries: impact on mortality and morbidity due to diarrhea, pneumonia and malaria. BMC Public Health 2011; 11 (suppl 3): S23. 198 Brown KH, Peerson JM, Baker SK, et al. Preventive zinc supplementation among infants, preschoolers, and older prepubertal children. Food Nutr Bull 2009; 30 (suppl 1): S1240.

38

www.thelancet.com

Series

199 Rice AL, West KP Jr, Black RE. Vitamin A deficiency. In: WHO. Comparative quantification of health risks: global and regional burden of disease attributes to selected major risk factors. Geneva: World Health Organization, 2004: 21156. 200 Lim SS, Vos T, Flaxman AD, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 19902010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2013; 380: 222460. 201 Imdad A, Bhutta ZA. Effect of preventive zinc supplementation on linear growth in children under 5 years of age in developing countries: a meta-analysis of studies for input to the lives saved tool. BMC Public Health 2011; 11 (suppl 3): S22. 202 Lamberti LM, Fischer Walker CL, Noiman A, et al. Breastfeeding and the risk for diarrhea morbidity and mortality. BMC Public Health 2011; 11 (suppl 3): S15. 203 Lamberti LM, Zakarija-Grkovic I, Fischer Walker CL, et al. Breastfeeding for reducing the risk of pneumonia morbidity and mortality in children under two: a systematic literature review and meta-analysis. BMC Public Health (in press). 204 Debes AK, Kohli A, Walker N, et al. Time to initation of breastfeeding and neonatal mortality and morbidity: a systematic review. BMC Public Health (in press). 205 Horta BL, Victora C. Long-term effects of breastfeeding: a series of systematic reviews. Geneva: World Health Organization, 2013. 206 Ezzati M, Lopez AD, Rodgers A, et al, and the Comparative Risk Assessment Collaborating Group. Selected major risk factors and global and regional burden of disease. Lancet 2002; 360: 134760. 207 Ezzati M, Hoorn SV, Rodgers A, et al, and the Comparative Risk Assessment Collaborating Group. Estimates of global and regional potential health gains from reducing multiple major risk factors. Lancet 2003; 362: 27180. 208 Richter LM, Victora CG, Hallal PC, et al. Cohort profile: the consortium of health-orientated research in transitioning societies. Int J Epidemiol 2012; 41: 62126. 209 Walker SP, Chang SM, Powell CA, et al. Early childhood stunting is associated with poor psychological functioning in late adolescence and effects are reduced by psychosocial stimulation. J Nutr 2007; 137: 246469. 210 Galler JR, Bryce CP, Waber D, et al. Early childhood malnutrition predicts depressive symptoms at ages 1117. J Child Psychol Psychiatry 2010; 51: 78998. 211 Cheung YB, Ashorn P. Linear growth in early life is associated with suicidal ideation in 18-year-old Filipinos. Paediatr Perinat Epidemiol 2009; 23: 46371. 212 Galler JR, Bryce CP, Zichlin ML, et al. Infant malnutrition is associated with persisting attention deficits in middle adulthood. J Nutr 2012; 142: 78894. 213 Maluccio JA, Hoddinott J, Behrman JR, et al. The impact of improving nutrition during early childhood on education among Guatemalan adults. Econ J 2009; 119: 73463.

214 Hoddinott J, Maluccio JA, Behrman JR, et al. Effect of a nutrition intervention during early childhood on economic productivity in Guatemalan adults. Lancet 2008; 371: 41116. 215 Stein AD, Wang M, Ramirez-Zea M, et al. Exposure to a nutrition supplementation intervention in early childhood and risk factors for cardiovascular disease in adulthood: evidence from Guatemala. Am J Epidemiol 2006; 164: 116070. 216 Lumey LH, Stein AD, Susser E. Prenatal famine and adult health. Ann Rev Public Health 2011; 32: 23762. 217 Chen Y, Zhou L-A. The long-term health and economic consequences of the 19591961 famine in China. J Health Econ 2007; 26: 65981.218 Huang C, Li Z, Wang M, et al. Early life exposure to the 19591961 Chinese famine has long-term health consequences. J Nutr 2010; 140: 187478. 219 Song S, Wang W, Hu P. Famine, death, and madness: schizophrenia in early adulthood after prenatal exposure to the Chinese great leap forward famine. Soc Sci Med 2009; 68: 131521. 220 Huang C, Phillips MR, Zhang Y, et al. Malnutrition in early life and adult mental health: evidence from a natural experiment. Soc Sci Med 2012; published online Dec 20. DOI:10.1016/j.socscimed.2012.09.051. 221 Huang C, Li Z, Venkat Narayan KM, et al. Bigger babies born to women survivors of the 19591961 Chinese famine: a puzzle due to survival selection? J Dev Orig Health Dis 2010; 1: 41218. 222 WHO. Discussion paper. Proposed global targets for maternal, infant and young child nutrition. Geneva: World Health Organization, 2012. 223 Lozano R, Naghavi M, Foreman K, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2013; 380: 2095128. 224 de Onis M, Blssner M, Borghi E. Global prevalence and trends of overweight and obesity among preschool children. Am J Clin Nutr 2010; 92: 125764. 225 McCormick DP, Sarpong K, Jordan L, et al. Infant obesity: are we ready to make this diagnosis? J Pediatr 2010; 157: 1519. 226 Evellen vanDijk C, Innis SM. Growth-curve standards and the assessment of early excess weight gain in infancy. Pediatrics 2009; 123: 10208. 227 Bruce KD, Hanson MA. The developmental origins, mechanisms, and implications of metabolic syndrome. J Nutr 2010; 140: 64852. 228 Gewa CA. Childhood overweight and obesity among Kenyan pre-school children: association with maternal and early child nutritional factorserratum. Pub Health Nutr 2009; 13: 496503. 229 Griffiths L, Hawkins S, Cole T, et al. Risk factors for rapid weight gain in preschool children: findings from a UK-wide prospective study. Int J Obes (Lond) 2010; 34: 62432.

www.thelancet.com

39

Series

Maternal and Child Nutrition 2 Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost?
Zulfiqar A Bhutta, Jai K Das, Arjumand Rizvi, Michelle F Gaffey, Neff Walker, Susan Horton, Patrick Webb, Anna Lartey, Robert E Black, The Lancet Nutrition Interventions Review Group, and the Maternal and Child Nutrition Study Group
Published Online June 6, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)60996-4 This is the second in a Series of four papers about maternal and child nutrition Aga Khan University, Karachi, Pakistan (Prof Z A Bhutta PhD, J K Das MBA, A Rizvi MSc); Hospital for Sick Children, Toronto, ON, Canada (M F Gaffey MSc); Johns Hopkins University, Bloomberg School of Public Health, Baltimore, MD, USA (N Walker PhD, Prof R E Black PhD); University of Waterloo, Waterloo, ON, Canada (Prof S Horton PhD); Tufts University, Medford, MA, USA (Prof P Webb PhD); and University of Ghana, Accra, Ghana (Prof A Lartey PhD) Correspondence to: Prof Zulfiqar A Bhutta, Center of Excellence in Women and Child Health, The Aga Khan University, Karachi 74800, Pakistan zulfiqar.bhutta@aku.edu

Maternal undernutrition contributes to 800 000 neonatal deaths annually; stunting, wasting, and micronutrient deficiencies are estimated to underlie nearly 31 million child deaths annually. Progress has been made with many interventions implemented at scale and the evidence for effectiveness of nutrition interventions and delivery strategies has grown since The Lancet Series on Maternal and Child Undernutrition in 2008. We did a comprehensive update of interventions to address undernutrition and micronutrient deficiencies in women and children and used standard methods to assess emerging new evidence for delivery platforms. We modelled the effect on lives saved and cost of these interventions in the 34 countries that have 90% of the worlds children with stunted growth. We also examined the effect of various delivery platforms and delivery options using community health workers to engage poor populations and promote behaviour change, access and uptake of interventions. Our analysis suggests the current total of deaths in children younger than 5 years can be reduced by 15% if populations can access ten evidence-based nutrition interventions at 90% coverage. Accelerated gains are possible and about a fifth of the existing burden of stunting can be averted using these approaches, if access is improved in this way. The estimated total additional annual cost involved for scaling up access to these ten direct nutrition interventions in the 34 focus countries is Int$96 billion per year. Continued investments in nutrition-specific interventions to avert maternal and child undernutrition and micronutrient deficiencies through community engagement and delivery strategies that can reach poor segments of the population at greatest risk can make a great difference. If this improved access is linked to nutrition-sensitive approachesie, womens empowerment, agriculture, food systems, education, employment, social protection, and safety netsthey can greatly accelerate progress in countries with the highest burden of maternal and child undernutrition and mortality.

Introduction
Stunting prevalence has been decreasing slowly and 165 million children were stunted in 2011.1 Under nutrition, consisting of fetal growth restriction, stunt ing, wasting, and deficiencies of vitamin A and zinc, along with sub optimum breastfeeding, underlies nearly 31 million deaths of children younger than 5 years annually world wide, representing about 45% of all deaths in this group.2 Maternal and child obesity have also increased in many low-income and middleincome countries.3 In a comprehensive review of nutrition interven tions, we previously assessed 43 nutrition-related inter ventions in detail and reported estimates of efficacy and effect for 11 core interventions.4 Much progress has been made since with many interventions implemented at scale, assessments of promising new interventions, and new delivery strategies. We used standard methods to do a comprehensive review of potential nutritionspecific inter ventions to address undernutrition and micro nutrient deficiencies in women and children. We modelled the potential effect of delivery of these inter ventions on lives saved in the 34 countries with 90% of the global burden of stunted children, and estimated the effect of various delivery platforms that could enhance equitable scaling up of nutritionspecific interventions.

Selection of interventions for review


We selected several nutrition-specific interventions across the lifecycle for assessment of evidence of benefit (figure 1); these interventions included those affect ing adolescents, women of reproductive age, pregnant women, newborn babies, infants, and children. We also reviewed the evidence for delivery platforms for nutri tion interventions and other emerging interventions of interest for nutrition of women and children. We identified and relied on the most recent reviews with good quality methods for all interventions and updated the evidence by incorporating newer studies, when available. For other identified interventions, when we did not find any relevant review, we did a de-novo review using the methodology described in panel 1.5 Additionally, we consulted the electronic library on nutrition actions (eLENA) for existing evidence used by WHO for development of guidelines and policies for action (appendix p 2).

See Online for appendix

Interventions to address adolescent health and nutrition


There is growing interest in adolescent health as an entry point to improve the health of women and chil dren, especially because an estimated 10 million girls younger than 18 years are married each year.6 A range of inter ventions exist in relation to adolescent health and
www.thelancet.com

40

Series

Key messages Globally, 165 million children are stunted; undernutrition underlies 31 million deaths in children younger than 5 years. A clear need exists to introduce promising evidence-based interventions in the preconception period and in adolescents in countries with a high burden of undernutrition and young age at first pregnancies; however, targeting and reaching a sufficient number of those in need will be challenging. Promising interventions exist to improve maternal nutrition and reduce fetal growth restriction and small-for-gestationalage (SGA) births in appropriate settings in developing countries, if scaled up before and during pregnancy. These interventions include balanced energy protein, calcium, and multiple micronutrient supplementation and preventive strategies for malaria in pregnancy. Replacement of iron-folate with multiple micronutrient supplements in pregnancy might have additional benefits for reduction of SGA in at-risk populations, although further evidence from effectiveness assessments might be needed to guide a universal policy change. Strategies to promote breastfeeding in community and facility settings have shown promising benefits on enhancing exclusive breastfeeding rates; however, evidence for long-term benefits on nutritional and developmental outcomes is scarce. Evidence for the effectiveness of complementary feeding strategies is insufficient, with much the same benefits noted from dietary diversification and education and food supplementation in food secure populations and slightly greater effects in food insecure populations. Further effectiveness trials are needed in food insecure populations with standardised foods (pre-fortified or non-fortified), duration of intervention, outcome definition, and cost effectiveness. Treatment strategies for severe acute malnutrition with recommended packages of care and ready-to-use therapeutic foods are well established, but further evidence is needed for prevention and management strategies for moderate acute malnutrition in population settings, especially in infants younger than 6 months. Data for the effect of various nutritional interventions on neurodevelopmental outcomes are scarce; future studies should focus on these aspects with consistency in measurement and reporting of outcomes. Conditional cash transfers and related safety nets can address the removal of financial barriers and promotion of access of families to health care and appropriate foods and nutritional commodities. Assessments of the feasibility and effects of such approaches are urgently needed to address maternal and child nutrition in well supported health systems. Innovative delivery strategies, especially community-based delivery platforms, are promising for scaling up coverage of nutrition interventions and have the potential to reach poor populations through demand creation and household service delivery. Nearly 15% of deaths of children younger than 5 years can be reduced (ie, 1 million lives saved), if the ten core nutrition interventions we identified are scaled up. The maximum effect on lives saved is noted with management of acute malnutrition (435000 [range 285000482000] lives saved); 221000 (135000293000) lives would be saved with delivery of an infant and young child nutrition package, including breastfeeding promotion and promotion of complementary feeding; micronutrient supplementation could save 145000 (30000216000) lives. These interventions, if scaled up to 90% coverage, could reduce stunting by 203% (335 million fewer stunted children) and can reduce prevalence of severe wasting by 614%. The additional cost of achieving 90% coverage of these proposed interventions would be Int$96 billion per year.

nutrition, which could also affect the period before first pregnancy or between pregnancies. Evi dence supporting reproductive health and family planning interventions in this age group suggests that it might be possible to reduce unwanted pregnancies and optimise age at first pregnancy. These aims might be important to reduce the risk of small-for-gestational age (SGA) births in populations in which a substantial proportion of births occur in adolescents. Opportunities might also exist to address micronutrient deficiencies and emerging issues of overweight and obesity in adoles cents through com munity and school-based edu cation platforms. Although evidence from robust randomised controlled trials is scarce, we identified a range of interventions in the adol es cent period affecting maternal, newborn, and child health and nutrition outcomes (panel 2718).
www.thelancet.com

Interventions in women of reproductive age and during pregnancy


Folic acid supplementation
Neural tube defects can be effectively prevented with peri conceptional folic acid supplementation. A review19 of five trials of periconceptional folic acid supplemen tation suggested a 72% reduction in risk of develop ment of neural tube defects and a 68% reduction in risk of recur rence compared with either no intervention, placebo, or micro nutrient intake without folic acid (table 11926). A review20 of folic acid supplementation during pregnancy showed that folic acid supplementation improved mean birthweight, with a 79% reduction in the incidence of megaloblastic anaemia (table 11926). Further more no evi dence of adverse effects was noted from folic acid supple mentation in pro gramme settings. Despite
41

Series

Preconception care: family planning, delayed age at rst pregnancy, prolonging of inter-pregnancy interval, abortion care, psychosocial care

Folic acid supplementation Multiple micronutrient supplementation Calcium supplementation Balanced energy protein supplementation Iron or iron plus folate Iodine supplementation Tobacco cessation

Delayed cord clamping Early initiation of breast feeding Vitamin K administration Neonatal vitamin A supplementation Kangaroo mother care

Exclusive breast feeding Complementary feeding Vitamin A supplementation (659 months) Preventive zinc supplementation Multiple micronutrient supplementations Iron supplementation Decreased maternal and childhood morbidity and mortality Improved cognitive growth and neurodevelopmental outcomes

Adolescent

WRA and pregnancy

Neonates

Infants and children

Disease prevention and treatment Malaria prevention in women Maternal deworming Obesity prevention

Disease prevention and treatment Management of SAM Management of MAM Therapeutic zinc for diarrhoea WASH Feeding in diarrhoea Malaria prevention in children Deworming in children Obesity prevention

Increased work capacity and productivity Economic development

Delivery platforms: Community delivery platforms, integrated management of childhood illnesses, child health days, school-based delivery platforms, nancial platforms, fortication strategies, nutrition in emergencies Bold=Interventions modelled Italics=Other interventions reviewed

Figure 1: Conceptual framework WRA=women of reproductive age. WASH=water, sanitation, and hygiene. SAM=severe acute malnutrition. MAM=moderate AM.

Panel 1: Methods, search strategy, and selection criteria As per the Child Health Epidemiology Reference Group (CHERG) systematic review guidelines,5 we searched PubMed, Cochrane libraries, electronic library on evidence on nutrition actions (eLENA), and WHO regional databases and included publications in every language available in these databases. We used Medical Subject Heading Terms (MeSH) and keyword search strategies with various combinations of relevant terms. We made every effort to gather unpublished data when reports were available for full abstraction. Inclusion and exclusion criteria were established for each area of review, and studies meeting these criteria were double data extracted and categorised according to outcome. Evidence was then summarised by outcome and study design, including study quality, generalisability, and summary outcome measures. We did meta-analyses for each outcome containing more than one study, using either the Mantel-Haenszel or the Der Simonian-Laird pooled relative risks (RR, with 95% CIs), when there was unexplained heterogeneity of effect. Heterogeneity was assessed by visual inspection of forest plots and by the 2 p value (p<010). The binary measure for individual studies and pooled statistics was reported as the RR between the experimental and control groups with 95% CIs. For the outcome of interest for each intervention, we applied the CHERG Rules for Evidence Review5 to generate a final estimate.

other foods might be a feasible way to reach the population in need.

Iron or iron and folic acid supplementation


A review21 of iron supplementation in non-pregnant women of reproductive age showed that intermittent iron supplementation (alone or with any other vitamins and minerals) reduced the risk of anaemia by 27% (table 11926). A Cochrane review22 of daily iron supplementation to women during pregnancy reported a 70% reduction in anaemia at term, a 67% reduction in iron deficiency anaemia (IDA), and 19% reduction in the incidence of low birthweight. Another review27 further suggests that the effects were much the same in women receiving intermittent iron supplementation, or daily iron, or iron and folic acid supplementation. Although some evidence suggests that side-effects are fewer with intermittent iron therapy in nonanaemic populations, WHO recommends daily iron supplementation during pregnancy as part of the standard of care in populations at risk of iron deficiency.28

strong evidence of benefit, reaching women of reproduc tive age in the peri conceptual period to provide folic acid supple ments through existing delivery platforms remains a logistical challenge. Fortification of cereals and
42

Maternal multiple micronutrient supplementation


Multiple-micronutrient deficiencies often coexist in lowincome and-middle-income countries (LMICs) and can be exacerbated in pregnancy with potentially adverse
www.thelancet.com

Series

maternal outcomes. A Cochrane review23 of multiple nutrient supplementation in pregnant women micro assessed 23 trials and reported an 1113% reduction in low birthweight and SGA births, whereas effects on anaemia and IDA were much the same when compared with iron and folic acid supplements (table 1). Despite earlier concerns about potential excess neonatal mortal ity with multiple micro nutrient use,29 present analyses suggest no births, perinatal, adverse effects on maternal mortality, still and neonatal mortality with insuffi cient data for neuro developmental outcomes. Although scarce, there are interesting data for benefits of maternal multiple micro nutrient supple mentation on growth in early childhood.30 Preliminary data from a large trial31 comparing multiple micronutrient with iron-folate supplementation in preg nancy in Bangladesh show a significant reduction in pre term births with no adverse effects. Inclusion of this study in our meta-analysis confirms the reduction in low birth weight (relative risk [RR] 088, 95% CI 085091) and SGA (089, 083096) and is also indicative of a small effect on preterm births (097, 094099). These find ings support the potential replacement of iron-folate supplements in pregnancy with multiple micronutrient supplements in populations at risk.

Panel 2: Interventions to address adolescent nutrition and preconception care Women of lower socioeconomic status and young age are at risk of being undernourished and underweight. Ronnenberg and colleagues7 assessed the association between preconception anaemia and poor fetal and neonatal outcomes. They showed that the risk of being born low birthweight was significantly greater with moderate preconception anaemia (odds ratio [OR] 65, 95% CI 16267) and fetal growth restriction (46, 15135). Important factors indirectly related to maternal, fetal, and neonatal nutritional status and pregnancy outcomes include young age at first pregnancy and repeated pregnancies. Young girls who are not physically mature might enter pregnancy with depleted nutrition reserves and anaemia.8 Adolescent pregnancy is associated with a 50% increased risk of stillbirths and neonatal deaths, and increased risk of preterm birth, low birthweight, and asphyxia.911 Adolescents are especially prone to complications of labour and delivery, such as obstructed and prolonged labour, vesico-vaginal fistulae, and infectious morbidity.11 In societies in which most births are within a marital relationship, interventions to increase the age at marriage and first pregnancy are important.12 Evidence suggests that programmes for adolescent mothers can reduce repeat adolescent pregnancies by 37% (95% CI 1251%) when they teach parenting skills through home visitation and provide young mothers with education and vocational or job support. Two reviews by Conde-Agudelo and colleagues13,14 assessed the association between inter-pregnancy intervals with maternal, newborn, and child health outcomes and found a J-shaped dose-response association for perinatal outcomes. Short interpregnancy intervals (<6 months) were associated with a higher probability of maternal anaemia (32%) and stillbirths (40%) whereas longer intervals (>60 months) were associated with an increased risk of pre-eclampsia.15 Both short and long birth intervals increase the risk for preterm births (OR 145 [95% CI 130161] for short term; OR 121 [95% CI 112130] for long intervals), low birthweight (OR 165 [95% CI 127214] for short intervals; relative risk [RR] 137 [95% CI 121155] for long intervals), and neonatal mortality (OR 131 [95% CI 096179] for short interval; RR 115 [95% CI 106125] for long intervals). With repeated pregnancies and advanced maternal age there is increased risk of chromosomal abnormalities, and increased risks of gestational diabetes and hypertension, stillbirths (RR 162, 95% CI 150176), perinatal mortality (RR 144, 95% CI 110189), and low birthweight (RR 161, 95% CI 116224).16,17 These findings support the need to optimise age at first pregnancy and family size and inter-pregnancy intervals. A global unmet need exists for family planning with more than 100 million unmarried women in developing countries not using contraception.18 Optimisation of age at first pregnancy must be coupled with promotion of effective contraceptive use and exclusive breastfeeding, so that women can ideally space their pregnancies 1824 months apart.

Maternal calcium supplementation


Gestational hypertensive disorders are the second leading cause of maternal morbidity and mortality and are associated with increased risk of preterm birth and fetal growth restriction.32,33 Calcium supplementation during pregnancy in women at risk of low calcium intake has been shown to reduce maternal hypertensive dis orders and preterm birth. A Cochrane review by Hofmeyr and colleagues34 assessed 13 trials and showed that calcium supplementation during pregnancy reduced the incidence of gestational hypertension by 35%, preeclampsia by 55%, and preterm births by 24% (table 1). These estimates have been updated in a review24 of 15 randomised controlled trials, which also showed a 52% reduction in the incidence of pre-eclampsia and confirmed that these effects were only noted in popu lations at risk of low calcium intake.

Maternal iodine supplementation or fortification


In nearly all regions affected by iodine deficiency, use of iodised salt is the most cost-effective way to avert deficiency. A Cochrane review35 suggests that although iodised salt is an effective means to improve iodine status, no conclusions can be drawn about physical and mental development in children and mortality. In some regions of the world with severe iodine deficiency, salt iodisation alone might not be sufficient for control of iodine deficiency in pregnancy; in these circumstances iodised oil supplementation during pregnancy can be a viable option (table 1). A review25 of five randomised trials of iodised oil supplementation in pregnancy in iodine-deficient populations showed a 73% reduction in cretinism and a 1020% increase in
www.thelancet.com

developmental scores in children. Existing evidence supports continued focus on effective universal salt iodisation for women of reproductive age and those who are pregnant. Further high-quality controlled studies are needed to address dosage and alternative strategies for iodine supplementation in different population groups and settings.

Addressing maternal wasting and food insecurity with balanced energy and protein supplementation
Maternal undernutrition is a risk factor for fetal growth restriction and adverse perinatal outcomes.1 Several nutritional interventions have been assessed in such situations, including dietary advice to pregnant women,
43

Series

Evidence reviewed Folic acid supplementation Women of reproductive age Systematic review of five trials19 of periconceptual folic acid supplementation Pregnant women Systematic review of 31 trials20

Setting Developing and developed countries Mostly developed countries

Estimates Significant effects: NTDs (RR 028, 95% CI 015052), recurrence of NTDs (RR 032, 95% CI 017060) Non-significant effects: other congenital abnormalities, miscarriages, still births Significant effects: mean birthweight (MD 13575, 95% CI 478522368), incidence of megaloblastic anaemia (RR 021, 95% CI 011038) Non-significant effects: preterm birth, still births, mean predelivery haemoglobin, serum folate, red cell folate Intermittent iron supplementation Significant effects: anaemia (RR 073, 95% CI 056095), serum haemoglobin concentration (MD 458 g/L, 95% CI 256659), serum ferritin concentration (MD 832, 95% CI 4971166) Non-significant effects: iron deficiency, adverse events, depression Daily iron-alone supplementation Significant effects: low birthweight (RR 081, 95% CI 068097), birthweight (MD 3081 g, 95% CI 5945568), serum haemoglobin concentration at term (MD 888 g/L, 95% CI 6961080), anaemia at term (RR 030, 95% CI 019046), iron deficiency (RR 043, 95% CI 027066), iron deficiency anaemia (RR 033, 95% CI 016069), side-effects (RR 236, 95% CI 096582) Non-significant effects: premature delivery, neonatal death, congenital anomalies Iron-folate supplementation Significant effects: birthweight (MD 577 g, 95% CI 76610779), anaemia at term (RR 034, 95% CI 021054), serum haemoglobin concentration at term (MD 1613 g/L, 95% CI 12741952) Non-significant effects: low birthweight, premature birth, neonatal death, congenital anomalies Significant effects: low birthweight (RR 088, 95% CI 085091), SGA (RR 089, 95% CI 083096), preterm birth (RR 097, 95% CI 094099) Non-significant effects: miscarriage, maternal mortality, perinatal mortality, stillbirths, and neonatal mortality Insufficient data for neurodevelopmental outcomes Significant effects: pre-eclampsia (RR 048, 95% CI 034067), birthweight 85 g (95% CI 37133), preterm birth (RR 076, 95% CI 060097) Non-significant effects: perinatal mortality, low birthweight, neonatal mortality Significant effects: cretinism at 4 years of age (RR 027, 95% CI 012060), developmental scores 1020% higher in young children, birthweight 382630% higher Significant effects: SGA (RR 066, 95% CI 049089), stillbirths (RR 062, 95% CI 040098, birthweight (MD 73g, 95% CI 30117) Non-significant effects: Bayley mental scores at 1 year

Iron and Iron-folate supplementation Women of reproductive age Systematic review of 21 RCTs and quasi-experimental studies21 Pregnant women Systematic review of 43 RCTs and quasi-experimental studies22 (34 iron alone, eight iron-folate) Developing and developed countries. Intervention mostly given in school settings. Mostly effectiveness studies Developed and developing countries. Intervention delivered in community or at facility antenatal clinic. Mostly effectiveness studies

MMN supplementation Pregnant women Systematic review of 21 RCTs23 Developed and developing countries. Studies compared MMN with two or fewer micronutrients

Calcium supplementation Pregnant women Systematic review of 15 RCTs24 Developed and developing countries. Mostly effectiveness trials Mostly developing countries. Mostly effectiveness trials

Iodine through iodisation of salt Pregnant women Systematic review of five RCTs25

Maternal supplementation with balanced energy protein Pregnant women Systematic review of 16 RCTs Developing and developed and quasi-experimental countries studies26

NTD=neural tube defects. RR=relative risk. MD=mean difference. RCT=randomised controlled trial. MMN=multiple micronutrient. SGA=small-for-gestational age.

Table 1: Review of nutrition interventions for women of reproductive age and during pregnancy

provision of balanced energy protein supplements, and high protein or isocaloric protein supplementation. In other contexts, prescription and promotion of low energy diets to pregnant women who are either overweight or exhibit high weight gain in early gestation have been assessed.36 Balanced energy protein supplementation, providing about 25% of the total energy supplement as protein, is deemed an important intervention for preven tion of adverse perinatal outcomes in mal nourished women.26,37 A Cochrane review38 con cluded that balanced energy protein supplementation reduced the incidence of SGA by 32% and risk of stillbirths by 45% (table 1). An updated meta-analysis showed that balanced energy protein supplementation increased birthweight by 73 g (95% CI 30117) and reduced risk of SGA by 34%, with more pronounced effects in mal nourished women.26
44

Nutrition interventions in neonates


Delayed cord clamping
Early clamping of the umbilical cord after birth is a common practice and permits immediate transfer of the baby for care as required, whereas delaying of clamping allows continued blood flow between the placenta and the baby for a longer duration. A Cochrane review39 suggested that delayed cord clamping in term neonates led to significant increase in newborn haemoglobin and higher serum ferritin concentration at 6 months of age (table 23945). Another review40 of studies in preterm neonates concluded that delayed cord clamping was associated with 39% reduction in need for blood transfusion and a lower risk of complications after birth. Although promising, these strategies have as yet not been assessed for effect or feasibility of implementation at scale in health systems.
www.thelancet.com

Series

Evidence reviewed Delayed cord clamping Term neonates Systematic review of 11 RCTs39

Setting Developing and developed countries. 24 and 36 weeks gestation at birth Developing and developed countries

Estimates Significant effects: increased newborn haemoglobin conentration (MD 217 g/dL, 95% CI 028406) Non-significant effects: postpartum haemorrhage, severe postpartum haemorrhage Delayed cord clamping was associated with an increased requirement for phototherapy for jaundice Significant effects: reduced need for blood transfusion (RR 061, 95% CI 046081), decrease in intraventricular haemorrhage (RR 059, 95% CI 041085), reduced risk of necrotising enterocolitis (RR 062, 95% CI 043090) Peak bilirubin concentration was higher for delayed cord clamping group (MD 1501 mmol/L, 95% CI 5622440) Significant effects: One dose of intramuscular vitamin K reduced clinical bleeding at 17 days and improved biochemical indices of coagulation status. Oral vitamin K also improved coagulation status

Preterm neonates

Systematic review of 15 RCTs40

Neonatal vitamin K administration Neonates Systematic review of two Developing and RCTs for intramuscular developed countries vitamin K and 11 RCTs for oral vitamin K41 Developed countries

Vitamin A supplementation Very low birthweight Systematic review of infants nine RCTs42 Term neonates Systematic review of five RCTs and quasiexperimental studies43 Systematic review of 34 RCTs44 Systematic review of 16 RCTs45 Significant effects: reduced number of deaths and oxygen requirement at 1 month of age. Non-significant effects: one large trial showed no significant effect on neurodevelopment assessment at 1822 months of age Significant effects: reduction in infant mortality at 6 months (RR 086, 95% CI 077097) Non-significant effects: infant mortality at 12 months (RR 103, 95% CI 087123) Little data available for cause specific mortality, morbidity, vitamin A deficiency, anaemia, and adverse events Significant effects: increase in breastfeeding at 14 months after birth (RR 127, 95% CI 106153), increased breastfeeding duration (MD 4255 days, 95% CI 1698679) Significant effects: reduction in the risk of mortality (RR 060, 95% CI 039093), reduction in nosocomial infection and sepsis (RR 042, 95% CI 024073), reduction in hypothermia (RR 023, 95% CI 010055), reduced length of hospital stay (MD 24 days, 95% CI 0741)

Developing countries

Kangaroo mother care for promotion of breastfeeding and care of preterm and SGA infants Healthy neonates Preterm neonates Developing and developed countries Developing and developed countries

RCT=randomised controlled trial. MD=mean difference. RR=relative risk. SGA=small-for-gestational age.

Table 2: Review of nutrition interventions in neonates

Neonatal vitamin K administration


Vitamin K deficiency can result in bleeding in the first weeks of life and vitamin K is commonly given prophylactically after birth for prevention of bleeding. In the absence of vitamin K prophylaxis there is a 0417% risk of development of clinically significant bleeding. A Cochrane review41 suggested that one dose of intra muscular vitamin K, when compared with placebo, reduced clinical bleeding at 17 days of life, including bleeding after circumcision (table 2). Oral and intra muscular vitamin K had much the same effects on improved biochemical indices of coagulation status at 17 days. Currently, vitamin K administration after birth is largely restricted to births in health facilities; no information is available on the public health significance of vitamin K deficiency-related bleeding in LMICs or population-based programmes for prevention.

review43 did report a 14% reduction in the risk of infant mortality at 6 months of age, four more trials are currently underway in Asia and Africa, and researchers agree that these additional data will be needed before development of recommendations for neonatal vitamin A supplementation.

Kangaroo mother care


Kangaroo mother care denotes early skin-to-skin contact between mother and baby at birth or soon thereafter, plus early and continued breastfeeding, parental support, and early discharge from hospital. A Cochrane review44 of 34 randomised controlled trials of early skin-to-skin care in healthy neonates showed a significant 27% increase in breastfeeding at 14 months of age and increased dura tion of breastfeeding (table 2). In a Cochrane review45 of 16 randomised trials, kangaroo mother care in preterm neonates was associated with a 40% reduction in the risk of mortality, a 58% reduction in nosocomial infection or sepsis, and a 77% reduction in prevalence of hypothermia. The trials included in these analyses were done in health facilities; although kangaroo mother care might also be useful for home deliveries, there is not yet evidence of effectiveness in community settings. Kangaroo mother care was also shown to increase some measures of infant growth, breastfeeding, and mother-infant attachment,45 but few studies provide objective evidence of any effect on early child development.
45

Neonatal vitamin A supplementation


A Cochrane review42 of oral or intramuscular vitamin A supplementation to very low birthweight infants showed reduced mortality and oxygen requirement at 1 month of age compared with placebo (table 2).42 Although neonatal vitamin A supplementation has also been shown to be effective in reduction of all-cause mortality by 6 months of age, evidence is conflicting, and might be related to maternal vitamin A status.46 Although a Cochrane
www.thelancet.com

Series

Nutrition interventions in infants and children


Promotion of breastfeeding and supportive strategies
WHO recommends initiation of breastfeeding within 1 h of birth, exclusive breastfeeding of infants till 6 months of age, and continued breastfeeding until 2 years of age or older.47 However, global progress on this intervention is both uneven and suboptimum.48 The exact scientific basis for the absolute early time window of feeding within the first hour after birth is weak.49,50 A systematic review51 feeding initiation within 24 h of birth suggests that breast is associated with a 4445% reduction in all-cause and infection-related neonatal mortality, and is thought to mainly operate through the effects of exclusive breast feeding. We updated the previous review by Imdad and colleagues,52 which assessed the effect of promotion interventions on occurrence of breastfeeding, and con cluded that counselling or educational inter ventions increased exclusive breastfeeding by 43% at day 1, by 30% till 1 month, and by 90% from 15 months. Significant reductions in occurrence of mothers not breast feeding were also noted; 32% reduction at day 1, 30% till 1 month, and 18% for 15 months53 (table 35362). Combined individual and group counselling seemed to be better than individual or group counselling alone. Although these results show the potential for scaling up, none of these trials address the issues of barriers around work environments and supportive strategies such as maternity leave provision. A Cochrane review63 of interventions in the workplace to support breast feeding for women found no trials. Although some trials are underway, much more needs to be done to assess innovations and strategies to promote breast feeding in working women, especially in underprivileged communities.

In an update of a previous review of complemen tary feeding,68 we assessed 16 randomised and nonrandomised controlled trials and programmes of moderate quality (table 3).54 We identified ten studies that assessed the effect of nutrition education and seven studies that assessed the effect of provision of additional complementary foods (one trial with three intervention groups was in both these categories). Studies of nutrition education in food secure populations showed a significant increase in height (standard mean difference [SMD] 035, 95% CI 008062, four studies), and HAZ (022, 001043, four studies), whereas the effect on stunting was not statistically significant (RR 070, 95% CI 049101, four studies). We identified a significant effect on weight gain (SMD 040, 95% CI 002078, four studies), whereas no effects were noted for weight-forage Z scores (WAZ; 012; 95% CI 002 to 026, four studies). Studies of nutrition education in food insecure populations (with an average daily per person income of less than US$125) showed significant effects on HAZ (SMD 025, 95% CI 009042, one study), stunting (RR 068, 95% CI 060076, one study), and WAZ (SMD 026, 95% CI 012041, two studies). The review54 did not find any eligible study that provided complementary feeding (with or without education) in a food secure population. Overall, the provision of complementary foods in food insecure populations was associated with significant gains in HAZ (SMD 039; 95% CI 005073, seven studies) and WAZ (SMD 026, 95% CI 004048, three studies), whereas the effect on stunting did not reach statistical significance (RR 033, 95% CI 011100, seven studies).

Promotion of dietary diversity and complementary feeding


Complementary feeding for infants refers to the timely introduction of safe and nutritionally rich foods in addition to breast-feeding at about 6 months of age and typically provided from 6 to 23 months of age.64 Different approaches have been used to create indicators of dietary diversity and to study its association with child mal nutrition. In seven Latin American surveys, Ruel and Menon65 noted significant associations between comple mentary feeding practices and height-for-age Z scores (HAZ). Similarly, analysis of Demographic Health Survey data to create a dietary diversity score based on seven food groups showed that increased dietary diversity was positively associated with heightfor-age HAZ in nine of 11 countries.66 More recently, WHO infant and young child feeding indicators were studied in 14 Demographic Health Survey datasets from low-income countries;67 consumption of a minimum acceptable diet with dietary diversity reduced the risk of both stunting and under weight whereas minimum meal frequency was associated with lower risk of underweight only.
46

Vitamin A supplementation in children


A Cochrane review55 of 43 randomised trials showed that vitamin A supplementation reduced all-cause mortality by 24% and diarrhoea-related mortality by 28% in children aged 659 months (table 3). Vitamin A supple mentation also reduced the incidence of diarrhoea and measles in this age group but there was no effect on mortality and morbidity related to respiratory infections. Although a large effectiveness study69 from India assessing the effect of vitamin A supplementation and deworming over several years did not show a significant effect on mortality from vitamin A supplementation (mortality ratio 096, 95% CI 089103), inclusion of these data with previous results still shows a significant, albeit lower, effect on mortality (RR 088, 95% CI 084094).55 We believe that vitamin A supplementation continues to be an effective intervention in children aged 659 months in populations at risk of vitamin A deficiency.

Iron supplementation in infants and children


A Cochrane review56 of 33 studies showed that inter mittent iron supplementation to children younger than 2 years reduced the risk of anaemia by 49% and iron
www.thelancet.com

Series

Setting Breast feeding promotion in infants

Estimates

Developing and developed Significant effects: educational or counselling interventions increased EBF by 43% (95% CI 987) at day 1, by 30% (1942) till Systematic review of countries 1 month, and by 90% (54134) from 16 months. Significant reductions in rates of no breastfeeding also noted; 32% (1346) at 110 RCTs and day 1, 30% (2038) 01 month, and 18% (1123) for 16 months. quasi-experimental studies53 Non-significant effects: predominant and partial breastfeeding Complementary feeding promotion in children 624 months of age Nutrition education in food secure populations 16 RCTs and Mostly from food secure quasi-experimental studies54 populations. Various foods Significant effects: increased height gain (SMD 035; 95% CI 008062), HAZ (SMD 022; 95% CI 001043), weight gain (SMD 040, 95% CI 002078) used Non-significant effects: stunting, WAZ Nutrition education in food insecure populations Significant effects: HAZ (SMD 025, 95% CI 009042), stunting (RR 068, 95% CI 060076), WAZ (SMD 026, 95% CI 012041) Complementary food provision with or without education in food insecure populations Significant effects: HAZ (SMD 039, 95% CI 005073), WAZ (SMD 026, 95% CI 004048) Non-significant effects: stunting (RR 033, 95% CI 011100) Preventive vitamin A supplementation in children 6 months to 5 years of age Systematic review of 43 RCTs55 Developing and developed Significant effects: reduced all-cause mortality (RR 076, 95% CI 069083), reduced diarrhoea-related mortality (RR 072, 95% CI countries 057091), reduced incidence of diarrhoea (RR 085, 95% CI 082087), reduced incidence of measles (RR 050, 95% CI 037067) Non-significant effects: measles-related and ARI-related mortality Intermittent iron supplementation Significant effects: decreased anaemia (RR 051, 95% CI 037072), decreased iron deficiency (RR 024, 95% CI 006091), increased haemoglobin concentration (MD 520 g/L, 95% CI 251788), increased ferritin concentration (MD 1417 mcg/L, 95% CI 3532481) Non-significant effects: HAZ, WAZ Evidence for mental development, motor skill development, school performance, and physical capacity was assessed by very few studies and showed no clear effect

Iron supplementation in children Systematic review of LMICs. Participants ages 33 RCTs and ranged from neonates to 56 quasi-experimental studies 19 years

Systematic review of 17 RCTs57

Developing and developed Significant effects: increased mental development score (SMD 030, 95% CI 015046), increased intelligence quotient scores countries. In children aged (8 years age; SMD 041, 95% CI 020062) Non-significant effects: Bayley mental development index in younger children (27 months old), motor development 6 months to 15 years Mostly developing countries. In children aged 6 months to 16 years Developing countries. Mostly effectiveness studies. In children aged 6 months to 11 years MMN supplementation Significant effects: increased length (MD 013, 95% CI 006021), increased weight (MD 014, 95% CI 003025) MMN might be associated with marginal increase in fluid intelligence and academic performance in healthy school children Micronutrient powders Significant effects: Reduced anaemia (RR 066, 95% CI 057077), reduced iron deficiency anaemia (RR 043, 95% CI 035052), reduced retinol deficiency (RR 079, 95% CI 064098). Improved haemoglobin concentrations (SMD 098, 95% CI 055140). MNP was associated with a significant increase in diarrhoea (RR 104, 95% CI 101106) Non-significant effects: serum ferritin, zinc deficiency, stunting, wasting, underweight, HAZ, WAZ, WHZ, fever, URI Preventive zinc supplementation Significant effects: mean height improved by 037 cm (SD 025) in children supplemented for 24 weeks, diarrhoea reduced by 13% (95% CI 619), pneumonia reduced by 19% (95% CI 1027) Non-significant effects: mortality (cause specific and all-cause)

MMN supplementation including iron in children Systematic review of 18 trials58 Systematic review of 17 RCTs59

Zinc supplementation in children Systematic review of 18 RCTs60,61 Mostly developing countries. In children younger than 5 years

Systematic review of 13 trials62

Developing and developed Non-significant effects: Mental developmental index, psychomotor development index countries. In children younger than 5 years

RCT=randomised controlled trial. EBF=exclusive breastfeeding. HAZ=height-for-age Z score. WAZ=weight-for-age Z score. WHZ=weight-for-height Z score. MMN=multiple micronutrient. ARI=acute respiratory infection. URI=upper-respiratory infection. SMD=standard mean difference. MD=mean difference. RR=relative risk.

Table 3: Review of evidence for nutrition interventions for infants and children

deficiency by 76% (table 3). The findings also suggested that inter mittent iron supplementation could be a viable public health intervention in settings in which daily supple mentation had not been implemented or was not feasible. A review57 of the effect of iron supplementation in children on mental and motor development showed only small gains in mental development and intelligence scores in supplemented school-age children who were initially anaemic or irondeficient. There was no con vincing evidence that iron
www.thelancet.com

treatment had an effect on mental development in children younger than 27 months. Since the demonstration of increased risk of admis sion to hospital and serious illnesses with iron supplementation,70 there has been concern about adminis tration of iron supplements in malaria endemic areas. WHO currently recommends administration of iron supple ments in malaria endemic areas on the stipulation that malaria prevention and treatment is made available.71
47

Series

Multiple micronutrient supplementation in children


Although the theoretical benefits of strategies to improve diet quality and micronutrient density of foods consumed by small children are well recognised, few resource-poor countries have clear policies in support of integrated strategies to control micronutrient deficiencies in young children.72 Available options include the provision of multiple micronutrients via supplements, micro nutrient powders, or fortified ready-to-use foods includ hensive ing lipid-based nutrient supplements. A compre review of the effects of multiple micronutrients compared with two or fewer micronutrients showed small benefits on linear growth (mean difference [MD] 013, 95% CI 006021) and weight gain (014, 003025) but with little evidence of effect on morbidity outcomes as suggested by individual studies (table 3).58 Another review73 of the effect of multiple micronutrient supple mentation on improvement of cognitive perfor mance in children concluded that multiple micro nutrient supple mentation might be associated with a marginal increase in reasoning abilities but not with acquired skills and knowledge. Micronutrient powders are increasingly in use at scale in programmes to address iron and multiple micronutrient deficiencies in children. We reviewed 16 randomised controlled trials to assess the effectiveness of micronutrient powders and estimated that they significantly improved haemoglobin concentration and reduced IDA by 57% and retinol deficiency by 21%.59 We noted no evidence of benefit on linear growth. However, in-line with findings from an earlier review of liquid iron supplementation trials,74 use of micronutrient powders was shown to be associated with a significant increase in the incidence of diarrhoea (RR 104, 95% CI 101106), largely because of results from a recent large clusterrandomised controlled trial of micronutrient powders in Pakistan in malnourished children.75 These find ings underscore the need for further assessment of micronutrient powder programmes in varying contexts for safety and benefits.

zinc-supplemented children compared with placebo.60 There is no convincing evidence that zinc supple mentation in infants or children results in improved motor or mental development.62

Disease prevention and management


Several interventions have the potential to affect health and nutrition outcomes through reduction in the burden of infectious diseases. Table 47788 summarises the evidence for interventions for disease prevention and management.

Prevention and treatment of severe acute malnutrition


A substantial global burden of wasting exists, especially severe acute malnutrition (SAM; weight-for-height Z score [WHZ] <3), which coexists with moderate acute mal nutrition (MAM; WHZ <2). In stable non-emer gency situations with endemic malnutrition, MAM can often present in combination with stunting. Most of the inter ventions previously discussed should be imple mented to prevent the development of SAM in food insecure popu lations. Several approaches for prevention and treatment are in use. Although the provision of comple mentary and supplementary foods could be considered in targeted food distribution pro grammes, other ways to stimulate access and purchasing power can be conceived. Where markets are fragmented or food access is constrained, appropriate food supple ments might be considered as in-kind transfers. WHO recommends inpatient treatment for children with complicated SAM, with stabilisation and appro priate treatment of infections, fluid management, and dietary therapy and also supports community-based care for uncomplicated SAM.89 Although facility-based treatment of SAM remains important, community manage ment of SAM continues to grow rapidly globally. This shift in treatment norms from centralised, inpatient care towards community-based models allows more affected children to be reached and is cost effective. Up to an estimated 15% of cases of SAM will need initial facilitybased care, whereas the rest can receive only communitybased treatment.90

Preventive zinc supplementation in children


Preventive zinc supplementation in populations at risk of zinc deficiency reduces the risk of morbidity from child hood diarrhoea and acute lower respiratory infec tions and might increase linear growth and weight gain in infants and young children.60,76 A review by Yakoob and colleagues61 assessed 18 studies from developing coun tries and showed that preventive zinc supple mentation reduced the incidence of diarrhoea by 13% and pneumonia by 19%, with a non-significant 9% reduction in all-cause mortality (table 3). However, subgroup analysis showed that there was a significant 18% reduction in all-cause mortality in children aged 1259 months. A daily dose of 10 mg zinc per day over 24 weeks in children younger than 5 years could lead to an estimated net gain of 037 cm (SD 025) in height in
48

Facility-based management of SAM according to the WHO protocol


A scientific literature review by Schofield and Ashworth91 showed that between the 1950s and 1990s, case fatality rates were typically 2030% in children with SAM treated in hospitals or rehabilitation units, and rates were higher (5060%) for oedematous malnutrition. A previous review4 of existing studies had estimated that following the WHO protocol, as opposed to standard care, would lead to a 55% reduction in deaths (RR 045, 95% CI 032062; random effects). In view of the limitations of analysis and variable quality of studies in the previous review, we updated the review to assess the effect of the WHO protocol or adaptations thereof on recovery and case fatality of children with SAM.
www.thelancet.com

Series

Case fatality rates ranged from 34% to 35%. The highest case fatality rate stemmed from a cohort of children with HIV infection.92,93 Only two studies provided information on recovery rates, which were 797% and 833%.94,95 In summary, the WHO protocol is substantiated through much evidence, based both on research and expert opinion. However, a clear need exists for continued work to improve staff training and quality96 to achieve high rates of survival across various resource-constrained settings.
Settings WASH interventions Overview of three systematic reviews77 Developing countries Estimates

Community-based management of SAM


The products used to deliver nutrients for management of SAM and MAM, and the approaches used to target and deliver these products, evolved rapidly during the past decade. Innovations include new formulations and packaging and a shift from institutional to communitybased management. We reviewed interventions to treat SAM in community settings, and were largely able to pool studies comparing

Significant effects: reduced risk of diarrhoea with hand washing with soap (RR 052, 95% CI 034065), with improved water quality, and with excreta disposal Significant effects: a recent World Bank report78 based on analysis of trends in DHS data suggests that open defecation explained 54% of international variation in child height by contrast with GDP, which only explained 29%. A 20 percentage point reduction in open defecation was associated with a 01 SD increase in child height A Cochrane review of the effect of WASH interventions on nutrition outcomes is underway87 Non-significant effects: one dose of anthelminthic in second trimester of pregnancy had a non-significant effect on maternal anaemia, low birthweight, preterm births, and perinatal mortality Non-significant effects: one-dose deworming had a non-significant effect on haemoglobin and weight gain. For multiple doses at 1 year follow up, there was a non-significant effect on weight, haemoglobin, cognition, and school attendance Treatment after confirmed infection Significant effects: one-dose of deworming drugs increased weight (058 kg, 95% CI 040076) and haemoglobin (037 g/dL, 95% CI 01064). Evidence on cognition was inconclusive These analyses are corroborated by the large-scale DEVTA trial88 of regular deworming and VAS over 5 years, which also did not show any benefits on weight gain or mortality Significant effects: in acute diarrhoea, lactose-free diets, when compared with lactose-containing diets, significantly reduced incidence of diarrhoea (SMD 036, 95% CI 062 to 010) and treatment failure (RR 053, 95% CI 040070) Non-significant effects: weight gain Significant effects: reduced all-cause mortality reduced by 46% (95% CI 1268), diarrhoea-related admissions to hospital by 23% (95% CI 1531) Non-significant effects: diarrhoea-specific mortality, diarrhoea-prevalence Zinc reduced duration of acute diarrhoea by 050 days and persistent diarrhoea by 068 days Significant effects: Anti-malarials to prevent malaria in all pregnant women reduced antenatal parasitemia (RR 053, 95% CI 033086), increased birthweight (MD 1267 g, 95% CI 886416475), reduced low birthweight by 43% (RR 057, 95% CI 046072) and severe antenatal anaemia 38% (RR 062, 95% CI 050078) Non-significant effects: perinatal deaths Significant effects: ITNs in pregnancy reduced low birthweight (RR 077, 95% CI 061098) and reduced fetal loss (first to fourth pregnancy; RR 067, 95% CI 047097) Non-significant effects: anaemia and clinical malaria Significant effects: Reduced clinical malaria episodes (RR 026; 95% CI 017038), reduced severe malaria episodes (RR 027, 95% CI 010076). IPTc also reduced risk of moderately severe anaemia (RR 071, 95% CI 052098) Non-significant effects: all-cause mortality Significant effects: ITNs improved packed cell volume of children by 17 absolute packed cell volume percent. When the control group used untreated nets, the difference was 04 absolute packed cell volume percent. ITNs and IRS reduced malaria-attributable mortality in children (159 months) by 55% (95% CI 4961) in Plasmodium falciparum settings

DHS data from 65 countries78 Developing countries

Maternal deworming Systematic review of five RCTs79 Systematic review of 34 RCTs80 Developing countries Developing countries

Deworming in children (for soil-transmitted intestinal worms)

Feeding practices in diarrhoea Review of 29 RCTs81 Developing countries

Zinc therapy for diarrhoea Systematic review of 13 studies82 Mostly Asia

IPTp/ITN for malaria in pregnancy Systematic review of 16 RCTs83 Mostly Africa

Systematic review of six RCTs84

Developing countries

Malaria prophylaxis in children Systematic review of seven RCTs85 Developing countries of West Africa Developing countries in Africa

Systematic review of 22 RCTs86

WASH=water, sanitation, and hygiene. RCT=randomised controlled trial. DHS=Demographic and Health Survey. GDP=gross domestic product. RR=relative risk. MD=mean difference. SMD=standard mean difference. WAZ=weight-for-age Z score. HAZ=height-for-age Z score. DEVTA=de-worming and enhanced vitamin A. IPTp=intermittent preventive treatment of malaria in pregnancy. IPTc=IPT in children. ITN=insecticide-treated bednets. IRS=indoor residual spraying.

Table 4: Review of evidence for disease prevention and management

www.thelancet.com

49

Series

ready-to-use therapeutic foods (RUTF) with standard care, as opposed to rigorous evaluation of effectiveness of the approach in programme settings.97 We identified no significant differences in mortality; however, children who received RUTF had faster rates of weight gain and had 51% greater likelihood to recover (defined as attaining WHZ 2) than did those receiving standard care. Notably, a new randomised controlled trial98 compared standard RUTF with RUTF and additional 7 day course of antibiotics, either amoxicillin or cefdinir, in children with uncomplicated SAM. This trial showed that the children receiving an antibiotic had a lower mortality rate, faster recovery rate, and higher weight gain com pared with children receiving placebo. Although further research on this topic is needed, especially in children with HIV infection, this study shows that effective community management of SAM might require an approach that goes beyond merely the choice of specially formulated foods to the entire package of care. Substantial programmatic evidence supports use of RUTF for community-based treatment,99 which has sub stantially changed the approach to treatment of SAM. Yet because of the nature of the evidence, establishing effect estimates for the overall approach to community manage ment has proved challenging. Available evidence shows some positive effects with the use of RUTF compared with standard care for the treatment of SAM in community settings, yet the differences were for the most part small and several outcomes had substantial hetero geneity. An emphasis not only on the choice of commodities, but also on the quality of programme design and implementation is crucial to improvement of outcomes for children with SAM, as is research to fill information gaps, such as opti mum treatment methods and approaches for treat ment of breastfed infants younger than 6 months.

adulthood.102107 All studies suggested a small protective effect of breastfeeding on obesity later in life, although the magnitude of the effect varied between reviews and the strength of the affect of confounding was unclear. The largest prospective follow up study in healthy term infants in Belarus showed that improving the duration and exclusivity of breastfeeding did not prevent over weight or obesity in children, nor did it affect insulinlike growth factor I concentrations at age 115 years.108 These findings suggest that despite the myriad advantages of breastfeeding, population strategies to increase the duration and exclusivity of breastfeeding are unlikely to curb the obesity epidemic. A Cochrane review107 examined the effects of obesity prevention interventions delivered for more than 12 weeks on changes in BMI and BMI Z scores in children and suggested a significant beneficial effect across age groups with a SMD of 015 kg/m (95% CI 021 to 009). The subgroup analysis showed sig nificant effects for children aged 612 years with non-significant effects in younger children and adoles cents. Interventions that combined physical activity and diet were more effective than either delivered alone. Findings suggested that short-term inter ventions (<12 months duration) were more effective than were those delivered over a longer duration (SMD 017, 95% CI 025 to 009 and SMD 012, 95% CI 021 to 003, respectively); however, there was sub stan tial heterogeneity in all pooled estimates. Another review109 of interventions to treat obesity in children showed that combined behavioural and life style inter ventions or self-help could benefit over weight children and adolescents. Overall the evidence of effectiveness of all obesity prevention and therapeutic interventions is weak, underscoring the need for high-quality research in this discipline.

Interventions for prevention and management of obesity


Obesity is increasing in many populations and is one of the most important challenges of the 21st century. Obese women are at an increased risk of adverse pregnancy outcomes. A Cochrane review100 assessed the effectiveness of interventions (eating, exercise, behaviour modification, or counselling) that reduce weight in obese pregnant women and identified no evaluable trials. Some studies assessed the effect of diet, exercise, or both for weight reduction in women after childbirth, and showed that women who exercised did not lose significantly more weight, but women who took part in a diet (MD 170 kg, 95% CI 208 to 132), or diet plus exercise programme (289 kg; 483 to 095), did so. These interventions did not seem to adversely effect breastfeeding performance in any setting.101 We identified six reviews that examined breastfeeding in infancy and its association with obesity prevalence or average body-mass index (BMI) in childhood or
50

Delivery platforms and strategies for implementation of nutrition-specific interventions


Delivery strategies are crucial to achieve coverage with nutrition-specific interventions and to reach popu lations in need. A range of channels can provide oppor tunities for scaling up and reaching large segments of the population.

Fortification of staple foods and specific foods


A detailed discussion of fortification strategies is beyond the scope of this review. As supported by the Copenhagen consensus,110 fortification is one the most cost-effective strategies to reach populations at large. Further discussion of fortification as a means for delivery of key micronutrients is provided in panel 3111118 and the accompanying report by Stuart Gillespie and colleagues.119

Cash transfer programmes


Financial incentives are widely used as policy strategies to ameliorate poverty, reduce financial barriers, and improve
www.thelancet.com

Series

Panel 3: Effect of fortification strategies Food fortification is safe and cost effective in the prevention of micronutrient deficiencies and has been widely practised in developed countries for more than a century.111 Foods can be fortified at three levels: mass or universal, targeted, and household. Mass or universal fortificationideally legislated to be mandatory for industrieshas the potential to produce foods and food products that are widely consumed by the general population (eg, salt iodisation and flour fortification with iron and folate). Mass fortification is by far the most cost-effective nutrition intervention, particularly when produced by medium-to-large scale industries.111 Targeted fortification (eg, nutrient-fortified complementary foods for children aged 624 months) is important for subgroups of nutritionally vulnerable populations and populations in emergency situations whose nutrient intake is insufficient through available diets. Targeted fortification is also effective in resource poor settings where family foods do not include animal-source foods that are typically necessary to meet nutrient requirements of young children. Home fortification involves addition of nutrients directly to food consumed by women or children, or both, in the form of micronutrient powders or small quantities of food-based fortified lipid spreads (eg, lipid-based nutrient supplement). Such direct addition of micronutrients to foods is different from foods fortified in the preparatory processes, has the advantage that it does not require changing dietary practices, and has little effect on the taste of food. However, addition of micronutrient powders to prepared foods has characteristics akin to supplementation as opposed to foods fortified at source. Biofortification of food crops (fortification of food at source) is an alternative to more common fortification interventions and is rapidly advancing in technology with much success, particularly with regard to increasing iron, provitamin A, zinc, and folate contents in staple foods.112 Despite many limitations to establishing causality during assessment of food fortification programmes, several studies have reported outcomes. Fortification for children shows significant benefits on serum micronutrient concentrations, which could indirectly be used to work out the population-level effect. A meta-analysis of multiple micronutrient fortification in children shows an increase in haemoglobin concentrations by 087 g/dL (95% CI 057116) and reduced risk of anaemia by 57% (relative risk [RR] 043; 95% CI 026071). The mean ferritin increase with fortification was 113 g/L (95% CI 33192) compared with control groups. Fortification also increased vitamin A serum concentrations compared with control groups (four studies, mean retinol increase 37 g/dL, 95% CI 1361).113 A meta-analysis of 60 trials showed that iron fortification of foods resulted in 41% reduction in the risks of anaemia (RR 059, 95% CI 048071, p< 0001) and a 52% reduction in iron deficiency (048, 038062, p<0001).114 Other studies have also shown that use of vitamin D fortified bread increased serum 25-hydroxyvitamin D concentration as effectively as the cholecalciferol supplement in women.115 Zinc fortification has also shown significantly higher zinc concentrations in serum and erythrocytes and lower serum copper concentrations compared with a placebo group in preterm infants.116 Fortification has the greatest potential to improve the nutritional status of a population when implemented within a comprehensive nutrition strategy. Key issues to ensure a sustainable programme include: identification of the right food (accounting for bioavailability, interaction with food, availability, acceptability, and cost) and target population, ensuring quality of product, and consumption of sufficient quantity of the fortified foods.117 To accomplish these aims, there needs to be demand that is sustained through behaviour change communication at the consumer level and ready access to a sufficient supply of products that maintain standards set through legislative process from production to point-of-consumption. Government monitoring of compliance to standards and public-private partnerships are essential to ensure a competitive market for fortified products.118 Fortification seems to be a potentially effective strategy but evidence of benefits on morbidity and functional outcomes from large-scale programmes in developing countries is scarce.

population health. We reviewed relevant studies reporting the effect of financial incentives on coverage of health and nutrition interventions and behaviours targeting children younger than 5 years.120 The affect of financial incentive programmes on five categories of interventions (breast feeding practices, immunisation coverage, diarrhoea man age ment, healthcare use, and other preventive strategies) was assessed. The review concluded that financial incen tives have the potential to promote increased coverage of several important child health interventions, but the quality of evidence available was low. The more pronounced effects seemed to be achieved by programmes that directly removed user fees for access to health services.120 Some indication of effect was also noted for programmes that conditioned finan cial incentives on participation in health education and attendance to health-care visits. Further
www.thelancet.com

information on the benefit of such programmes for health and nutrition outcomes is provided in the accompanying report by Stuart Gillespie and colleagues.119

Community-based platforms for nutrition education and promotion


Community-based interventions to improve maternal, newborn, and child health are now widely recognised as important strategies to deliver key maternal and child survival interventions121 and have been shown to reduce inequities in childhood pneumonia and diarrhoea deaths.122 These interventions are delivered by health-care personnel or lay individuals, and implemented locally in homes, villages, or any defined community group. A full spectrum of promotive, preventive, and curative inter ventions can be delivered via community platforms,
51

Series

including provision of basic antenatal, natal, and postnatal care; preventive essential newborn care; breastfeeding counselling; man age ment and referral of sick neonates; development of skills in behaviour change communication; and com munity mobilisation strategies to promote birth and new born care prepared ness. For example, a review123 of community-based ventions can packages of care suggested that these inter improve rates of facility births by 28% (RR 128, 95% CI 104159) and result in a doubling of the rate of initiation of breastfeeding within 1 h (RR 225, 95% CI 170297). Lewin and colleagues124 reviewed 82 studies with lay health workers and showed moderate quality evidence of effect on initiation of breastfeeding (RR 136, 95% CI 114161), any breastfeeding (124, 110139), and exclusive breastfeeding (278, 174444) when compared with usual care. Although much of the evidence from large-scale programmes using community health workers is of poor quality, process indicators and assessments do suggest that community health workers are able to implement many of these projects at scale, and have substantial potential to improve the uptake of child health and nutrition outcomes among difficult to reach populations.125 It is important to underscore the
Panel 4: Nutrition in emergencies Irrespective of the underlying cause, humanitarian emergencies are often characterised by high and rising rates of severe acute malnutrition (SAM), moderate acute malnutrition (MAM), and micronutrient deficiencies in children (and sometimes adults). The foremost intent of nutrition-specific interventions in such situations is to prevent mortality, and involves management of wasting and resolution of specific nutrient deficiencies, and ensuring adequate food consumption. The humanitarian community largely agrees that emergency nutrition interventions have improved in the past 1015 years in terms of coverage, scale of operations, reporting standards, and effectiveness (assessed by Sphere133 and other standards of practice). Until the early 2000s, nutrition programming in emergencies was dominated by facility-based therapeutic care, targeted or blanket supplementary feeding, and provision of micronutrient supplements.134136 More recently, the focus has widened, with attention being given to both short-term and longer-term concerns, and to a choice of actions from a more comprehensive range of interventions.137 The options for effective management for both SAM and MAM in emergencies have improved in the past 1015 years as products used have been improved and coverage has increased through community-based treatment. Potential alternatives to the use of food to address seasonal or emergency-driven peaks in wasting are being explored, including combinations of food plus cash, cash alone, or vouchers; however, cost-effectiveness studies of various strategies are scarce. There is evidence that in contexts in which markets have not been seriously disrupted, appropriate foods are readily accessible, and rates of

crucial importance of community engagement and buyin to ensure effective community outreach programmes, behaviour change, and access.

Integrated management of childhood illnesses


WHO, in collaboration with UNICEF and other agencies, developed the Integrated Management of Childhood Illness (IMCI) strategy in the 1990s.126 IMCI includes both curative and preventive interventions targeted at improve ment of health practices at health facilities and at home. The strategy includes three components: improve ments in case management; improvements in health systems; and improvements in family and com munity practices. Assessments of IMCI in Uganda, Tanzania, Bangladesh, Brazil, Peru, South Africa, China, Armenia, Nigeria, and Morocco have shown various benefits in health service quality, mortality reduction, and healthcare cost savings.127 In Tanzania, implemen tation of IMCI was associated with significant improve ments in equity differentials for six child health indicators, with the largest improvements noted for stunting in children between 24 and 59 months of age.128 Much the same findings were reported from Bangladesh, where imple men tation of IMCI was asso ciated with a significant

undernutrition are not dangerously high, alternatives or complements to food-based rations might be viable and potentially cost-effective.138,139 Concern has also grown about the potential trade-offs between long-term versus short-term objectives of emergency nutrition interventions. Although life-saving actions are justifiably prioritised over the prevention of chronic diseases, food assistance programmes suitable for acute emergencies might be less appropriate for protracted situations.140 This has important implications when thinking through seasonal blanket distribution of ready-to-use foods to prevent a worsening of levels of acute malnutrition. As a result of the difficulty of generating experimental data specific to programming in emergencies,141 the discipline has evolved relying less on randomised controlled trials and more on the sharing of lessons learned, which are used to inform technical or operational guidelines disseminated by WHO and UN bodies.142144 Although practice must still be improved in many areas, and outcomes better documented, it remains crucially important to secure appropriate resources to support nutrition actions in this most challenging of disciplines and to assess outcomes for future learning. The nutritional status of individuals assessed and treated in emergency contexts overlaps substantially with non-emergency settings. Although high-quality programmatic research can and must help improve the design and outcome of effective emergency nutrition interventions, these interventions should be seen as entry points that support, rather than supplant, longer-term actions seeking to address underlying causes of poor nutrition.

52

www.thelancet.com

Series

increase in exclusive breastfeeding and com paratively faster reduction in the prevalence of stunting in children aged 2459 months.129

School-based delivery platforms


Many countries have school feeding programmes targeting children who are older than 5 years. The main purpose of such programmes is to provide incentives for school enrolment and evidence of nutrition benefits is scarce. A Cochrane review130 of 18 relevant studies of the effectiveness of school feeding programmes in improving physical and psychosocial health for disadvantaged school pupils reported an increase in school attendance by 46 days annually and weight gains averaging 039 kg (95% CI 011067) over 11 months and 071 kg (048095) over 19 months. The results were inconclusive for height gain, so there must be caution that these programmes do not lead to obesity. A detailed discussion of school feeding programmes is provided in the accompanying report by Stuart Gillespie and colleagues.119 Notwithstanding the scarce evidence, schools offer an enormous opportunity for promotion of health and nutrition for older children and adolescents and could have an important role in future.

survival interventions. There are few robust assessments or reported experiences with child health days, which commonly include delivery of vitamin A supplements, immunisations, insecticide-treated nets, and deworming drugs. Available evidence suggests that these days can achieve greater coverage than stand-alone campaigns in previously low-coverage countries.131 A descriptive review132 of scale-up of child health days from 1999 to 2009 suggests that these days were more effective than stand-alone campaigns, provided that the number of interventions did not exceed four. The overall equity effect of these approaches are uncertain and further studies are needed to establish how best to integrate this approach within routine health-care services.

Delivery of nutrition interventions in humanitarian emergency settings


Delivery strategies for nutrition interventions in human i tarian emergencies necessitate a different approach to what might be deemed optimum in stable circumstances. In view of variability in the characteristics of emergencies and protracted population displace ment, humanitarian emergencies might closely mirror situ ations of endemic malnutrition in food insecure settings. Hence prevention and health promotion strategies, such as breastfeeding and complementary feeding education and support, should also become essential parts of the packages of interventions in emergency contexts (panel 4133144).

Child health days


Child health days have been introduced in weak health systems to rapidly enhance coverage of essential child
Panel 5: Evidence for emerging interventions Household air pollution Household air pollution (HAP) from solid fuels used in simple stoves for cooking and heating, is recognised as a risk factor for several health outcomes with important consequences for child survival, including pneumonia,146 low birthweight, and stillbirths.147 A review of observational studies148 shows significant risk reduction estimates for HAP for low birthweight (29%), stillbirth (34%), stunting (21%), and all-cause mortality (27%). Reduction of exposure to HAP could substantially reduce risk of several important outcomes for child survival. One randomised controlled trial in rural Guatemala,149 with an improved stove intervention, reduced average exposure to indoor air pollution by 50% and resulted in a reduction in physician-diagnosed pneumonia (relative risk [RR] 084, 95% CI 063113) although this difference was not statistically significant. However, this finding was supported by the results of an exposure-response analysis which showed a statistically significant reduction in the same outcome (082, 070098). This intervention also resulted in a reduction in low birthweight (074, 95% CI 033166), with babies weighing 89 g more (95% CI 27 to 204) than those in the control group.150 A range of interventions, including both clean fuels and improved solid fuel stoves are available, but substantial challenges remain in achieving sustained use of low-cost low-emission technologies at scale in low-income households.

Maternal vitamin D supplementation Vitamin D is an essential requirement of the body at any age. Vitamin D can be acquired through three main channels: through the skin via exposure to sunlight, from the diet, and from supplements or fortified foods. However, natural low-cost sources of dietary sources of vitamin D are very scarce. A systematic review151 assessing the association of vitamin D status in pregnancy, suggests that women with circulating 25-hydroxyvitamin D (25[OH]D) concentrations of less than 50 nmol/L in pregnancy have an increased risk of preeclampsia (odds ratio [OR] 209, 95% CI 150290), gestational diabetes mellitus (138, 112170), preterm birth (158, 108231) and small-for-gestation age ([SGA] 152, 108215). A long-term cohort study152 did not find any association of low maternal vitamin D concentrations with bone mineral content in late childhood. Similarly, a Cochrane review153 assessed the effectiveness of vitamin D supplementation in pregnancy and revealed little evidence of benefits on functional pregnancy outcomes, although significant increase in serum vitamin D concentrations at term were noted and borderline reduction in low birthweight was reported in three trials (RR 048, 95% CI 023101). The number of high-quality trials with maternal vitamin D supplementation is too small to draw conclusions on its usefulness and safety. (Continues on next page)

www.thelancet.com

53

Series

(Continued from previous page) Maternal zinc supplementation A Cochrane review154 suggests that zinc supplementation in pregnancy results in a 14% reduction in preterm birth (RR 086, 95% CI 076097). This decrease was not accompanied by a similar reduction in stillbirths, neonatal death, SGA, or low birthweight. No subgroup differences were identified in women with low versus normal zinc nutrition levels or in women who complied with their treatment versus those who did not. We conclude that there is presently insufficient evidence for a beneficial role of isolated zinc supplementation in pregnancy. Omega-3 fatty acid supplementation Several reviews155162 have been done to assess the effectiveness of maternal supplementation with omega-3 fatty acids during pregnancy and its effects on various outcomes including nutritional, morbidity, mortality, cognitive, and neurodevelopmental measures. Findings from these reviews, consisting of studies done in developed countries and of variable quality, suggest that marine omega-3 fatty acids administered in pregnancy reduce the rate of preterm birth and increase birthweight. However, a Cochrane review155 suggests that there is not enough evidence to support the routine use of marine oil supplements or other prostaglandin precursors during pregnancy to reduce the risk of pre-eclampsia, preterm birth, low birthweight, or SGA. A review163 of the intake of omega-3 and omega-6 fatty acids in low-income countries showed that the total omega-3 fatty acid supply was below the recommended intake range for infants and young children, and below the minimum recommended level for pregnant and lactating women, in the nine countries with the lowest gross domestic product. The review noted that supply of omega-3 fatty acids could be increased by using vegetable oils with higher alpha-linolenic acid and by increasing fish production through fish farming. Another review164 on the effect of fatty acid status on immune function of children in low-income countries suggested that fatty acid interventions could yield immune benefits in children in poor settings, especially in non-breastfed children and in relation to inflammatory disorders, such as persistent enteropathy, although more trials are needed for a conclusive association. Antenatal psychosocial assessment and mental health support Stable maternal mental health during pregnancy is crucial for the development of the early motherchild relationship and for health. Although there is ample evidence of the link between maternal mental health and child health and growth,165 there is insufficient evidence to support routine psycho-social screening for all pregnant women.166 There is promising evidence that cognitive-behaviour therapy-based interventions provided by community health workers to pregnant women, can effectively reduce depression at 3 months post-partum (adjusted OR 022, 95% CI 014036) and at 1-year follow-up (023, 015036).167 However, there was no effect on weight gain or linear growth in infancy. There is a need for further robust trials of maternal mental health interventions with longer term follow up.

Role of massage for promoting growth in preterm Infants Preterm infants have been noted to benefit from massage therapy and the suggested mechanisms include increased vagal activity and gastric motility, which leads to increased concentrations of insulin and Insulin-like growth factor 1.168 A Cochrane review169 of the effect of massage in preterm infants showed that massage increased daily weight gain by 5 g, reduced the length of hospital stay by 45 days, and had a slight effect on development and weight gain at 46 months, although the evidence was of weak quality. A more recent review170 of the effects of massage therapy for preterm infants showed that 510 days of moderate pressure massage, typically 15 min three-times daily, resulted in improved weight gain (mean for studies 2848%) and bone density, and reduced length of hospital stay. Related evidence from studies of emollient therapy in preterm infants from the developing world suggest potential synergistic benefits of skin barrier protection, thermoregulation, and light massage. Vitamin D supplementation in children In view of the widespread deficiency of vitamin D and associated health consequences and rickets, preventive vitamin D supplementation to high-risk populations, including infants and toddlers, might be a useful strategy. A Cochrane review of vitamin D supplementation in children in at-risk populations is underway, and an existing review171 of postnatal supplementation shows relatively few studies assessing effects on bone density, growth, and other functional outcomes. Zinc supplementation for treatment of newborn infections and childhood pneumonia A Cochrane review172 suggests that zinc supplementation in addition to antibiotics in children with severe and non-severe pneumonia did not have a significant effect on clinical recovery or duration of hospital stay. Other recent studies show mixed effects across a range of severity of disease,173176 showing the need for larger well-powered studies for the treatment of severe pneumonia with zinc in populations at-risk of deficiency. Two trials177,178 of adjunctive zinc supplementation in presumed serious infections in neonates and young infants show disparate findings, underscoring the need for further well-designed and adequately powered studies of zinc as an adjunct to the treatment of serious infections in infancy. Lipid-based nutrient supplementation Lipid-based nutrient supplements (LNS, in the form of vegetable oil, peanut butter, milk powder, sugar, vitamins, and minerals) are used in small quantities (20 g) to meet micronutrient requirements in children, in combination with a normal diet. Randomised controlled trials in Malawi179,180 and Ghana181,182 have shown significant benefits on iron status and linear growth. Further evidence of benefits and absence of adverse effects are needed to assess the feasibility of use of LNS in programme settings and randomised controlled trials are underwaythree in Africa and one in Asiawhich should provide more information.

54

www.thelancet.com

Series

Panel 6: Overview of the Lives Saved Tool (LiST) To model the effect of scaling up the ten proven nutritionspecific interventions on the health of children we used the Lives Saved Tool (LiST). This model has been developed under the auspices of the Child Health Epidemiology Reference Group (CHERG) to allow users to estimate the effect of scaling up interventions on maternal and child health. The present version of the model is based on previous modelling exercises, including The Lancets 2008 Maternal and Child Undernutrition series.4,122,145,184 A more detailed description of the LiST model is provided in appendix pp 37 and at the LiST website. The LiST has been characterised as a linear, mathematical model that is deterministic.185 It describes fixed associations between inputs and outputs that will produce the same outputs each time the model is run. In LiST the primary inputs are coverage of interventions and the outputs are changes in population level of risk factors (such as wasting or stunting rates, or birth outcomes such as prematurity or size at birth) and cause-specific mortality (neonatal, mortality in children aged 159 months, maternal mortality, and stillbirths). The association between an input (change in intervention coverage) with one or more outputs is specified in terms of the effectiveness of the intervention for reduction of the probability of that outcome. The outcome can be cause-specific mortality or a risk factor. The overarching assumption in LiST is that mortality rates and cause of death structure will not change except in response to changes in coverage of interventions. The model assumes that changes in distal variables, such as increase in income per person or mothers education, will affect mortality by increasing coverage of interventions or reducing risk factors. Figure 2 shows the linkages in LiST between risk factors, interventions, and mortality. For example, the input of multiple micronutrients consumed by pregnant women has an effect on birth outcomes. It directly affects the probability of a child being born small-for-gestational-age (SGA). A child born SGA will in turn have an increased risk of dying during the neonatal period and those who survive through the first month of life then have an increased risk of being stunted. Stunted children have higher risks of mortality from 1 to 59 months of age. A second example is promotion of breastfeeding. Scaling up breastfeeding promotion will affect breastfeeding practice, shifting the distribution of mothers who exclusively, predominately, or partly breastfeed their child or do not breastfeed. Within the model, this intervention is examined separately for the first month of life and for the period of 15 months. Within each of the two time periods there is a different relative risk of dying of pneumonia and diarrhoea associated with each breastfeeding practice and an effect on diarrhoea incidence. So within the model, breastfeeding promotion has an effect on breastfeeding practices, which in turn has a direct effect on mortality in the neonatal and 123 month period. Additionally, there is an effect on diarrhoea incidence, which then has an effect on stunting rates and mortality. The assumptions of the effects of the ten nutrition interventions we used in the modelling are shown in appendix pp 1316. For each of the interventions we have also shown the 95% CIs around the estimates, which were used in the sensitivity analyses. The source and methods used to develop these assumptions and others used in the model are described in a series of reports.186188

Emerging interventions that need further evidence


We also reviewed interventions that are not currently recommended but that have potential and future prospects for inclusion in regular programmes. These interventions, which have possible effects on nutritional outcomes in women and children, include strategies to reduce house hold air pollution, maternal vitamin D supple mentation, maternal zinc supplementation, omega 3 fatty acids supple mentation in pregnancy, antenatal psycho social assessment and cognitive behaviour therapy for depres sion, emollient and massage therapy for preterm infants, vitamin D supplementation in children, zinc therapy for pneumonia, and lipid-based nutrient supple ments. Some of the existing evidence around these interventions is summarised in panel 5.146182

Folic acid fortication

MMN, BEP

Breastfeeding promotion

Complementary feeding, education, and supplementation Breastfeeding practices

Vitamin A supplementation

Zinc supplementation

Neonatal mortality by cause Birth risk dened by maternal age, parity, and spacing Birth outcomes (SGA and preterm)

Diarrhoea incidence

Stunting Mortality by cause 159 months Wasting

Risk factors Interventions Mortality

Management of SAM

Management of MAM

Figure 2: Linkages between risk factors, interventions, and mortality in LiST LiST=Lives Saved Tool. MMN=multiple micronutrients. BEP=balanced energy protein. SGA=small-for-gestational-age. SAM=severe acute malnutrition. MAM=moderate AM.

Modelling the effect of scaling up coverage of nutrition interventions in countries with the highest burden
We used the Lives Saved Tool (LiST) to model the potential effect on child health and mortality in 2012 of
www.thelancet.com

scaling up a set of ten nutrition-specific interven tions that could affect stunting and severe wasting183 (panel 6,4,122,145,184188 figure 2). For modelling, we selected 34 countries with more than 90% of the burden of
55

Series

Afghanistan Egypt Chad Burkina Faso Mali Niger Guatemala C te dIvoire Ghana Nigeria Cameroon Congo Angola Zambia Sudan Yemen Ethiopia Kenya Tanzania Madagascar Malawi Mozambique South Africa High burden countries Other countries Uganda Rwanda Indonesia Iraq Pakistan Nepal India Philippines Vietnam Bangladesh Myanmar

Figure 3: Countries with the highest burden of malnutrition These 34 countries account for 90% of the global burden of malnutrition.
1 400 000 1 200 000 1 000 000 Number of deaths 800 000 600 000 400 000 200 000 0 Baseline Nutrition interventions scaled up

Diarrhoea deaths

Pneumonia deaths

Measles Neonatal deaths in diarrhoea children <5 years deaths

Neonatal pneumonia deaths

Neonatal congenital anomalies

Neonatal asphyxia

Cause-specic deaths in children <5 years and neonates

Figure 4: Effect of scale up of interventions on cause-specific deaths Error bars are ranges.

For more on the Lives Saved Tool see http://www.jhsph.edu/ iip/LiST

stunting (figure 3; appendix pp 812) and took 2011 as the base year. The present coverage level for each intervention was taken from the latest available esti mates from large-scale surveys and effective ness of interventions (see LiST for details). We modelled the effect of scaling up the follow ing ten nutrition inter ventions: periconceptional folic acid supplementation or fortifi ca tion, maternal balanced energy protein supplementation, maternal calcium supplementation, multiple micronutrient supplementation in pregnancy, promotion of breastfeeding, appropriate complementary feeding, vitamin A and preventive zinc supplementation in chil dren 659 months of age, management of SAM, and management of MAM, from their present level of coverage to 90% (or retention of present coverage when

higher than 90%). Appendix pp 1316 list the estimates of effect considered for each intervention. The con version of intervention effects of preventive zinc supplementation and complementary feeding strategies from linear growth to stunting effects in LiST is detailed in appendix pp 1722. We assessed the effect of this scale up scenario on mortality in children younger than 5 years, rates of breastfeeding, stunting, and wasting. Our model suggested that if these ten nutrition interventions were scaled up to 90% coverage, mortality in children younger than 5 years could be reduced by 15% (range 919), with a 35% (1943) reduction in diarrhoea-specific mortality, a 29% (1637) reduction in pneumonia-specific mortality, and a 39% (2347) reduction in measles-specific mortality (figure 4). The analysis also showed fewer deaths attributable to congenital anomalies and birth asphyxia related to periconceptual folic acid use and a reduction in SGA (figure 4; appendix pp 2324). This scale up had a little effect on maternal mortality (data not shown). Scaling up of all ten interventions to 90% coverage was also associated with a mean 203% (range 111289) reduc tion in stunting and a 614% (35772) reduction in severe wasting. The maximum effect for severe wasting was noted in children in the 1223 months age group (appendix p 25). The analysis suggested that the interventions with the largest potential affect on mortality in children younger than 5 years are management of SAM, preventive zinc supplementation, and promotion of breastfeeding (figure 5). Analysis of community support strategies for breastfeeding suggested that achieving 90% coverage of breastfeeding promotion could increase exclusive breast feeding by 15% (722) in children younger than 1 month and by 20% (1326) in children aged 15 months.
www.thelancet.com

56

Series

Management of SAM Preventive zinc supplementation Promotion of breastfeeding Appropriate complementary feeding Interventions Management of MAM Periconceptual folic acid supplementation or fortication Maternal balanced energy protein supplementation Maternal multiple micronutrient supplementation Vitamin A supplementation Maternal calcium supplementation 0 50 000 100 000 150 000 200 000 250 000 300 000 Number of deaths of children <5 years averted 350 000 400 000

Figure 5: Effect of scale up of interventions on deaths in children younger than 5 years Error bars are ranges. SAM=severe actute malnutriton. MAM=moderate AM.

Implementation of nutrition-specific packages of care


We also assessed the potential effect of nutrition-specific packages of care by scaling up these interventions to 90% coverage. Four packages were assessed for effect on child survival: optimum maternal nutrition during preg nancy (maternal multiple micronutrients, use of iodised salt, calcium, and balanced energy protein supple men tation), an infant and young child nutrition package (breast feeding promotion and appropriate comple mentary feeding education or provision), micronutrient supplemen tation (preventive zinc and vitamin A supple mentation), and management of acute malnutrition (management of MAM, management of SAM). Analysis of these nutrition-specific packages showed that the most lives could be saved by the therapeutic feeding for severe acute mal nutrition, followed by the infant and young child nutrition package (table 5189).
Optimum maternal nutrition during pregnancy Maternal multiple micronutrient supplements to all Calcium supplementation to mothers at risk of low intake Maternal balanced energy protein supplements as needed Universal salt iodisation Infant and young child feeding

Number of lives saved* 102000 (49000146000)

Cost per lifeyear saved $571 (3981191)

Promotion of early and exclusive breastfeeding for 6 months 221000 (135000293000) and continued breastfeeding for up to 24 months Appropriate complementary feeding education in food secure populations and additional complementary food supplements in food insecure populations Micronutrient supplementation in children at risk Vitamin A supplementation between 6 and 59 months age 145000 Preventive zinc supplements between 12 and 59 months of age (30000216000) Management of acute malnutrition Management of moderate acute malnutrition Management of severe acute malnutrition 435000 (285000482000)

$175 (132286)

$159 (106766)

$125 (119152)

Can these interventions promote equitable access?


To assess the potential benefit of community-based delivery strategies on reaching and engaging poor and marginalised populations, we assessed the effect of community-based promotion and delivery of seven nutrition-specific interventions (multiple micronutrient supple ment ation in pregnancy, pro motion of breast feeding, appropriate complementary feeding, manage ment of SAM, vitamin A supplementaton, preventive zinc supple mentation, and treatment of diarrhoea with zinc) across various wealth quintiles in three target countries Pakistan, Bangladesh, and Ethiopia (figure 6). Baseline data were stratified by wealth quin tiles by reanalysing the most recently available Demographic Health Survey for each country. Since no recent estimates existed for cause-specific mortality across wealth quintiles for Bangladesh and Ethiopia, we used LiST to recompute the cause of death structure using the procedures described by Amouzou and colleagues.190 For Pakistan, we used the recent national verbal autopsy study191 for the cause of death structure and distribution of deaths
www.thelancet.com

Data are number (95% CI) or cost in 2010 international dollars (95% CI). *Effect of each of package when all four packages are scaled up at once. Cost per life-year saved assumes that a life saved of a child younger than 5 years saves on average 59 life-years, based on WHO data (2011189) that life expectancy at birth on average in low-income countries is 60, and that most deaths of children younger than 5 years occur in the first year of life. To convert to cost per discounted life-year saved multiply these estimates by 59/32 (ie, 184). Intervention has effect on maternal or child morbidity, but no direct effect on lives saved. Cost per life-year saved by management of severe acute malnutrition only, costs for supplementary feeding for moderate acute malnutrition are currently unavailable.

Table 5: Effect of packages of nutrition interventions at 90% coverage

by asset quintiles from a recent analysis.192 As shown in figure 6 and appendix p 26, the effect of this scale up is greatest in the poorest quintiles, suggesting that scal ing up these interventions through community-based approaches would not only reduce the overall burden of childhood mortality but also substantially reduce existing disparities in access and mortality.

Cost analysis
We used a so-called ingredients approach to work out the cost of nutrition interventions, based on the UN One Health Tool,193 which allows for regional variation due to personnel costs. We constructed cost estimates as addons to existing antenatal, postnatal, and standard infant
57

Series

A
Richest Richer Middle Poorer Poorest

Pakistan

Cost Salt iodisation Multiple micronutrient supplementation in pregnancy (includes iron-folate) Calcium supplementation in pregnancy Energy-protein supplementation in pregnancy Vitamin A supplementation in childhood Zinc supplementation in childhood Breastfeeding promotion Complementary feeding education Complementary food supplementation SAM management Total
Data are 2010 international dollars, millions.

$68 $472 $1914 $972 $106 $1182 $653 $269 $1359 $2563 $9559

50

150

200

200

250

B
Richest Richer Middle Poorer Poorest

Bangladesh

Table 6: Total additional annual cost of achieving 90% coverage with nutrition interventions, in 34 countries with more than 90% of the burden

20

40

60

80

100

120

C
Richest Richer Middle Poorer Poorest

Ethiopia

50 100 150 Proportion of deaths in children <5 years averted (%)

200

Figure 6: Equity analysis showing effect of scale up of nutrition interventions on proportion of deaths of children younger than 5 years averted in Pakistan, Bangladesh, and Ethiopia Error bars are ranges.

visits as part of WHOs Expanded Program on Immuni sation, plus five stand-alone nutrition visits between 6 and 35 months of age. The few interventions targeted at children between 36 and 59 months of age were assumed to be delivered opportunistically (at clinic visits, or during outreach visits for younger siblings). The base delivery platform assumed was outreach pro grammes for sub-Saharan Africa, and primary health-care clinics elsewhere (appendix pp 2728 provides details). We compared the unit costs from the ingredients method with actual costs as used in the Scaling up Nutrition (SUN) costing.194 Although the ingredients method allows greater detail than an actual costs method such as SUN for planning purposes, the comparison to actual costs serves as a useful check on the appropriateness of assumptions made. Costs were estimated for ten nutrition interventions (one population-wide, three in
58

pregnancy, and six after birth; intervention definitions and assumptions are provide in appendix pp 2930). We calculated unit costs separately for WHO sub regions (appendix p 31).195 Unit costs were higher in Africa compared with elsewhere, because of higher labour costs and the extra travel time required for delivery using outreach (associated with lower population density in many areas, and also the lower coverage of primary care facilities). The unit costs for interventions were much the same between the ingredients and SUN approaches, allowing for some difference in interventions based on updated recommendations. Our analysis shows that the estimated total additional cost involved to achieve 90% coverage of the population in need in the 34 focus countries with the selected set of ten nutrition interventions is Int$96 billion per annum (table 6). Of this $96 billion, $37 billion (39%) is for micronutrient interventions, $09 billion (9%) for educational interventions, and $26 billion (27%) for SAM management. The amount required for provision of supplementary food for pregnant women and for chil dren aged 623 months in poor households (those with <$125 per person per day) constitutes the remaining $23 billion (24%). When these costs are broken down by region, $34 billion is needed in the 20 countries included from sub-Saharan Africa, $48 billion in the four in south Asia plus Myanmar (Burma), $10 billion for the six in eastern Mediterranean, and $05 billion for the three remaining countries (Vietnam and the Philippines in western Pacific region, plus Guatemala; appendix p 32). The $96 billion estimate for the nutrition interventions is lower than the 2008 SUN estimate of $118 billion.194 The SUN figure included $12 billion for capacitybuilding and monitoring and assessment, which we excluded from the present analysis because we do not have a mechanism to allocate this cost by region or country and category.
www.thelancet.com

Series

There are differences in the details of our results compared with the earlier SUN estimates. Unit costs are similar but not identical. The list of focus countries is likewise similar but not identical (using 2005 data, Turkey, Peru, Cambodia, and Burundi were included in the list of countries with 90% of the worlds stunted children, but not with 2010 data; Rwanda and Chad entered this list with 2010 data). Some interventions are excluded from the new total (deworming, therapeutic zinc for diarrhoea), whereas some new ones are included (calcium supple ments and balanced energy protein supplements in preg nancy). Complementary food supple mentation is targeted only to the 623 month age group in this new analysis. Population in need has changed since the SUN estimates were calculated, with changes in coverage of some inter ventions (notably, management of SAM has begun to scale up).

Discussion
This update of nutrition interventions differs from past exercises in several ways. First, we included a wider range of nutrition-specific interventions and applied more stringent assessment criteria, using the Grades of Recom mendation Assessment, Development and Evalu ation system and Child Health Epidemiology Reference Group criteria for most inverventions.5 Second, in view of emerging evidence of the importance of maternal nutrition, SGA, and early stunting,1 we specifically focused on interventions that might affect prevalence and out comes in SGA births and early stunting. We also reviewed and modelled a range of delivery strategies and platforms and specifically explored the potential of reaching poor and disadvantaged popu lations through community platforms and outreach services, an approach used to assess the effect of interventions to address childhood diarrhoea and pneumonia.122 Finally, the LiST model was substantially updated to include age-specific effects and the interrelationship of new interventions and their effect on maternal and child undernutrition, a much more complex exercise than what was undertaken pre viously.4 To under take this exercise we substantively updated LiST in a way that more accurately captures the role of undernutrition and the effect of proven interventions on maternal and child health. Several limitations should be recognised in considering our findings. A large proportion of the evidence on inter ventions is still derived from efficacy trials as opposed to effectiveness studies and hence variations exist in esti mates of effect size for various interventions. Few robust assessments have been done in programme settings and available data from observational studies do not permit ready assessment of intervention effective ness. There are also very few studies that report morbidity and neuro developmental outcomes. We reviewed the available evidence of effects on neurodevelopmental outcomes from studies of maternal and child nutrition interventions and identified little data with evaluation methodologies for assessment. We are therefore greatly limited in the
www.thelancet.com

inferences that can be drawn on neuro development and long-term outcomes from nutrition interventions. Notwithstanding these limitations, our estimates of the effect of nutrition-specific interventions, though more conservative than previous findings, still suggest great benefits from a core set of interventions delivered ante natally or postnatally. Our assessments of benefits from interventions to reduce SGA are hindered by the limited range of interventions in pregnancy. Even though ante natal care services offer a unique opportunity for maternal screening and interventions, the difficulty of reaching women early enough in pregnancy is a major limitation in ensuring adequate uptake of interventions for a reasonable length of time. In parts of the world with high rates of maternal malnutrition, micronutrient deficiencies, and SGA births, these factors remain major determinants of stunting in early childhood. This finding underscores the nutrition early in need to address determinants of under the lifecycle through appropriate strategies, such as enhancing adolescent nutrition and family planning to delay the age of first pregnancy or increase spacing between births.196 Achieving high cover age of multiple micronutrient supplementation in preg nancy offers a new avenue to reduce SGA births and their consequences for mortality and growth in early childhood. The absence of appreciation of the crucial links of maternal and fetal nutrition to fertility and repeated preg nancies has been a major barrier in targeting of inter ventions to address these factors appropriately. Although the overall effect on stunting alleviation seems modest, the rate of decline suggested from the package of nutrition-specific interventions is plausible and within the broad range of observed effects across countries. A review of global stunting trends by Stevens and colleagues197 showed average rates of reduction in stunting in the best performing countries ranging from 2142% over the past decade, broadly consistent with what our model predicts from scaling up a core set of nutrition-specific inter ventions. Importantly, the countries that have made tremendous strides in improving nutrition and health outcomes (such as Brazil, China, Saudi Arabia, Kuwait, and Chile) have implemented nutrition-specific inter ventions, but also have been settings with exceptional economic growth, and investments in nutrition-sensitive interventions to address population health, education, and social sector development. Much the same conclusions were drawn by UNICEF in a report on undernutrition.198 A major advance on our previous review of inter ventions is the addition of delivery platforms that allow us to assess strategies to reach populations who are not being reached currently. Our findings suggest that com munity platforms offer a unique opportunity to engage and reach poor and difficult to access populations through com munication and outreach strategies. These strategies could also lead to potential integration of nutrition with maternal, newborn, and child health interventions. Since several countries are investing in
59

Series

community health worker programmes to address maternal, newborn, and child health,125 much potential tion promotion and therapeutic exists for scaling up nutri interventions through such platforms and hence integrating the two at point-of -service delivery. This integration could also help achieve reductions in inequities in the short term as has been noted by universal scaling up of selected maternal and child survival interventions.199 However, importantly, imple mentation of such programmes involves unique combinations and sequencing of health system policies, actions, and advocacy. Community-based nutrition pro rammes need meticulous planning, a rights-based framework for engagement of communities and other sectors, and piloting. Other health system pillars are crucial to success, including training and support for community health workers, strengthening of the supply formation systems, monitoring, and chain, simplified in regular feedback. The model used in this review estimates feasible reductions in mortality and stunting with enhanced investments. The same investments that can achieve these results will also lead to other improvements in cognitive and socioemotional development. Although these outcomes are not included in the model, partly because the costing database differs (ie, LiST mainly addresses mortality), substantial evidence from a range of models and longitudinal studies confirms that the benefits in terms of overall development on human capacity are appreciable. In terms of cost, an annual outlay of an additional $96 billion to bring to scale a range of nutrition interventions that would save nearly 1 million lives is reasonable since many interventions would be scaled up from negligible coverage rates. The cost per (discounted) life-year saved is about $370 for a set of interventions that could effectively deliver optimum nutrition to pregnant women, infants, and children, and manage SAM. These figures suggest that the nutrition interventions are well within the cost-effectiveness benchmark (less than three-times per person income) for all countries. More than half the $96 billion is accounted for by two large countries that could rely heavily on domestic resources (India and Indonesia). Consumables (whether drugs, or other items such as for transport or administration) account for slightly less than half the $96 billion, and all but the poorest countries can be expected to cover most of the expenditures on personnel; $34 billion from external donors could make a substantial difference to child nutrition. What proportion of development assistance for health is earmarked for nutrition is unclear. Global tracking data from the Institute for Health Metrics and Evaluation were unable to disaggregate nutrition-related funding from the annual funding of $517 billion for maternal, newborn, and child health programmes.200 The Countdown col laboration estimates nutrition-specific funding for the same year (2012) at $3245 million.201
60

Much funding for nutrition-relevant programmes prob ably overlaps with existing programmes for maternal, newborn, and child health and health systems strengthen ing and there might also be potential synergies, making it possible to share costs. The evidence from carefully conducted cohort studies202 of benefits of higher birthweight and early linear growth on education and improved health outcomes is con vincing and consistent with the overall message from our review and modelling exercise. As the world moves towards the post-2015 development agenda, it is important to draw attention to the unfinished agenda of maternal and child undernutrition and to the emerging issues of obesity. Our review reconfirms the existence of feasible and low cost evidence-based interventions and the fact that coverage rates for many of these inter ventions remain poor203 and for some, nonexistent. In view of the importance of fetal nutrition and poverty alleviation strategies to reach those in greatest need, priority must be given to scaling up nutrition specific and sensitive interventions in some of the highest burden countries. At the same time, in view of the increasing importance of non-communicable diseases, concerted efforts must be made to develop and implement interventions to reduce the risk of obesity.
Contributors ZAB conceptualised the review in consultation with the coordinators (PW, AL, SH, and REB) and wrote the first draft of the paper with substantial inputs from JKD. NW, YT, and AR developed the modification of LiST for assessment of effect and equity. Costing for selected interventions was done by MFG and SH. VW contributed to the scientific literature search, screening, collection, and analysis of data for the deworming review. LL led the review of severe and moderate acute malnutrition. KW led the obesity prevention review and contributed to the severe and moderate acute malnutrition reviews. CM and SZ oversaw the obesity reviews. BAH assessed the effect of maternal multiple micronutrient supplements, neonatal vitamin A supplementation, and deworming in pregnant women. ZL contributed to reviews of complementary feeding strategies and community platforms. RAS contributed to reviews of micronutrient powders and breastfeeding. AI contributed to the reviews of calcium, balanced protein supplementation in pregnancy, and vitamin A supplementation. All authors and members of the review groups (below) saw successive drafts of the paper and provided input. ZAB finalized the paper and is the overall guarantor. The Lancet Nutrition Interventions Review Group Zulfiqar A Bhutta, Arjumand Rizvi, Jai K Das, Rehana A Salam, Aisha Yousafzai (Aga Khan University, Pakistan), Zohra S Lassi (University of Adelaide, Australia), Lindsey Lenters, Ceilidh McPhail, Kerri Wazny, Michelle F Gaffey, Stanley Zlotkin (Hospital for Sick Children, Canada), Aamer Imdad (SUNY Upstate Medical University, USA), Batool Azra Haider (Harvard School of Public Health, USA), Vivian Welch (University of Ottawa, Canada), Reynaldo Martorell (Emory University, USA), Robert E Black, Neff Walker, Yvonne Tam (Johns Hopkins University, USA), Tahmeed Ahmad (International Center for Diarrheal Diseases Research, Bangladesh). Maternal and Child Nutrition Study Group Robert E Black (Johns Hopkins Bloomberg School of Public Health, USA), Cesar Victora (Universidad de Federal de Pelotas, Brazil), Susan Walker (The University of the West Indies, Jamaica), Harold Alderman (International Food Policy Research Institute, USA), Zulfiqar A Bhutta (Aga Khan University, Pakistan), Stuart Gillespie (International Food Policy Research Institute, USA), Lawrence Haddad (Institute of Development Studies, UK), Susan Horton (University of

www.thelancet.com

Series

Waterloo, Canada), Anna Lartey (University of Ghana, Ghana), Venkatesh Mannar (The Micronutrient Initiative, Canada), Marie Ruel (International Food Policy Research Institute, USA), Patrick Webb (Tufts University, USA) Series Advisory Committee Marc Van Ameringen (Gain Health Organization, Switzerland), Mandana Arabi (New York Academy of Sciences, USA), Shawn Baker (Helen Keller International, USA), Martin Bloem (United Nations World Food Programme, Italy), Francesco Branca (WHO, Switzerland), Leslie Elder (The World Bank, USA), Erin McLean (Canadian International Development Agency, Canada), Carlos Monteiro (University of So Paulo, Brazil), Robert Mwadime (Makerere School of Public Health, Uganda), Ellen Piwoz (Bill & Melinda Gates Foundation, USA), Werner Schultink (UNICEF, USA), Lucy Sullivan (1000 Days, USA), Anna Taylor (Department for International Development, UK), Derek Yach (The Vitality Group, USA). The Advisory Committee provided advice in a meeting with Series Coordinators for each paper at the beginning of the process to prepare the Series and in a meeting to review and critique the draft reports. Conflicts of interest REB serves on the Boards of the Micronutrient Initiative, Vitamin Angels, the Child Health and Nutrition Research Initiative, and the Nestl Creating Shared Value Advisory Committee. VM serves on the Nestl Creating Shared Value Advisory Committee. The other authors declare that they have no conflicts of interest. As corresponding author Zulfiqar A Bhutta states that he had full access to all data and final responsibility to submit for publication. Acknowledgments Funding for the preparation of the Series was provided to the Johns Hopkins Bloomberg School of Public Health through a grant from the Bill & Melinda Gates Foundation. The sponsor had no role in the analysis and interpretation of the evidence or in writing the paper and the decision to submit for publication. We thank the Aga Khan University, Karachi, Pakistan and the Program for Program for Global Pediatric Research, Global Child Health, Hospital for Sick Children (Toronto, ON, Canada) for support to ZAB during the course of this work. References 1 Black RE, Victora CG, Walker SP, and the Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S01406736(13)60937-X. 2 Liu L, Johnson HL, Cousens S, et al, for the Child Health Epidemiology Reference Group of WHO and UNICEF. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. Lancet 2012; 379: 215161. 3 Lim SS, Vos T, Flaxman AD, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 19902010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012; 380: 222460. 4 Bhutta ZA, Ahmed T, Black RE, et al, for the Maternal and Child Undernutrition Group. What works? Interventions for maternal and child undernutrition and survival. Lancet 2008; 371: 41740. 5 Walker N, Fischer-Walker C, Bryce J, Bahl R, Cousens S. Standards for CHERG reviews of intervention effects on child survival. Int J Epidemiol 2010; 39 (suppl 1): i2131. 6 WHO, The Partnership for Maternal, Newborn and Child Health. Reaching child brides. 2012. http://www.who.int/pmnch/topics/ part_publications/knowledge_summary_22_reaching_child_brides/ en/ (accessed March 25, 2013). 7 Ronnenberg AG, Wood RJ, Wang X, et al. Preconception hemoglobin and ferritin concentrations are associated with pregnancy outcome in a prospective cohort of Chinese women. J Nutr 2004; 134: 2586. 8 Mehra S, Agrawal D. Adolescent health determinants for pregnancy and child health outcomes among the urban poor. Indian Pediatr 2004; 41: 137.

10 11 12 13 14

15 16 17 18

19

20 21

22 23 24 25 26 27 28 29

30

31

Haldre K, Rahu K, Karro H, Rahu M. Is a poor pregnancy outcome related to young maternal age? A study of teenagers in Estonia during the period of major socio-economic changes (from 1992 to 2002). Eur J Obstet Gynecol Reprod Biol 2007; 131: 4551. Paranjothy S, Broughton H, Adappa R, Fone D. Teenage pregnancy: who suffers? Arch Dis Child 2009; 94: 239. WHO. Adolescent pregnancy: unmet needs and undone deeds. Geneva: World Health Organization, 2007. UNFPA. Marrying too young: end child marriage. New York, NY: UN Population Fund, 2012. Conde-Agudelo A, Rosas-Bermudez A, Kafury-Goeta AC. Birth spacing and risk of adverse perinatal outcomes: a meta-analysis. JAMA 2006; 295: 1809. Conde-Agudelo A, Rosas-Bermudez A, Castao F, Norton MH. Effects of birth spacing on maternal, perinatal, infant, and child health: a systematic review of causal mechanisms. Stud Fam Plann 2012; 43: 93114. WHO. Meeting report. Meeting to develop a global consensus on preconception care to reduce maternal and childhood mortality and morbidity; Geneva; Feb 67, 2012. Rosenthal AN, Paterson Brown S. Is there an incremental rise in the risk of obstetric intervention with increasing maternal age? BJOG 1998; 105: 106469. Carolan M. The graying of the obstetric population: implications for the older mother. J Obstet Gynecol Neonatal Nurs 2003; 32: 1927. Alkema L, Kantorova V, Menozzi C, Biddlecom A. National, regional, and global rates and trends in contraceptive prevalence and unmet need for family planning between 1990 and 2015: a systematic and comprehensive analysis. Lancet 2013; published online March 13. http://dx.doi.org/10/1016/S0140-6736(12)62204-1. De-Regil LM, Fernandez-Gaxiola AC, Dowswell T, Pena-Rosas JP. Effects and safety of periconceptional folate supplementation for preventing birth defects. Cochrane Database Syst Rev 2010; 10: CD007950. Lassi ZS, Salam RA, Haider BA, Bhutta ZA. Folic acid supplementation during pregnancy for maternal health and pregnancy outcomes. Cochrane Database Syst Rev 2013; 3: CD006896. Fernandez-Gaxiola AC, De-Regil LM. Intermittent iron supplementation for reducing anaemia and its associated impairments in menstruating women. Cochrane Database Syst Rev 2011; 12: CD009218. Pena-Rosas JP, De-Regil LM, Dowswell T, Viteri FE. Daily oral iron supplementation during pregnancy. Cochrane Database Syst Rev 2012; 12: CD004736. Haider BA, Bhutta ZA. Multiple-micronutrient supplementation for women during pregnancy. Cochrane Database Syst Rev 2012; 11: CD004905. Imdad A, Bhutta ZA. Effects of calcium supplementation during pregnancy on maternal, fetal and birth outcomes. Paediatr Perinat Epidemiol 2012; 26: 13852. Zimmermann MB. The effects of iodine deficiency in pregnancy and infancy. Paediatr Perinat Epidemiol 2012; 26 (suppl 1): 10817. Imdad A, Bhutta ZA. Maternal nutrition and birth outcomes: effect of balanced protein-energy supplementation. Paediatr Perinat Epidemiol 2012; 26 (suppl 1): 17890. Pena-Rosas JP, De-Regil LM, Dowswell T, Viteri FE. Intermittent oral iron supplementation during pregnancy. Cochrane Database Syst Rev 2012; 7: CD009997. WHO. Guideline: daily iron and folic acid supplementation in pregnant women. Geneva: World Health Organization, 2012. Ronsmans C, Fisher DJ, Osmond C, Margetts BM, Fall CH. Multiple micronutrient supplementation during pregnancy in low-income countries: a meta-analysis of effects on stillbirths and on early and late neonatal mortality. Food Nutr Bull 2009; 30: S54755. Vaidya A, Saville N, Shrestha BP, Costello AM, Manandhar DS, Osrin D. Effects of antenatal multiple micronutrient supplementation on childrens weight and size at 2 years of age in Nepal: follow-up of a double-blind randomised controlled trial. Lancet 2008; 371: 49299. West KP Jr, Shamim AA, Labrique AB, et al. Efficacy of antenatal multiple micronutrient (MM) vs iron-folic acid supplementation in improving gestational and postnatal viability in rural Bangladesh: the JiVitA-3 Trial. FASEB J 2013; 27: 5225 (abstr).

www.thelancet.com

61

Series

32 Srinivas SK, Edlow AG, Neff PM, Sammel MD, Andrela CM, Elovitz MA. Rethinking IUGR in preeclampsia: dependent or independent of maternal hypertension? J Perinatol 2009; 29: 68084. 33 Duley L. The global impact of pre-eclampsia and eclampsia. Semin Perinatol 2009; 33: 13037. 34 Hofmeyr GJ, Lawrie TA, Atallah AN, Duley L. Calcium supplementation during pregnancy for preventing hypertensive disorders and related problems. Cochrane Database Syst Rev 2010; 8: CD001059. 35 Wu T, Liu GJ, Li P, Clar C. Iodised salt for preventing iodine deficiency disorders. Cochrane Database Syst Rev 2006; 3: CD003204. 36 Kulier R, de Onis M, Gulmezoglu AM, Villar J. Nutritional interventions for the prevention of maternal morbidity. Int J Gynaecol Obstet 1998; 63: 23146. 37 de Onis M, Villar J, Gulmezoglu M. Nutritional interventions to prevent intrauterine growth retardation: evidence from randomized controlled trials. Eur J Clin Nutr 1998; 52 (suppl 1): S8393. 38 Kramer MS, Kakuma R. Energy and protein intake in pregnancy. Cochrane Database Syst Rev 2003; 4: CD000032. 39 McDonald SJ, Middleton P. Effect of timing of umbilical cord clamping of term infants on maternal and neonatal outcomes. Cochrane Database Syst Rev 2009; 2: CD004074. 40 Rabe H, Diaz-Rossello JL, Duley L, Dowswell T. Effect of timing of umbilical cord clamping and other strategies to influence placental transfusion at preterm birth on maternal and infant outcomes. Cochrane Database Syst Rev 2011; 8: CD003248. 41 Puckett RM, Offringa M. Prophylactic vitamin K for vitamin K deficiency bleeding in neonates. Cochrane Database Syst Rev 2010; 4: CD002776. 42 Darlow BA, Graham PJ. Vitamin A supplementation to prevent mortality and short and long-term morbidity in very low birthweight infants. Cochrane Database Syst Rev 2007; 4: CD000501. 43 Haider BA, Bhutta ZA. Neonatal vitamin A supplementation for the prevention of mortality and morbidity in term neonates in developing countries. Cochrane Database Syst Rev 2011; 10: CD006980. 44 Moore ER, Anderson GC, Bergman N, Dowswell T. Early skin-to-skin contact for mothers and their healthy newborn infants. Cochrane Database Syst Rev 2012; 5: CD003519. 45 Conde-Agudelo A, Belizan JM, Diaz-Rossello J. Kangaroo mother care to reduce morbidity and mortality in low birthweight infants. Cochrane Database Syst Rev 2011; 3: CD002771. 46 WHO. Meeting report. Technical consultation on vitamin a in newborn health: mechanistic studies; Geneva; Dec 13, 2009. 47 Dyson L, McCormick FM, Renfrew MJ. Interventions for promoting the initiation of breastfeeding. Cochrane Database Syst Rev 2005; 2: CD001688. 48 Cai X WT, Brown DW. Global trends in exclusive breastfeeding. Int Breastfeed J 2012; 7: 12. 49 Edmond KM, Zandoh C, Quigley MA, Amenga-Etego S, Owusu-Agyei S, Kirkwood BR. Delayed breastfeeding initiation increases risk of neonatal mortality. Pediatrics 2006, 117: e3806. 50 Mullany LC, Katz J, Li YM, et al. Breast-feeding patterns, time to initiation, and mortality risk among newborns in southern Nepal. J Nutr 2008; 138: 599603. 51 Debes AK, Kohli A, Walker N, Edmond K, Mullany LC. Time to initiation of breastfeeding and neonatal mortality and morbidity: a systematic review. BMC Public Health (submitted). 52 Imdad A, Yakoob MY, Bhutta ZA. Effect on breastfeeding promotion interventions on breastfeeding rates, with special focus on developing countries. BMC Public Health 2011; 11 (suppl 3): S24. 53 Haroon S, Das JK, Salam RA, Bhutta ZA. Breastfeeding promotion interventions and breastfeeding practices: a systematic review. BMC Public Health (in press). 54 Lassi ZS, Zahid GS, Das JK, Bhutta ZA. Impact of complementary food and education on complementary food on growth and morbidity of children less than 2 years of age in developing countries: a systematic review. BMC Public Health (in press). 55 Imdad A, Herzer K, Mayo-Wilson E, Yakoob MY, Bhutta ZA. Vitamin A supplementation for preventing morbidity and mortality in children from 6 months to 5 years of age. Cochrane Database Syst Rev 2010; 12: CD008524.

56 De-Regil LM, Jefferds MED, Sylvetsky AC, Dowswell T. Intermittent iron supplementation for improving nutrition and development in children under 12 years of age. Cochrane Database Syst Rev 2011; 12: CD009085. 57 Sachdev H, Gera T, Nestel P. Effect of iron supplementation on mental and motor development in children: systematic review of randomised controlled trials. Public Health Nutr 2005; 8: 11732. 58 Allen LH, Peerson JM, Olney DK. Provision of multiple rather than two or fewer micronutrients more effectively improves growth and other outcomes in micronutrient-deficient children and adults. J Nutr 2009; 139: 102230. 59 Salam RA, MacPhail C, Das JK, Bhutta ZA. Effectiveness of micronutrient powders (MNP) in women and children. BMC Public Health (in press). 60 Imdad A, Bhutta ZA. Effect of preventive zinc supplementation on linear growth in children under 5 years of age in developing countries: a meta-analysis of studies for input to the lives saved tool. BMC Public Health. 2011; 11 (suppl 3): S22. 61 Yakoob MY, Theodoratou E, Jabeen A, et al. Preventive zinc supplementation in developing countries: impact on mortality and morbidity due to diarrhea, pneumonia and malaria. BMC Public Health 2011; 11 (suppl 3): S23. 62 Gogia S, Sachdev HS. Zinc supplementation for mental and motor development in children. Cochrane Database Syst Rev 2012; 12: CD007991. 63 Abdulwadud OA, Snow ME. Interventions in the workplace to support breastfeeding for women in employment. Cochrane Database Syst Rev 2012; 3: CD006177. 64 WHO. Meeting report. Informal meeting to review and develop indicators for complementary feeding. Washington DC: World Health Organization, 2002. 65 Ruel MT, Menon P. Child feeding practices are associated with child nutritional status in Latin America: innovative uses of the demographic and health surveys. J Nutr 2002; 132: 118087. 66 Arimond M, Ruel MT. Dietary diversity is associated with child nutritional status: evidence from 11 demographic and health surveys. J Nutr 2004; 134: 257985. 67 Marriott BP, White A, Hadden L, Davies JC, Wallingford JC. World Health Organization (WHO) infant and young child feeding indicators: associations with growth measures in 14 low-income countries. Matern Child Nutr 2012; 8: 35470. 68 Dewey KG, Adu-Afarwuah S. Systematic review of the efficacy and effectiveness of complementary feeding interventions in developing countries. Matern Child Nutr 2008; 4 (suppl 1): 2485. 69 Awasthi S, Peto R, Read S, et al, and the DEVTA (Deworming and Enhanced Vitamin A) team. Vitamin A supplementation every 6 months with retinol in 1 million pre-school children in north India: DEVTA, a cluster-randomised trial. Lancet 2013; 381: 146977. 70 Sazawal S, Black RE, Ramsan M, et al. Effects of routine prophylactic supplementation with iron and folic acid on admission to hospital and mortality in preschool children in a high malaria transmission setting: community-based, randomised, placebo-controlled trial. Lancet 2006; 367: 13343. 71 WHO. Guideline: intermittent iron supplementation in preschool and school-age children, Geneva: World Health Organization, 2011. 72 Neufeld LM, Ramakrishnan U. Multiple micronutrient interventions during early childhood: moving towards evidence-based policy and program planning. J Nutr 2011; 141: 206465. 73 Eilander A, Gera T, Sachdev HS, et al. Multiple micronutrient supplementation for improving cognitive performance in children: systematic review of randomized controlled trials. Am J Clin Nutr 2010; 91: 11530. 74 Gera T, Sachdev HP. Effect of iron supplementation on incidence of infectious illness in children: systematic review. BMJ 2002; 325: 1142. 75 Soofi SB, Cousens S, Iqbal SP, et al. Effect of provision of daily zinc and iron with several micronutrients on growth and morbidity among young children in Pakistan: a cluster-randomised trial Lancet 2013; published online April 18. http://dx.doi.org/10.1016/ S0140-6736(13)60437-7. 76 Brown KH, Peerson JM, Baker SK, Hess SY. Preventive zinc supplementation among infants, preschoolers, and older prepubertal children. Food Nutr Bull 2009; 30 (suppl 1): S140.

62

www.thelancet.com

Series

77 Cairncross S, Hunt C, Boisson S, et al. Water, sanitation and hygiene for the prevention of diarrhoea. Int J Epidemiol 2010; 39 (suppl 1): i193205. 78 Spears D. How much international variation in child height can sanitation explain? World Bank policy research working paper, no WPS 6351 2013. http://go.worldbank.org/SZE5WUJBI0 (accessed May 15, 2013). 79 Haider BA, Humayun Q, Bhutta ZA. Effect of administration of antihelminthics for soil transmitted helminths during pregnancy. Cochrane Database Syst Rev 2012; 2: CD005547. 80 Taylor-Robinson D, Jones A, Garner P. Deworming drugs for treating soil-transmitted intestinal worms in children: effects on growth and school performance. Cochrane Database Syst Rev 2012; 4: CD000371. 81 Gaffey MF, Wazny K, Bassani DG, Bhutta ZA. Dietary management of childhood diarrhea in low- and middle-income countries: a systematic review. BMC Public Health (in press). 82 Walker CLF, Black RE. Zinc for the treatment of diarrhoea: effect on diarrhoea morbidity, mortality and incidence of future episodes. Int J Epidemiol 2010; 39 (suppl 1): i6369. 83 Garner P, Gulmezoglu AM. Drugs for preventing malaria in pregnant women. Cochrane Database Syst Rev 2009; 4: CD000169. 84 Gamble C, Ekwaru JP, ter Kuile FO. Insecticide-treated nets for preventing malaria in pregnancy. Cochrane Database Syst Rev 2009; 2: CD003755. 85 Meremikwu MM, Donegan S, Sinclair D, Esu E, Oringanje C. Intermittent preventive treatment for malaria in children living in areas with seasonal transmission. Cochrane Database Syst Rev 2012; 2: CD003756. 86 Lengeler C. Insecticide-treated bed nets and curtains for preventing malaria. Cochrane Database Syst Rev 2009; 2: CD000363. 87 Dangour AD, Watson L, Cumming O, et al. Interventions to improve water quality and supply, sanitation and hygiene practices, and their effects on the nutritional status of children. Cochrane Database Syst Rev 2011; 3: CD009382. 88 Awasthi S, Peto R, Read S, Richards SM, Pande V, Bundy D. Population deworming every 6 months with albendazole in 1 million pre-school children in north India: DEVTA, a cluster-randomised trial. Lancet 2013; published online March 14. http://dx.doi. org/10.1016/S0140-6736(12)62126-6. 89 WHO. Guideline update: technical aspects of the management of severe acute malnutrition in infants and children. Geneva: World Health Organization, 2013. 90 Collins S, Sadler K, Dent N, et al. Key issues in the success of community-based management of severe malnutrition. Food Nutr Bull 2006; 27: S4982. 91 Schofield C, Ashworth A. Why have mortality rates for severe malnutrition remained so high? Bull World Health Organ 1996; 74: 22329. 92 Chinkhumba J, Tomkins A, Banda T, Mkangama C, Fergusson P. The impact of HIV on mortality during in-patient rehabilitation of severely malnourished children in Malawi. Trans R Soc Trop Med Hyg 2008; 102: 63944. 93 Fergusson P, Chinkhumba J, Grijalva-Eternod C, Banda T, Mkangama C, Tomkins A. Nutritional recovery in HIV-infected and HIV-uninfected children with severe acute malnutrition. Arch Dis Child 2009; 94: 51216. 94 Hossain MM, Hassan MQ, Rahman MH, Kabir A, Hannan AH, Rahman A. Hospital management of severely malnourished children: comparison of locally adapted protocol with WHO protocol. Indian Pediatr 2009; 46: 21317. 95 Khanum S, Ashworth A, Huttly SR. Growth, morbidity, and mortality of children in Dhaka after treatment for severe malnutrition: a prospective study. Am J Clin Nutr 1998; 67: 94045. 96 Puoane T, Cuming K, Sanders D, Ashworth A. Why do some hospitals achieve better care of severely malnourished children than others? Five-year follow-up of rural hospitals in Eastern Cape, South Africa. Health Policy Plan 2008; 23: 42837. 97 Lenters LM, Wazny K, Webb P, Ahmed T, Bhutta ZA. Treatment of severe and moderate acute malnutrition in low- and middle-income settings: a systematic review, meta-analysis and delphi process. BMC Public Health (in press). 98 Trehan I, Goldbach HS, LaGrone LN, et al. Antibiotics as part of the management of severe acute malnutrition. N Engl J Med 2013; 368: 42535.

99 Collins S. Community-based therapeutic carea new paradigm for selective feeding in nutritional crises. Humanitarian Policy Network paper 48. London: Overseas Development Institute, 2004. 100 Furber CM, McGowan L, Bower P, Kontopantelis E, Quenby S, Lavender T. Antenatal interventions for reducing weight in obese women for improving pregnancy outcome. Cochrane Database Syst Rev 2013; 1: CD009334. 101 Amorin Adegboye AR, Linne YM, Lourenco PMC. Diet or exercise, or both, for weight reduction in women after childbirth. Cochrane Database Syst Rev 2012; 2: CD005627. 102 Arenz S, Ruckerl R, Koletzko B, von Kries R. Breast-feeding and childhood obesitya systematic review. Int J Obes Relat Metab Disord 2004; 28: 124756. 103 Harder T, Bergmann R, Kallischnigg G, Plagemann A. Duration of breastfeeding and risk of overweight: a meta-analysis. Am J Epidemiol 2005; 162: 397403. 104 Owen CG, Martin RM, Whincup PH, Davey-Smith G, Gillman MW, Cook DG. The effect of breastfeeding on mean body mass index throughout life: a quantitative review of published and unpublished observational evidence. Am J Clin Nutr 2005; 82: 1298307. 105 Owen CG, Martin RM, Whincup PH, Smith GD, Cook DG. Effect of infant feeding on the risk of obesity across the life course: a quantitative review of published evidence. Pediatrics 2005; 115: 136777. 106 Plagemann A, Harder T. Breast feeding and the risk of obesity and related metabolic diseases in the child. Metab Syndr Relat Disord 2005; 3: 22232. 107 Waters E, de Silva-Sanigorski A, Hall BJ, et al. Interventions for preventing obesity in children. Cochrane Database Syst Rev 2011; 12: CD001871. 108 Martin RM, Patel R, Kramer MS, et al. Effects of promoting longer-term and exclusive breastfeeding on adiposity and insulin-like growth factor-I at age 115 years: a randomized trial. JAMA 2013; 309: 100513. 109 Oude Luttikhuis H, Baur L, Jansen H, et al. Interventions for treating obesity in children. Cochrane Database Syst Rev 2009; 1: CD001872. 110 Horton S, Alderman H, Rivera J. Hunger and malnutrition. Copenhagen consensus 2008: malnutrition and hunger. http:// thousanddays.org/wp-content/uploads/2011/05/CopenhagenConsensus-2008-summary.pdf (accessed March 15, 2013). 111 WHO, FAO. Guidelines on food fortification with micronutrients. Geneva: World Health Organization; Food and Agricultural Organization, 2006. 112 Mayer JE, Pfeiffer WH, Beyer P. Biofortified crops to alleviate micronutrient malnutrition. Curr Opin Plant Biol 2008; 11: 16670. 113 Eichler K, Wieser S, Ruthemann I, Brugger U. Effects of micronutrient fortified milk and cereal food for infants and children: a systematic review. BMC Public Health 2012; 12: 506. 114 Gera T, Sachdev HS, Boy E. Effect of iron-fortified foods on hematologic and biological outcomes: systematic review of randomized controlled trials. Am J Clin Nutr 2012; 96: 30924. 115 Natri AM, Salo P, Vikstedt T, et al. Bread fortified with cholecalciferol increases the serum 25-hydroxyvitamin D concentration in women as effectively as a cholecalciferol supplement. J Nutr 2006; 136: 12327. 116 Diaz-Gomez NM, Domenech E, Barroso F, Castells S, Cortabarria C, Jimenez A. The effect of zinc supplementation on linear growth, body composition, and growth factors in preterm infants. Pediatrics 2003; 111: 100209. 117 Harvey PWJ, Dary O. Governments and academic institutions play vital roles in food fortification: iron as an example. Public Health Nutr 2012; 1: 15. 118 Mannar MG, van Ameringen M. Role of public-private partnership in micronutrient food fortification. Food Nutr Bull 2003; 24 (suppl 4): S15154. 119 Gillespie S, Haddad L, Mannar V, Menon P, Nisbettt N, and the Maternal and Child Nutrition Study Group. The politics of reducing malnutrition: building commitment and accelerating progress. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/ S0140-6736(13)60842-9. 120 Bassani DG, Arora P, Wazny K, Gaffey MF, Lenters L, Bhutta ZA. Financial incentives and coverage of child health interventions: a systematic review and meta-analysis. BMC Public Health (in press).

www.thelancet.com

63

Series

121 Haines A, Sanders D, Lehmann U, et al. Achieving child survival goals: potential contribution of community health workers. Lancet 2007; 369: 212131. 122 Bhutta ZA, Das JK, Walker N, et al, for The Lancet Diarrhoea and Pneumonia Interventions Study Group. Interventions to address deaths from childhood pneumonia and diarrhoea equitably: what works and at what cost? Lancet 2013; 381: 141729. 123 Lassi ZS, Haider BA, Bhutta ZA. Community-based intervention packages for reducing maternal and neonatal morbidity and mortality and improving neonatal outcomes. Cochrane Database Syst Rev 2010; 11: CD007754. 124 Lewin S, Munabi-Babigumira S, Glenton C, et al. Lay health workers in primary and community health care for maternal and child health and the management of infectious diseases. Cochrane Database Syst Rev 2010; 3: CD004015. 125 GHWA. Global experience of community health workers for delivery of health related millennium development goals: a systematic review, country case studies, and recommendations for integration into national health systems. Geneva: Global Health Workforce Alliance, 2010. 126 Gove S. Integrated management of childhood illness by outpatient health workers: technical basis and overview. The WHO Working Group on Guidelines for Integrated Management of the Sick Child. Bull World Health Organ 1997; 75 (suppl 1): 724. 127 Ahmed HM, Mitchell M, Hedt B. National implementation of Integrated Management of Childhood Illness (IMCI): policy constraints and strategies. Health Policy 2010; 96: 12833. 128 Schellenberg JRM, Adam T, Mshinda H, et al. Effectiveness and cost of facility-based Integrated Management of Childhood Illness (IMCI) in Tanzania. Lancet 2004; 364: 158394. 129 Arifeen SE, Hoque DM, Akter T, et al. Effect of the Integrated Management of Childhood Illness strategy on childhood mortality and nutrition in a rural area in Bangladesh: a cluster randomised trial. Lancet 2009; 374: 393403. 130 Kristjansson B, Petticrew M, MacDonald B, et al. School feeding for improving the physical and psychosocial health of disadvantaged students. Cochrane Database Syst Rev 2009; 1: CD004676. 131 Doherty T, Chopra M, Tomlinson M, Oliphant N, Nsibande D, Mason J. Moving from vertical to integrated child health programmes: experiences from a multi-country assessment of the Child Health Days approach in Africa. Trop Med Int Health 2010; 15: 296305. 132 UNICEF. Child health days 19992009: key achievements and the way forward. A report prepared for the UNICEF Joint Working Group on Child Health Days. New York, NY: UNICEF, 2011. 133 The Sphere Project. Sphere handbook. 2011. http://www. spherehandbook.org/en/how-to-use-this-chapter-3/ (accessed May 15, 2013). 134 Young H, Borrel A, Holland D, Salama P. Public nutrition in complex emergencies. Lancet 2004; 364: 1899909. 135 World Food Programme. Nutrition in emergencies: WFP experiences and challenges. Food Nutr Bull 2006; 27: 5766. 136 The Johns Hopkins Universitys Bloomberg School of Public Health and International Committee of the Red Cross. Public health guide in emergencies, 2nd edn. Geneva: ICRC, 2008. 137 Global Nutrition Cluster of the Inter Agency Standing Committee. A toolkit for addressing nutrition in emergency situations. New York, NY: United Nations International Childrens Fund, 2008. 138 Bailey S, Hedlund K. The impact of cash transfers on nutrition in emergency and transitional contexts: a review of the evidence. HPG Synthesis Paper. London: Overseas Development Institute, 2012. 139 Langendorf C, Roederer T. Evaluation de differentes strategies de distributions preventives de la malnutrition au Niger. 2012 District de Madarounfa Rgion de Maradi Aot 2011Novembre 2012. Rapport intermdiaire. Paris: Mdecins Sans Frontires, 2012. 140 Grijalva-Eternod CS, Wells JCK, Cortina-Borja M, et al. The double burden of obesity and malnutrition in a protracted emergency setting: a cross-sectional study of western Sahara refugees. PLoS Med 2012; 9: e1001320. 141 IFRC. Focus on hunger and malnutrition. World Disasters Report 2011. Geneva: International Federation of Red Cross and Red Crescent Societies, 2011.

142 Hall A, Blankson B, Shoham J. The impact and effectiveness of emergency nutrition and Nutrition related interventions: a review of published evidence 20042010. Oxford: Emergency Nutrition Network, 2011. 143 UNHCR, WFP. Guidelines for selective feeding: the management of malnutrition in emergencies. In collaboration with the United Nations Standing Committee on Nutrition and the World Health Organization. Geneva: United Nations High Commission for Refugees, World Food Programme, 2011. 144 WHO, WFP, SCN, UNICEF. Community-Based Management of Severe Acute Malnutrition: A Joint Statement by the World Health Organization, the World Food Programme, the United Nations System Standing Committee on Nutrition and the United Nations Childrens Fund. Geneva: World Health Organization, World Food Programme, United Nations System Standing Committee on Nutrition, United Nations Childrens Fund, 2007. 145 Jones G, Stekettee RW, Black RE, Bhutta ZA, Morris SS, and the Bellagio Child Survival Study Group. How many child deaths can we prevent this year? Lancet 2003; 362: 6571. 146 Dherani M, Pope D, Mascarenhas M, Smith KR, Weber M, Bruce NG. Indoor air pollution from unprocessed solid fuel use and pneumonia risk in under-5 children: systematic review and meta-analysis. Bull World Health Organ 2008; 86: 39098. 147 Pope DP, Mishra V, Thompson L, et al. Risk of low birth weight and stillbirth associated with indoor air pollution from solid fuel use in developing countries. Epidemiol Rev 2010; 32: 7081. 148 Bruce N, Dherani M, Das J, et al. Control of household air pollution for child survival: estimates for intervention impacts. BMC Public Health (in press). 149 Smith KR, McCracken JP, Weber MW, et al. Effect of reduction in household air pollution on childhood pneumonia in Guatemala (RESPIRE): a randomised controlled trial. Lancet 2011; 378: 171726. 150 Thompson LM, Bruce N, Eskenazi B, Diaz A, Pope D, Smith KR. Impact of reduced maternal exposures to wood smoke from an introduced chimney stove on newborn birth weight in rural Guatemala. Environ Health Perspect 2011; 119: 148994. 151 Wei SQ, Qi HP, Luo ZC, Fraser WD. Maternal vitamin D status and adverse pregnancy outcomes: a systematic review and meta-analysis. J Matern Fetal Neonatal Med 2013; published online Feb 11. DOI:10.31 09/14767058.2013.765849. 152 Lawlor DA, Wills AK, Fraser A, Sayers A, Fraser WD, Tobias JH. Association of maternal vitamin D status during pregnancy with bone-mineral content in offspring: a prospective cohort study. Lancet 2013; published online March 19. http://doi.dx.org/10.1016/S01406736(12)62203-X. 153 De-Regil LM, Palacios C, Ansary A, Kulier R, Pena-Rosas JP. Vitamin D supplementation for women during pregnancy. Cochrane Database Syst Rev 2012; 2: CD008873. 154 Mori R, Ota E, Middleton P, Tobe-Gai R, Mahomed K, Bhutta ZA. Zinc supplementation for improving pregnancy and infant outcome. Cochrane Database Syst Rev 2012; 7: CD000230. 155 Makrides M, Duley L, Olsen SF. Marine oil, and other prostagladin precursor, suppementation for pregnancy uncomplicated by pre-eclampsia or intrauterine growth restriction. Cochrane Database Syst Rev 2007; 3: CD003402. 156 Salvig JD, Lamont RF. Evidence regarding an effect of marine n-3 fatty acids on preterm birth: a systematic review and meta-analysis. Acta Obstet Gynecol Scand 2011; 90: 82538. 157 Dziechciarz P, Horvath A, Szajewska H. Effects of n-3 long-chain polyunsaturated fatty acid supplementation during pregnancy and/ or lactation on neurodevelopment and visual function in children: a systematic review of randomized controlled trials. J Am Coll Nutr 2010; 29: 44354. 158 Klemens CM, Berman DR, Mozurkewich EL. The effect of perinatal omega-3 fatty acid supplementation on inflammatory markers and allergic diseases: a systematic review. BJOG 2011; 118: 91625. 159 Rodriguez G, Iglesia I, Bel-Serrat S, Moreno LA. Effect of n-3 long chain polyunsaturated fatty acids during the perinatal period on later body composition. Br J Nutr 2012; 107: S11728. 160 Szajewska H, Horvath A, Koletzko B. Effect of n-3 long-chain polyunsaturated fatty acid supplementation of women with low-risk pregnancies on pregnancy outcomes and growth measures at birth: a meta-analysis of randomized controlled trials. Am J Clin Nutr 2006; 83: 133744.

64

www.thelancet.com

Series

161 Ryan AS, Astwood JD, Gautier S, Kuratko CN, Nelson EB, Salem N Jr. Effects of long-chain polyunsaturated fatty acid supplementation on neurodevelopment in childhood: a review of human studies. Prostaglandins Leukot Essent Fatty Acids 2010; 82: 30514. 162 Muhlhausler BS, Gibson RA, Makrides M. Effect of long-chain polyunsaturated fatty acid supplementation during pregnancy or lactation on infant and child body composition: a systematic review. Am J Clin Nutr 2010; 92: 85763. 163 Michaelsen KF, Dewey KG, Perez-Exposito AB, Nurhasan M, Lauritzen L, Roos N. Food sources and intake of n-6 and n-3 fatty acids in low-income countries with emphasis on infants, young children (624 months), and pregnant and lactating women. Matern Child Nutr 2011; 7: 12440. 164 Prentice AM, van der Merwe L. Impact of fatty acid status on immune function of children in low-income countries. Matern Child Nutr 2011; 7: 8998. 165 Rahman A, Patel V, Maselko J, Kirkwood B. The neglected m in MCH programmeswhy mental health of mothers is important for child nutrition. Trop Med Int Health 2008; 13: 57983. 166 Satyanarayana VA, Lukose A, Srinivasan K. Maternal mental health in pregnancy and child behavior. Indian J Psychiatr 2011; 53: 351. 167 Austin MP, Priest SR, Sullivan EA. Antenatal psychosocial assessment for reducing perinatal mental health morbidity. Cochrane Database Syst Rev 2008; 4: CD005124. 168 Diego MA, Field T, Hernandez-Reif M, Deeds O, Ascencio A, Begert G. Preterm infant massage elicits consistent increases in vagal activity and gastric motility that are associated with greater weight gain. Acta Paediatr 2007; 96: 158891. 169 Vickers A, Ohlsson A, Lacy JB, Horsley A. Massage for promoting growth and development of preterm and/or low birth-weight infants. Cochrane Database Syst Rev 2009; 2: CD000390. 170 Field T, Diego M, Hernandez-Reif M. Preterm infant massage therapy research: a review. Infant Behav Dev 2011; 33: 11524. 171 Lerch C, Meissner T. Interventions for the prevention of nutritional rickets in term born children. Cochrane Database Syst Rev 2007; 4: CD006164. 172 Haider BA, Lassi ZS, Ahmed A, Bhutta ZA. Zinc supplementation as an adjunct to antibiotics in the treatment of pneumonia in children 2 to 59 months of age. Cochrane Database Syst Rev 2011; 10: CD007368. 173 Basnet S, Shrestha PS, Sharma A, et al. A randomized controlled trial of zinc as adjuvant therapy for severe pneumonia in young children. Pediatrics 2012; 129: 70108. 174 Das RR, Singh M, Shafiq N. Short-term therapeutic role of zinc in children <5 years of age hospitalised for severe acute lower respiratory tract infection. Paediatr Respir Rev 2012; 13: 18491. 175 Shah GS, Dutta AK, Shah D, Mishra OP. Role of zinc in severe pneumonia: a randomized double bind placebo controlled study. Ital J Pediatr 2012; 38: 36. 176 Srinivasan MG, Ndeezi G, Mboijana CK, et al. Zinc adjunct therapy reduces case fatality in severe childhood pneumonia: a randomized double blind placebo-controlled trial. BMC Med 2012; 10: 14. 177 Bhatnagar S, Wadhwa N, Aneja S, et al. Zinc as adjunct treatment in infants aged between 7 and 120 days with probable serious bacterial infection: a randomised, double-blind, placebo-controlled trial. Lancet 2012; 379: 207278. 178 Mehta K, Bhatta NK, Majhi S, Shrivastava MK, Singh RR. Role of zinc in neonatal sepsis: a double blinded, randomized, placebo controlled trial. Indian Pediatr 2013; 50: 39093. 179 Phuka J, Malete K, Thakwalakwa C. Complementary feeding with fortified spread and incidence of severe stunting in 68 month-old rural Malawians. Arch Pediatr Adolesc Med 2008; 162: 61926. 180 Phuka JC, Maleta K, Thakwalakwa C, et al. Postintervention growth of Malawian children who received 12-mo dietary complementation with a lipid-based nutrient supplement or maize-soy flour. Am J Clin Nutr 2009; 89: 38290. 181 Adu-Afarwuah S, Lartey A, Brown KH, Zlotkin S, Briend A, Dewey KG. Randomized comparison of 3 types of micronutrient supplements for home fortification of complementary foods in Ghana: effects on growth and motor development. Am J Clin Nutr 2007; 86: 41220.

182 Adu-Afarwuah S, Lartey A, Brown KH, Zlotkin S, Briend A, Dewey KG. Home fortification of complementary foods with micronutrient supplements is well accepted and has positive effects on infant iron status in Ghana. Am J Clin Nutr 2008; 87: 92938. 183 Walker N, Friberg I. Overview of the Lives Saved Tool (LiST). BMC Public Health (in press). 184 Darmstadt G, Walker N, Lawn JE, Bhutta ZA, Haws RA, Cousens S. Saving newborn lives in Asia and Africa: cost and impact of phased scale-up of interventions within the continuum of care. Health Policy Plann 2008; 23: 10117. 185 Garnett GP, Cousens S, Hallett TB, Steketee, R, Walker N. Mathematical models in the evaluation of health programmes. Lancet 2011; 378: 51525. 186 Sachdev HPS, Hall A, Walker N, eds. Development and use of the Lives Saved Tool (LiST): a model to estimate the impact of scaling up proven interventions on maternal, neonatal and child mortality. Int J Epidemiol 2010; 39 (suppl 1): i1205. 187 Fox M, Marterell R, van den Broek N, Walker N, eds. Technical inputs, enhancements and applications of the Lives Saved Tool (LiST). BMC Public Health 2011; 11 (suppl 3): S22. 188 Walker N, ed. Updates of assumptions and methods for the Lives Saved Tool (LiST). BMC Public Health (in press). 189 WHO, Global Health Observatory Data Repository. Life expectancy at birth, low income countries, both sexes combined, 2011. http:// apps.who.int/gho/data/view.main.700?lang=en (accessed April 8, 2013). 190 Amouzou A, Richard SA, Friberg IK, et al. How well does LiST capture mortality by wealth quintile? A comparison of measured versus modelled mortality rates among children under-five in Bangladesh. Int J Epidemiol 2010; 39 (suppl 1): i18692. 191 National Institute of Population Studies. Pakistan Demographic and Health Survey 200607. Islamabad, Pakistan. http://www. measuredhs.com/pubs/pdf/FR200/FR200.pdf (accessed on March 2, 2013). 192 Bhutta ZA, Hafeez A, Rizvi, et al. Reproductive, maternal, newborn, and child in Pakistan: challenges and opportunities. Lancet 2013; published online May 17. http://dx.doi.org/10.1016/S01406736(12)61999-0. 193 Futures Institute. UN integrated costing tool. http://www.who.int/ workforcealliance/knowledge/toolkit/10/en/ (accessed Dec 7, 2012). 194 Horton MS, McDonald C, Mahal A. Scaling-up nutrition: what will it cost? Washington DC: World Bank Directions in Development, 2009. 195 WHO. Health systems and information. http://www.who.int/ healthinfo/global_burden_disease/definition_regions/en/ (accessed April 20, 2013). 196 Winkvist A, Rasmussen KM, Habicht JP. A new definition of maternal depletion syndrome. Am J Public Health 1992; 82: 69194. 197 Stevens GA, Finucane MM, Paciorek CJ, et al. Trends in mild, moderate, and severe stunting and underweight, and progress towards MDG 1 in 141 developing countries: a systematic analysis of population representative data. Lancet 2012; 380: 82434. 198 UNICEF. Improving child nutrition: the increasing imperative for global progress. New York, NY: United Nations International Childrens Fund, 2013. 199 Victora CG, Barros AJ, Axelson H, et al. How changes in coverage affect equity in maternal and child health interventions in 35 Countdown to 2015 countries: an analysis of national surveys. Lancet 2012; 380: 114956. 200 IHME. Financing global health 2012: the end of the golden age? Seattle, WA: Institute for Health Metrics and Evaluation, 2012. 201 Hsu J, Pitt C, Greco G, Berman P, Mills A. Countdown to 2015: changes in official development assistance to maternal, newborn, and child health in 2009-10, and assessment of progress since 2003. Lancet 2012; 380: 115768. 202 Adair L, Fall CHD, Osmond C, et al, for the COHORTS group. Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: findings from five birth cohort studies. Lancet 2013; published online March 28. http://dx.doi.org/10.1016/S10406736(13)60103-8. 203 Bhutta ZA, Chopra M. The Countdown for 2015: what lies ahead? Lancet 2012; 380: 112527.

www.thelancet.com

65

Series

Maternal and Child Nutrition 3 Nutrition-sensitive interventions and programmes: how can they help to accelerate progress in improving maternal and child nutrition?
Marie T Ruel, Harold Alderman, and the Maternal and Child Nutrition Study Group*
Published Online June 6, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)60843-0 This is the third in a Series of four papers about maternal and child nutrition *Members listed at end of paper Poverty, Health and Nutrition Division, International Food Policy Research Institute, Washington, DC, USA (M T Ruel PhD, H Alderman PhD) Correspondence to: Dr Marie T Ruel, Poverty, Health and Nutrition Division, International Food Policy Research Institute, Washington, DC 20006, USA m.ruel@cgiar.org

Acceleration of progress in nutrition will require effective, large-scale nutrition-sensitive programmes that address key underlying determinants of nutrition and enhance the coverage and effectiveness of nutrition-specific interventions. We reviewed evidence of nutritional effects of programmes in four sectorsagriculture, social safety nets, early child development, and schooling. The need for investments to boost agricultural production, keep prices low, and increase incomes is undisputable; targeted agricultural programmes can complement these investments by supporting livelihoods, enhancing access to diverse diets in poor populations, and fostering womens empowerment. However, evidence of the nutritional effect of agricultural programmes is inconclusiveexcept for vitamin A from biofortification of orange sweet potatoeslargely because of poor quality evaluations. Social safety nets currently provide cash or food transfers to a billion poor people and victims of shocks (eg, natural disasters). Individual studies show some effects on younger children exposed for longer durations, but weaknesses in nutrition goals and actions, and poor service quality probably explain the scarcity of overall nutritional benefits. Combined early child development and nutrition interventions show promising additive or synergistic effects on child developmentand in some cases nutritionand could lead to substantial gains in cost, efficiency, and effectiveness, but these programmes have yet to be tested at scale. Parental schooling is strongly associated with child nutrition, and the effectiveness of emerging school nutrition education programmes needs to be tested. Many of the programmes reviewed were not originally designed to improve nutrition yet have great potential to do so. Ways to enhance programme nutrition-sensitivity include: improve targeting; use conditions to stimulate participation; strengthen nutrition goals and actions; and optimise womens nutrition, time, physical and mental health, and empowerment. Nutrition-sensitive programmes can help scale up nutrition-specific interventions and create a stimulating environment in which young children can grow and develop to their full potential.

Introduction
The food system is threatened by food and oil price volatility, diversion of resources from production of food to biofuels, climate change and related water shortages, persistent conflicts and emergencies, and natural disasters affecting agriculture production and yields.14 These challenges are compounded by changes in demand for food that are brought about by growing populations,
Key messages Nutrition-sensitive interventions and programmes in agriculture, social safety nets, early child development, and education have enormous potential to enhance the scale and effectiveness of nutrition-specific interventions; improving nutrition can also help nutrition-sensitive programmes achieve their own goals. Targeted agricultural programmes and social safety nets can have a large role in mitigation of potentially negative effects of global changes and man-made and environmental shocks, in supporting livelihoods, food security, diet quality, and womens empowerment, and in achieving scale and high coverage of nutritionally at-risk households and individuals. Evidence of the effectiveness of targeted agricultural programmes on maternal and child nutrition, with the exception of vitamin A, is limited; strengthening of nutrition goals and actions and rigorous effectiveness assessments are needed. (Continues on next page)

increasing incomes, and urbanisationshifts that raise concerns about diet quality and food safety, while threatening water, land, and other finite natural resources.58 In view of these challenges, pro tection of nutrition, let alone acceleration of progress, will entail more than bringing nutrition-specific inter ventions to scale. It will require a new and more aggressive focus on coupling effective nutrition-specific interventions (ie, those that address the immediate determinants of nutrition) with nutrition-sensitive pro grammes that address the underlying causes of under nutrition (panel 19,10). Nutrition-sensitive programmes draw on comple mentary sectors such as agriculture, health, social pro tection, early child development, education, and water and sanitation to affect the underlying determinants of nutrition, including poverty; food insecurity; scarcity of access to adequate care resources; and to health, water, and sanitation services.11 Key features that make pro grammes in these sectors potentially nutrition-sensitive are: they address crucial underlying determinants of nutrition; they are often implemented at large scale and can be effective at reaching poor populations12 who have high malnutrition rates; and they can be leveraged to serve as delivery platforms for nutrition-specific
www.thelancet.com

66

Series

interventions. Nutrition-sensitive programmes might fore help to accelerate progress in improving there nutrition by enhancing the household and community ment in which children develop and grow, and environ by increasing the effectiveness, coverage, and scale of nutrition-specific interventions. Nutrition-sensitive programmes can help protect poor populations from the negative consequences of global cial, food security threats and mitigate the effects of finan weather-related, and man-made shocks (eg, conflicts). Such shocks make poor populations increasingly vul nerable to undernutrition, as shown by food and fuel price crises in the past 6 years,4 and documented effects of conflicts on morbidity and mortality among affected populations.13,14 Climate change and the expected increased frequency of droughts and flooding are likely to reduce food availability and dietary diversity, and increase rates of infectious diseases such as diarrhoea or malaria.15 Under these circumstances, nutrition-sensitive programmes can help to protect the assets and welfare of poor people and their investments in the health, nutri tion, and development of their children. Nutrition-sensitive programmes are likely to affect nutrition through changes in food and non-food prices and income, and through womens empowerment. Panel 216,17 and figures 118,19 and 218,19 show results of analyses of the links between income growth and maternal and child anthropometry and anaemia (appendix p 1). Appendix p 2 summarises evidence regarding the association between womens empower ment and child nutrition. We review evidence of the nutritional effect of pro grammes from different sectors, and discuss how such investments could be made more nutrition-sensitive. We selected sectors on the basis of: relevance for nutrition (eg, address crucial underlying determinants of nutrition); availability of assessments of their nutritional effect; high coverage of poor populations; and targeting (programmes are, or could be, targeted to reach nutritionally vulnerable groups). The two sectors that most closely meet these criteria are agriculture and social safety nets. Early child development programmes do not meet the high coverage criteria but they are included because child development and nutrition outcomes share many of the same risk factors, and there is a growing interest in examination of potential integration and synergies in programming and outcomes.20,21 Schooling is also included, despite failing to meet all criteria, because of the importance of parental education for child nutrition and development. Health, water and sanitation, and family planning are covered in the accompanying report by Zulfiqar Bhutta and colleagues.22 Investments and policies in several other sectors (eg, transportation; communication and infor mation technology; and global food, agriculture, and trade) have the potential to affect nutrition, as do more targeted policies (eg, maternity leave); however,
www.thelancet.com

(Continued from previous page) The feasibility and effectiveness of biofortified vitamin A-rich orange sweet potato for increasing maternal and child vitamin A intake and status has been shown; evidence of the effectiveness of biofortification continues to grow for other micronutrient and crop combinations. Social safety nets are a powerful poverty reduction instrument, but their potential to benefit maternal and child nutrition and development is yet to be unleashed; to do so, programme nutrition goals and interventions, and quality of services need to be strengthened. Combinations of nutrition and early child development interventions can have additive or synergistic effects on child development, and in some cases, nutrition outcomes. Integration of stimulation and nutrition interventions makes sense programmatically and could save cost and enhance benefits for both nutrition and development outcomes. Parental schooling is consistently associated with improved nutrition outcomes and schools provide an opportunity, so far untapped, to include nutrition in school curricula for prevention and treatment of undernutrition or obesity. Maternal depression is an important determinant of suboptimum caregiving and health-seeking behaviours and is associated with poor nutrition and child development outcomes; interventions to address this problem should be integrated in nutrition-sensitive programmes. Nutrition-sensitive programmes offer a unique opportunity to reach girls during preconception and possibly to achieve scale, either through school-linked conditions and interventions or home-based programmes. The nutrition-sensitivity of programmes can be enhanced by improving targeting; using conditions; integrating strong nutrition goals and actions; and focusing on improving womens physical and mental health, nutrition, time allocation, and empowerment.
See Online for appendix

Panel 1: Definition of nutrition-specific and nutrition-sensitive interventions and programmes Nutrition-specific interventions and programmes Interventions or programmes that address the immediate determinants of fetal and child nutrition and developmentadequate food and nutrient intake, feeding, caregiving and parenting practices, and low burden of infectious diseases Examples: adolescent, preconception, and maternal health and nutrition; maternal dietary or micronutrient supplementation; promotion of optimum breastfeeding; complementary feeding and responsive feeding practices and stimulation; dietary supplementation; diversification and micronutrient supplementation or fortification for children; treatment of severe acute malnutrition; disease prevention and management; nutrition in emergencies Nutrition-sensitive interventions and programmes Interventions or programmes that address the underlying determinants of fetal and child nutrition and developmentfood security; adequate caregiving resources at the maternal, household and community levels; and access to health services and a safe and hygienic environmentand incorporate specific nutrition goals and actions Nutrition-sensitive programmes can serve as delivery platforms for nutrition-specific interventions, potentially increasing their scale, coverage, and effectiveness Examples: agriculture and food security; social safety nets; early child development; maternal mental health; womens empowerment; child protection; schooling; water, sanitation, and hygiene; health and family planning services
Adapted from Scaling Up Nutrition9 and Shekar and colleagues, 2013.10

67

Series

Panel 2: How responsive is nutrition to income growth? As economies grow, stunting rates typically decrease, but the predicted decrease is far slower than the corresponding poverty reduction associated with economic growth (figure 1). Country fixed-effects regressions show that a 100% increase in gross domestic production (GDP) per person predicts a 59% (95% CI 4176) reduction in stunting and an 110% (86134) decrease in the World Banks poverty measure of individuals living on $125 per person, per day. The effect of growth in gross national product on nutrition comes from a combination of increased household resources, and improved infrastructure and nutrition-relevant services. Much unexplained variability exists in the effect of national income on stunting. As shown in figure 1, countries such as Guatemala, South Africa, and India have higher stunting rates than expected for their income levels. By contrast, the Dominican Republic, Senegal, Ghana, China, and Sri Lanka are among the best performers. The association between prevalence of child underweight and GDP growth is stronger than for stunting, with the rate of decrease with 100% GDP growth being 70% (95% CI 5388; appendix p 1). This estimate is larger than reported with earlier datasets.16 Anaemia defined as haemoglobin concentrations below 109 g/Ldecreases at a slower rate; a 100% improvement in income would decrease child anaemia by only 24% (1336) and maternal anaemia by 18% (0431). However, severe anaemiadefined as haemoglobin below 70 g/Ldecreases at a much higher rate with income growth for both mothers (65%; 95% CI 4288) and children (90%; 51129).17 Although data for low birthweights are not as reliable as those for other nutritional indicators, estimates using World Bank data suggest that a 100% increase in GDP per person typically reduces low birthweight prevalence by only 23% (95% CI 0841). Bangladesh, India, Sudan, and Haiti have particularly high rates of low birthweight prevalence relative to their national levels of income. For women underweight (body-mass index <185 kg/m2), a 100% growth in national income results in a 40% (95% CI 1758) decrease in underweight prevalence, a rate substantially lower than the reduction in child underweight. The association between national income growth and women overweight and obesity is much stronger than for women underweight: a 100% increase in GDP per person is estimated to increase prevalence of overweight and obesity in women by 70% (95% CI 40100; figure 2). These findings show that when GDP per person increases, prevalence of womens overweight or obesity increases faster than prevalence of women underweight decreases. Again, some countries are clear outliers, with Egypt and several Latin American countries having very high levels and several Asian countries (eg, Vietnam, India, China, Thailand) having lower than expected levels of overweight and obesity in view of their GDP per person.

cost effectiveness studies cannot be easily applied to assess or rank these programmes. Similarly, although costbenefit analysis can be used in principle, this analysis needs a common metric for all outputs, generally in monetary terms. However, a conversion of a death averted into monetary values requires an arbitrary assessment of the value of premature deaths averted. Similarly, although equity is usually deemed socially desirable, its value cannot be easily quantified.25 Therefore, the nutrition outcomes in the programmes we discuss cannot be directly compared with those in the accompanying report by Zulfiqar Bhutta and colleagues.22 grammes we review are However, as we explain, the pro an integral component of an overall strategy to improve global nutrition.

Agriculture
Agriculture systems have a crucial role in provision of food, livelihoods, and income.1 Agriculture is the main occupation of 80% of poor populations in rural areas, including women. In Africa, women account for 70% of agricultural labour and 80% of food processing labour.26,27 Growing concerns about how to meet the food needs of an estimated global population of 9 billion by 2050 have spurred renewed efforts to boost agriculture production and productivity in the face of increasing threats that affect the global food system. Agriculture growth has been shown to reduce undernutrition;28 an additional investment of US$8 billion per year globally would reduce the number of underweight children by 10 million and of hungry people by 201 million by 2050, and raise the income of many of the worlds poorest people.28 Moreover, the economic returns to investments in agriculture are high compared with many other economic investments.29 Although investments to enhance agriculture produc tivity and boost global food supply are crucial for longterm reductions in poverty, hunger, and malnutrition, they might not solve the problem of scarcity of access to nutritious and diverse diets (as opposed to scarcity of calories) that poor people face. A new emphasis on making agricultural systems and food and agriculture policies more nutrition-sensitive is called for and several reports discuss approaches and instruments to do so.1,3033 An approach that can complement efforts to raise agri cultural productivity and food supply globally is targeted agricultural programmes aimed at enhancing poor households income and access to high-quality diets. Our review focuses on these types of programmes, most specifically homestead food production systems, and the biofortification of staple crops, because they both meet our selection criteria, with the exception of scale. Targeted agricultural programmes can affect nutrition through several pathways (panel 334,35). Despite variations in the way researchers use these pathways, all concur that women their social status, empowerment, con trol over resources, time allocation, and health and nutritional statusare key mediators in the pathways
www.thelancet.com

we excluded these sectors because of the absence of assess ments of nutritional effects. Consistent with the Maternal and Child Nutrition Series, we focus on adolescent girls and women, infants, and young children during the first 1000 days of life (period from conception to a childs second birthday). Interventions to improve nutrition and child development during this period have high rates of return because of their importance in enhancing economic productivity later in life23 fostered by a combination of improved health, nutrition, and cognition, which lead to more schooling, higher-paying jobs, and overall enhancement of physical, cognitive, and reproductive performance.24 The programmes we reviewed generally have several objectives, including improving income, food security, womens empowerment, and nutrition. For this reason
68

Series

Stunting prevalence in children aged 05 years

between agriculture inputs, intra-household resource allocation, and child nutrition.26,30,3438 The recognised importance of development of new approaches to stimulate agricultures contribution to nutri tion has led to an increased interest in examination of the so far untapped potential of leveraging value chains to improve nutrition. Since this approach is still at a nascent stage, experience and evidence of effectiveness are scarce. Panel 43946 provides a brief overview of the approach.

60
Niger ETH DRC

Burundi Afghanistan MDG Nepal BGD SDN PAK NGA CIV Haiti VTN KEN Ghana Senegal PHN IDN China IRQ Syria Egypt Algeria Thailand Colombia DOM Turkey Brazil Belarus Peru South Africa Ecuador Botswana Yemen Guatemala Angola

India

40

MMR

Home gardens and homestead food production systems


Several scientific literature reviews of homestead food production systems have been done in the past decade.34,38,4752 These reviews focused on different types of programmes and nutritional outcomes and used different search strategies and inclusion and exclusion criteria. Despite these differences, key conclusions were largely consistent across all reviews (appendix pp 34). First, these reviews note that there is little evidence of effectiveness of homestead food production programmes on maternal or child nutritional status (anthropometry or micronutrient status), with the possible exception of vitamin A status. For child anthropometry, a few studies reported an effect on at least one indicator,5357 but effects were generally small. Although meta-analysis might not be the method of choice for synthesis of evidence from such diverse programmes, the results of a four-study meta-analysis52 showed no overall effect of targeted agricultural pro grammes on underweight, wasting, or stunting. Another four-study meta-analysis51 for vitamin A status, however, reports a small overall difference in serum retinol between intervention and control areas (008 mol/L); a clusterrandomised effectiveness assessment of a biofortified orange sweet potato intervention in Uganda also showed a 95 percen tage point reduction in the prevalence of low serum retinol (<105 mol/L) in intervention compared with control children aged 35 years at baseline.58 The second con sistent message is that nutritional effect is more likely when agriculture interventions target women and include womens empowerment activities, such as improvement in their knowledge and skills through behaviour-change com munications or promotion of their increased control over income from the sale of targeted commodities. No studies, however, have specifically com pared targeting of men versus women, or main streaming gender versus not doing so in the programmes reviewed. The third key message is that, with the exception of two studies of biofortified orange sweet potato,58,59 impact evaluation studies have generally been too poor and sample sizes often too small to draw definite con clusions about effects on nutritional status. One review,38 which specifically looked at effects of home stead food production systems on intermediary outcomes along the impact pathway, concluded that, when measured, positive effects are shown for several underlying determinants of nutrition, including house hold production and consumption, maternal and child
www.thelancet.com

Predicted stunting line


Kazakhstan

20

Kyrgyzstan UzbekistanSri Lanka Moldova

Predicted poverty line 0 0 2000 4000 6000 GDP per person (2005 international $) 8000

10 000

Figure 1: Prevalence of stunting in children aged 05 years and GDP per person Most observations for prevalence of stunting are from 200008. The fitted curves are locally weighted regressions of prevalence of stunting in children aged 05 years and poverty (<$125 per person, per day), against GDP per person. The adjustment to international dollar units converts income expressed in nominal dollars to one that is expressed in terms of international dollars, which have the same estimated purchasing power as a dollar in the USA, accounting for local prices. The size of the circles represents the estimated population of stunted children aged 05 years, in about 2005, on the basis of multiplication of stunting prevalence by UN estimates of the population of children aged 05 years. Data are sourced principally from the Demographic and Health Surveys,18 with observations for some countries sourced from WHO.19 GDP=gross domestic product. BGD=Bangladesh. CIV=Cte dIvoire. DOM=Dominican Republic. DRC=Democratic Republic of the Congo. ETH=Ethiopia. IDN=Indonesia. IRQ=Iraq. MDG=Madagascar. MMR=Myanmar (Burma). KEN=Kenya. NGA=Nigeria. PAK=Pakistan. PHN=Philippines. SDN=Sudan. VTN=Vietnam.

80
Egypt

Women overweight prevalence, BMI>25 (%)

Jordan

Bosnia-Herzegovina Peru Turkey Serbia South Africa Brazil

60
Bolivia Nicaragua Lesotho Honduras Guatemala Swaziland

40

Sierra Leone

Ghana Cameroon Kenya

Morocco Kyrgyzstan Moldova

Armenia

Dominican Republic Thailand

20

Nigeria Pakistan

Philippines Indonesia China

Congo (DRC)

Bangladesh Ethiopia

India Vietnam

2000

4000 6000 GDP per person (2005 PPP$)

8000

10 000

Figure 2: Prevalence of women overweight (BMI>25) and GDP per person, for low-income and middle-income countries Most observations for prevalence of women overweight are from 200010. The fitted curve is a locally weighted regression of prevalence of women overweight against GDP per person. The correlation between prevalence of women underweight and the log of GDP per person is 071 and is significant at the 1% level. The size of the circles represents the estimated population of overweight women aged 1549 years, in about 2005, on the basis of multiplication of prevalence of women overweight by the UN population estimates of the female population aged 1549 years. Data are sourced principally from the Demographic and Health Surveys18 and WHO.19 DRC=Democratic Republic of the Congo. GDP=gross domestic product. PPP=purchasing power parity.

69

Series

Panel 3: Pathways by which agriculture can affect nutrition outcomes As a source of food: increases household availability and access to food from own production As a source of income: increases income from wages earned by agricultural workers or through the marketing of agriculture commodities produced Food prices: agricultural policies (national and global) affect a range of supply and demand factors that establish the price of marketed food and non-food crops; this price in turn, affects the income of net seller households, the purchasing power of net buyers, and the budget choices of both Womens social status and empowerment: womens participation in agriculture can affect their access to, or control over, resources and assets, and increase their decisionmaking power regarding intra-household allocation of food, health, and care Womens time: womens participation in agriculture can affect their time allocation and the balance between time spent in income generating activities and time allocated to household management and maintenance, caregiving, and leisure Womens own health and nutritional status: womens participation in agriculture can affect their health (eg, through exposure to agriculture-associated diseases) and nutritional requirements (eg, through increased energy expenditure); their health and nutritional status can, in turn, affect their agricultural productivity and hence their income from agriculture
Adapted from the World Bank34 and Gillespie and colleagues, 201235

Panel 4: Value chains for nutrition Food supply chains are defined as the series of processes and actors that take a food from its productionincluding inputs into productionto consumption and disposal as waste.41 Broadly defined steps along the supply chain include production, processing, distribution, retailing, promotion, labelling, and consumption. The concept of value chain refers to the addition of value (usually economic) for chain actors at different steps along the chain. In the past 5 years, value chains have been singled out as one potential strategy to leverage agriculture to improve nutrition.3941 The approach could be particularly relevant for traditional value chains for micronutrient-rich foods such as dairy, meat, fish, poultry, and fruits and vegetables, which are generally lacking in the diets of low-income households because of scarce availability, perishability, and high prices often compounded by a scarcity of information and knowledge about their health and nutritional benefits. Food value chains are therefore a possible entry point to stimulate both supply and demand (especially among poor populations) for micronutrient-rich foods. Value chain concepts and analysis have unique features that make them a promising approach for tackling both undernutrition and overnutrition: 1) they focus on coordination between actors, because all value chain processes and actors are tightly linked by each action affecting the others along the chain; 2) they are analytical, versatile, and solution-oriented and can therefore be used to assess the constraints that affect availability, affordability, acceptability, or quality of nutritious foods in a given context, and identify and test solutions that can be implemented at specific leverage points along the chain; 3) they focus on addition of economic value, and could therefore be used to identify points before, during, and after production at which nutritional (and economic) value could be added, or losses in nutrients prevented. In view of the importance of coordination across sectors and of development of joint solutions to stimulate agriculture and nutrition linkages, value chain concepts and analysis might provide a useful framework and platform to achieve these goals. (Continues on next page)

intake of target foods and micronutrients, and overall dietary diversity. This finding is consistent with results from an impact-pathway focused assessment of a home stead food production system in Cambodia, which showed no effect on child anthropometry or anaemia despite effects on household production, consumption, and dietary diversity.60 Despite explicit targeting of women in many agri cultural programmes, few studies have measured specific aspects of womens empowerment as a pathway to improved nutrition, and results are mixed. Assess ments of home stead food production systems in Bangladesh and Nepal report positive effects on womens income, control over resources, or influence in decision making on a range of issues.53,6163 In Kenya, a project promoting orange sweet potato production among women farmers showed that women gained control over selling the product, whereas men maintained control over income.64 Livestock and dairy projects in Kenya and Bangladesh report increases in womens income or influence in decision making,65,66 whereas in India, mens but not womens income improved as a result of a dairy project.67 Very few studies have measured the effect of agriculture interventions on womens time, know ledge, practices, health, or nutritional status and none have modelled the potential mediating role of these maternal resources on child nutrition.38

Biofortification
Biofortification is a uniquely nutrition-sensitive agri culture intervention because it focuses on breeding of staple crops that are rich in essential micronutrients.68 The many advantages of the approach are well docu mented.68,69 Bio fortification, however, cannot achieve the high con centrations of micronutrients needed to treat severe deficien cies or to fulfil the high requirements (eg, for iron and zinc) of pregnant and lactating women and infants; it is more suited for provision of a daily dose of micronutrients (about 50% or more of daily needs) to help prevent deficiencies in individuals through out the lifecycle, outside of the 1000 days window. As is true for all approaches, biofortification should be considered as one component of a larger strategy to eliminate micro nutrient deficiencies, and the optimum mix of supple mentation, dietary diversi fication, fortifi cation, bio forti fication, and health services should be defined depending on local context. Three broad milestones need to be achieved for bio fortification to succeed: 1) breeding objectives (mini mum nutrient) must be met; target concentration for each micro 2) retention and bioavailability of micro nutrients must be satisfactory so that intake leads to expected improvements in status; and 3) farmer adoption rates and intakes by target populations must be adequate. HarvestPlus, a programme that has led a global effort to breed and disseminate biofortified staple food crops since 2003, has made substantial progress in research to test these three steps for vitamin A, zinc, and iron in seven crops: cassava,
www.thelancet.com

70

Series

maize, sweet potato, bean, pearl millet, rice, and wheat (table). In addition to achieving major progress in breeding and releasing crops, the programme will complete all planned retention and bioavailability studies in 2013, and 11 efficacy trials in 2014. Two effectiveness trials,58,59 which assessed the rollout of orange sweet potato in Uganda and Mozambique (milestone three), have been completed. They showed high farmer adoption and sig nifi cant increases in vitamin A intakes in both countries and in child vitamin A status in Uganda.58,59 Effectiveness trials for the other target crops are expected to be completed by 2018. Thus, present evidence regarding biofortification is concentrated on the first two milestonesproof of concept that breeding for micronutrient-rich crops is feasible and that micronutrients are retained and bio availableand a growing evidence of efficacy. Results on bioconversion of carotene to retinol in humans and bioavailability of zinc and iron from biofortified com pared with common varieties are very encouraging, suggesting that extra minerals will lead to net increases in quantities absorbed. Efficacy studies also confirm that intakes of iron-biofortified rice70 and beans71,72 improve iron status, and that all biofortified crops released so far have favourable agronomic qualities, including equal or higher yields than common varieties, and greater disease resistance and drought tolerance. Evidence regarding the effective ness of biofortification, however, is still confined to vitamin A in orange sweet potato, and the scalability of delivery is yet to be shown.

(Continued from previous page) Value chains also have important limitations. First, they focus on one food at a time, as opposed to the whole diet and the many nutrients required for healthy living. Efforts to integrate nutrition into value chains should therefore focus on complementary value chains to fill the specific dietary and nutrient gaps identified in target populations. Second, the focus on addition of economic value plus incorporation of nutrition goals might create insurmountable trade-offs for value chains actors. Third, although they might be well-suited to enhance access to micronutrient-rich foods for girls and women during the reproductive period, their role for addressing the special needs of young children might be limited to fortified complementary foods or products, and a few target foods such as dairy products and biofortified crops (eg, biofortified orange sweet potatoes). Case studies4143 and a review of on-going programmes44 suggest that several research initiatives and value chain actors are currently exploring the potential of value chains to improve nutrition. One such initiative is homegrown school feeding programmes, which use value chains to link agriculture and nutrition, with potential livelihood and income benefits for farmers and nutrition benefits for young children and their families.45,46 Existing efforts to incorporate nutrition in value chains should also consider addressing food safety issues, especially since most of the micronutrient-rich foods of interest are also highly perishable and susceptible to food safety problems. Tackling of food-borne diseases would improve nutrition.

Social safety nets


Social safety nets are programmes that distribute transfers to low-income households. These programmes raise income among vulnerable groups and enhance resilience by preventing destitution brought about by loss of assets or reduced investment in human capital during times of crises. Transfers can be in the form of
Country (year of first release)*

cash or food, although with improved technology for tracking income transfers, cash transfers are increasingly the preferred means to support chronically poor house holds. Between 075 and 10 billion people in lowincome and middle-income countries currently receive cash support.73 Although many transfer programmes reach only a small share of the vulnerable population, some have extensive coverage, such as Ethiopias Productive Safety Net Programme, which reaches 10% of the countrys population,74 and transfer programmes in Brazil and Mexico that reach 25%, and in Ecuador 40%, of their populations.75 The generosity of transfers varies widely, ranging from transfers that increase total income marginally to those that boost income by up to a third for the poorest recipients.75

For more on the HarvestPlus programme see http://www. harvestplus.org

Status of nutrition studies Dietary intake and retention Bio-availability Efficacy Effectiveness

Vitamin A crops (released) Cassava Maize Iron crops (released) Bean Pearl Millet Rice Wheat Rwanda (2012) India (2012) Bangladesh and India (2013) India and Pakistan (2013) P P P P P P 2013 P Continuing P 201314 201314 201315 201416 201416 Nigeria, Democratic Republic of Congo (2012) Nigeria, Zambia (2012) P P P P P P 201314 Continuing P 201315 201315 P

Orange sweet potato Uganda (2007), Mozambique (2002)

Zinc crops (under developmentto be released in 2013)

References are provided in appendix pp 56. *Approved for release by National Governments after intensive multi-location testing for agronomic traits and micronutrient performance. Completed though not necessarily reported.

Table: Release schedule for biofortified crops and status of related nutrition studies

www.thelancet.com

71

Series

The main goal of social transfers is to augment income, but programmes sometimes include additional inter ventions or conditions that can enhance their nutrition sensitivity such as: linking of transfers to health and nutrition services (eg, through conditionality); targeting of households with nutritionally vulnerable members, on the basis of age or physiological status; inclusion of nutrition-specific interventions for selected individuals within the household (eg, nutrition behaviour-change communi cations or distribution of fortified foods or supplements); administration of transfers in a sexsensitive manner (eg, by directing transfers to women or designing them to accommodate time constraints of caregivers); and targeting of populations facing climatic or economic stress related to seasonality or other shocks, or focusing on emergencies.

Conditional cash transfers


Conditional cash transfers aim to stimulate households to invest in the health, nutrition, and education of their children (enhancing human capital) by promotion of the use of these services as conditions (conditionalities) for receipt of transfer. Most conditional cash transfers target transfers to women, on the premise that increasing womens control over resources will lead to greater investments in children (appendix p 2). Although conditional cash transfers are implemented worldwide, experimental evidence of effectiveness comes mostly from Latin America. In addition to their positive effects on poverty reduction, household food consumption, and dietary diversity,76,77 almost all programmes assessed increased the use of preventive and curative health and nutrition services.78,79 The Mexico, Brazil, and Nicaragua programmes8083 also showed improvements in womens control over additional resources, enhanced self-esteem, heightened knowledge and awareness of health and nutrition, and increased opportunities for women to strengthen their social networks. Despite the many benefits of conditional cash transfers for households and women, evidence of effects on nutritional outcomes is mixed.76,78 A review77 using pooled estimates shows that, overall, conditional cash transfers have had a small, but not statistically significant, effect on child anthropometry. A forest plot analysis of 15 programmes,77 combining conditional cash transfers and unconditional cash transfers, shows an average effect of 004 in height-for-age Z score, an effect size that is neither statistically significant nor biologically meaning ful; similarly, no significant effect was identified for conditional cash transfers only. In view of the heterogeneity of populations and pro gramme designs and methods, meta-analyses might not be the most appropriate approach for assessment of effect, but analyses of individual studies are consistent with the findings. Only a few conditional cash transfer studies show effects on anthropometry, and these effects are shown in the youngest or poorest children, or those
72

exposed to the programme for long durations.76,77,84 Evidence of effects on micronutrient nutrition is equally scant and comes from only a few studies that have looked at these outcomes.76 The Mexico conditional cash transfer programme, which distributed a micronutrient-fortified food to beneficiary mothers and children, showed a positive effect on child intake of iron, zinc, and vitamin A among those who consumed the product, but only a small effect on mean haemoglobin or anaemia reduc tion.76 Two other programmes, in Honduras and Nicaragua, that assessed effect on haemoglobin showed no effect.76 The Mexico programme showed reductions in low birthweight attributed to changes in womens empower ment, which in turn were attributed to womens increased demand for better quality prenatal care as a result of participation in the programme.85 Evidence is also emerging of small effects of conditional cash trans fers on child develop ment outcomes.86

School feeding programmes


School feeding programmes are a type of conditional transfer, albeit in kind. Similar to other transfers, they are mainly a form of social assistance for consumption. The links to nutrition are less direct than transfers targeted to mothers and children during the first 1000 days, but school feeding can reduce hunger and stimulate learning.87 These programmes, however, are implemented in nearly every country in the world.88 Results from a meta-analysis show that school feeding programmes have small effects on school-age childrens anthropometry, particularly in low-income settings.89 Major effects on height are not expected in school-age children and weight gains can be either positive (in under weight populations) or negative (when risks of obesity are high). In middle-income countries, school meals might also serve as an opportunity to combat obesity; Brazil and Chile have redesigned their pro grammes with this risk in mind.90 School meals might also benefit other members of the household when the food provided is shared or when the school-aged childs intake at home is reduced.91,92 Randomised controlled trials in Burkina Faso and Uganda showed effects on weight among preschoolaged boys (ie, <5 years) whose siblings received school meals or take-home rations compared with a control group93 (Gilligan D, International Food Policy Research Institute, personal communication). Another oppor tunity offered by school feeding programmes is to affect iron nutrition, especially for adolescent girls. A review92 of randomised evaluations of iron-rich school meals (fortified or providing animal-source foods) documents that three of four studies improved iron status, irre spective of initial status. The addition of a micro nutrient mix to school meals in India improved total body iron, but not anaemia, possibly because of worm loads.94 Deworming can, however, be included as part of a larger school health programme, although the timing of
www.thelancet.com

Series

delivery differs from the daily meal programme. Since school-aged children are the main reservoir of worm loads in a population, such an intervention could benefit younger children as well.

Panel 5: Unintended, negative effects of in-kind and cash transfers in Mexico Social safety nets can reduce poverty and increase use of health and education services.79 Depending on their target populations, however, these interventions can have unintended negative consequences. The effects of Mexicos Programa de Apoyo Alimentario (PAL; Food Support Programme) on excess weight gain in women is one such example. PAL is a dual conditional cash and in-kind transfer programme targeted to poor and remote communities in rural Mexico. An assessment of PAL99 showed improved household dietary quality, but also increases in total energy consumption in a population that was not energy-deficient and had a high prevalence of overweight and obesity among women at baseline (65%). PAL also increased the already steep annual weight gain in adult women in the control group (425 g, SD 80) by 291 g (111) per year in the food basket group (a 68% increase) and by 222 g (122) per year in the cash group (a 52% increase). The most substantial effect was recorded in adult women who were already obese before the programme started (518g [153] per year in the food basket group and 354 [169] per year in the food basket group; figure 3).100 The PAL programmes food basket included several energy-dense staple and basic food products including oil, cookies, and whole milk, and provided an additional 450 kcal per day per adult. To avoid negative effects on populations experiencing the double burden of child stunting and adult obesity, transfer programmes should be designed to respond to the identified needs of target populations, and for food transfers, their specific nutrient gaps.

Unconditional transfers
Unconditional transfers, either as cash or in kind, have also been popular, particularly outside of Latin America. Households commonly spend more on food and health with cash transferseven when they are only indirectly linked to nutrition and healththan they spend out of other increases in income.95,96 Moreover, some uncon ditional cash transfers use so-called soft conditions in the form of broadly targeted behaviour-change com muni cations or social marketing to encourage health-seeking behaviour. In Africa, soft conditions or uncon ditional cash transfers are more common than conditional cash transfers.97 One randomised trial showed that a con ditional cash transfer with health con ditionality increased clinic visits in Burkina Faso, where as an unconditional cash transfer did not,98 showing the importance of the condition for achieving health-seeking behaviour changes in this setting. Evidence, however, shows an absence of overall effect of both unconditional cash transfers and conditional cash transfers on child nutritional status.77 Costs will differ between the two approaches as will the distribution of benefits for different outcomes; as such, no method dominates in all situations.

7000 Estimated eect on annual weight gain (g) 6000 5000 4000 3000 2000 1000 0

Food basket

Cash

* *

In-kind household food distributions


In-kind household food distributions are currently less prominent than they were in previous decades, mostly because of cost considerations. They are now largely used as part of an emergency response or in places where the logistics of cash transfers are constrained. Evidence from Mexico suggests that in-kind food transfer programmes might have unintended effects on overweight and obesity when the energy contribution of the food basket exceeds the energy gap in the targeted population (panel 5,99,100 figure 3). In addition to general family rations, food distribution programmes often provide micronutrientfortified foods (eg, corn-soy or wheat-soy blend) to mothers and young children. In Haiti, distribution of such food rations to all mothers and children within the first 1000 days of life had a greater effect on child growth than did targeting of underweight children younger than 5 years.101 In view of the severity of food insecurity in this popu lation, no unintended effects on overweight or obesity were identified. Complementation of this pro gramme with the distribution of iron-fortified micro nutrient powders reduced anaemia prevalence by half in as little as 2 months.102

All women

<25

25 to 30 BMI group

>30

Figure 3: Estimated effect of Mexicos Programa de Apoyo Alimentario programme on annual weight gain in women, by initial BMI BMI=body-mass index. *p<005. p<001.

Transfer programmes in emergencies


Transfer programmes in emergencies also usually combine nutritionally enhanced complementary foods for pregnant and lactating women and their young child with
www.thelancet.com

family rations or cash. Disasters, particularly sudden onset emergencies such as earthquakes and hurricanes, often disrupt normal market channels, which might dampen the logistical advantages of cash compared with food transfers. Although food aid deliveries overall declined from 15 million metric tonnes (t) in 1999 to 41 million t in 2011, emergency deliveries have remained almost constant; they now account for more than 67% of total food aid.103 Even when targeted towards overall household subsistence, aid during disasters can prevent major deteriorations in child undernutrition.104,105 Agebased targeting of fortified foods can help prevent under nutrition and complement efforts to tackle cases of severe acute malnutrition with specially formulated products. The nutritional effects of emergency deliveries can be
73

Series

enhanced by inclusion of lipid-based nutrient supplements in the package of assistance to families.106108

Early child development


Stunting and impaired cognitive development share several of the same risk factors, including deficiencies in protein, energy, and some micronutrients, intrauterine growth retardation, and social and economic conditions, such as maternal depression and poverty.109 Some of the key phases of brain growth and development also encompass the first 1000 days of life, the period of peak susceptibility to nutritional insults. Therefore, some key interventions can protect children from both nutritional and developmental risks; these include core maternal and child nutrition interventions, psychosocial stimu lation and responsive parenting, and interventions to alleviate poverty, food insecurity, maternal depression, and gender inequity.21 Evidence of the effect of early child development inter ventions, with or without a nutrition component, on child development outcomes has been extensively reviewed in two previous series in The Lancet.20,21 We focus on evidence of how child stimulation and nutrition interventions can have complementary effects on nutrition outcomes (appendix pp 79). The most compre hensive, long-term study110 of interventions that provided both child stimu lation and food supplemen tation to stunted children aged 924 months in Jamaica showed an additive effect of the two interventions on cognitive development, but not on growth. At adoles cence, the additive effects on cognition were not sus tained, but the group having received stimulation had long-term benefits, which ranged from improved develop ment outcomes to educational attain ment and social behaviour.111,112 In Bangladesh, addition of stimu lation and home visits to standard nutrition and health care for severely malnourished children improved development outcomes and weight-for-age Z score (WAZ).113 Another trial in Bangladesh, which added responsive parenting (includ ing feeding) to an informal nutrition and child development education programme, showed benefits on several feeding and parenting behaviours, child self-feeding, and develop ment out comes; addition of iron-fortified micronutrient powders to the intervention improved weight gain and WAZ but had no additional effect on development outcomes.114 A zinc supple men tation and responsive stimulation inter vention in under weight children in Jamaica showed synergistic effects on child development between the two inter ven tions: greater benefits on development outcomes were identified in the zinc and stimulation group, compared with little or no effect in the groups receiving either.115 This synergistic effect, however, was not noted for morbidity (reduced only in zinc group) or growth (no effect in either group). A randomised controlled trial of cash transfers to households linked to preschool enrolment in Uganda provides an example of joint benefits on cog nition and nutrition; findings showed significant effects of cash
74

transfers on child cognitive development, which were mediated by increased preschool partici pation, improved diets, and reduced anaemia (Gilligan D, Roy S, Inter national Food Policy Research Institute, personal communication). Other trialsin India, Pakistan, and Bangladesh116118 although successful at improving child development or nutrition outcomes, or both, failed to show additive or synergistic effects between nutrition and stimulation interventions. In India,116 beneficiaries of the Integrated Child Development Services (ICDS) programme were allocated to groups receiving breastfeeding and comple mentary feeding counselling, or this package plus responsive feeding and psychosocial stimulation skills. Compared with ICDS only, both intervention packages improved child dietary intake and haemoglobin and reduced morbidity, but only the nutrition intervention ing increased length gain, and only the full package includ stimulation benefited development outcomes.116 A factorial design trial in Pakistan117 showed no evidence of additive or synergistic effects of a nutrition (counselling and micronutrient powders) and stimulation intervention (monthly group meetings and home visits for children aged 024 months) on child development or nutrition outcomes. Preliminary results show effects of all three intervention packages on developmental scores com pared with control, with larger effect sizes among the two stimulation groups. In Bangladesh,118 psychosocial stimulation with or without food supplements among severely underweight children aged 624 months on discharge from hospital had an effect on mental development and a small effect on WAZ, but no additive or synergistic effects were noted between the two interventions. One intervention that closely ties feeding practices with child stimulation is responsive feeding. Few studies of this approach, however, have been designed to distinguish messaging on complementary feeding from those on psychosocial care114 and few so far have shown a clear association with nutritional outcomes.119 Reduction of maternal depression is another way to address risk factors common to both nutrition and child development.120,121 Efforts are being made to link basic health services with a wide range of social support for women. Efficacy trials.122,123 show that the benefits might accrue to both mothers and their newborn babies, justifying ongoing efforts to bring these initiatives to scale.

Schooling
Although children are beyond the crucial 1000 days window when they enter school, their schooling experience might be a strong determinant of the nutrition of the next generation. Parental schooling has been consistently associated with child nutritional status, with maternal education often, but not always, having a larger explanatory power than paternal education, controlling for income and schooling choices.124,125 The positive global trends in schooling are, therefore, encouraging for
www.thelancet.com

Series

nutrition. Data from developing countries show an average increase in years of schooling from 260 to 762 for boys and from 150 to 664 for girls between 1950 and 2010.126 The girl to boy ratio shows a substantial improvement over this period, from 577% to 859%. Enrolment data show parity in girls primary schooling in most countries;127 moreover, in many countries more girls than boys are now in secondary school. Still, only about a fifth of adolescent girls in sub-Saharan Africa and two fifths of girls in south Asia are enrolled in secondary education.128 We assessed the level of parental education necessary for a meaningful reduction in child undernutrition by analysing 19 datasets from the Demographic and Health Survey (collected since 1999) and derived estimates of the risk of child stunting associated with maternal and paternal primary and secondary education, controlling for household wealth, rural versus urban residence, and child age and sex.The analysis showed that the risk of stunting is significantly lower among mothers with at least some primary schooling (odds ratio [OR] 089, 95% CI 085093), and even lower (p<0001) among mothers with some secondary schooling (075, 071079). Paternal education at both the primary and secondary levels also reduced the risk of stunting although the respective ORs (096, 093101; and 085, 081089) are smaller than for maternal schooling. Despite this overall association, there is appreciable heterogeneity in effect sizes for both maternal and paternal education in individual countries, probably indicative of differences in both quality of education and quality of data. Schooling directly increases individual earnings and national income and, through these pathways, can affect nutrition in the long term. Thus, programmes to increase schooling via the supply of inputs or through fee waivers or cash transfers can be expected to reduce the risk of undernutrition for the next generation. There is a lack of clarity, however, about which aspects of schooling, beyond the income effect, benefit nutrition. At least five overlapping pathways have been suggested, but not formally tested. Schooling might: 1) transmit information about health and nutrition directly; 2) teach numeracy and literacy, thereby assisting caregivers in acquiring information and possibly nutrition know ledge;129,130 3) expose individuals to new environments, making them receptive to modern medicine; 4) impart self-confidence, which enhances womens roles in decision making, and their interactions with health-care professionals; and 5) provide women with the opportunity to form social networks, which can be of particular importance in isolated rural areas. The question remains as to whether schooling could do even more to directly affect nutrition, both in the short term for school children and in the long term as they transition into their parental role. Although nutrition modules are available in some school health education programmes,131,132 assessments of the effect of a school
www.thelancet.com

health and nutrition curriculum in developing countries on undernutrition or health knowledge, let alone on parenting skills decades later, are absent. Schools are also suitable venues to introduce pro grammes to combat obesity. Such programmes can focus on healthy diets and promotion of physical activity. A systematic review133 of 22 studies in low-income and middle-income countries noted that 82% of such programmes had a favourable effect on physical activity, diet, or both.

Discussion
In 2008, The Lancet Maternal and Child Undernutrition Series included conditional cash transfer programmes and dietary diversification approaches as general nutrition support strategies, and noted small positive effects of conditional cash transfers on child anthrop ometry in three Latin American countries, and an absence of a statistically significant of effect of dietary diversification strategies on child nutrition outcomes.134 In the present series, we discuss evidence regarding the nutritional contribution of programmes in four sectors and the potential for enhancing their nutrition-sensitivity. Although the concept of nutrition sensitivity is not new, investments in development and implementation of nutrition-sensitive programmes have intensified in the past few years, prompted by the 2008 series, and spearheaded by the Scaling Up Nutrition movement.9 It is important to recognise, when interpreting the results of our review, that most of the programmes included were retrofitted and tagged as nutrition sensitive without having been originally designed as such. Targeted agricultural programmes have an important role in supporting livelihoods, improving household food security and healthy diets, and in fostering womens empowerment. Yet, our review shows inconclusive evi dence of effects on child nutritional status, with the possible exception of benefits on vitamin A intake and, to a lesser extent, vitamin A status. These findings are probably the result of a combination of factors, including: weaknesses in programme design and implementation (especially the nutrition, behaviour-change communi cation, and health components);60 inclusion of house holds with children outside of the 1000 days window with little potential to benefit in linear growth; and the fact that other pressing constraints to nutritionsuch as infec tious diseases, helminths, and environ mental enter opathy associated with scarcity of access to appropriate water, sanitation, and hygienemight not be addressed by the programmes. Additionally, the assessments of most of the programmes we reviewed had crucial weaknesses such as an absence absence of valid comparison and control groups, a possibly too-short duration of intervention, small sample sizes, the inclusion of the wrong age group in effectiveness assessments, and the failure to control for potential confounding factors in the analysis. All these assessment

For more on the Scaling Up Nutrition movement see http:// scalingupnutrition.org

75

Series

Panel 6: Research priorities Rigorous, theory-based effectiveness and cost-effectiveness assessments of complex and large scale nutrition-sensitive programmes. These assessments should include: Use of experimental randomised controlled trials, where feasible, to test different methods of delivery and joint packages of interventions. Careful assessment of programme impact pathways and quality of service delivery, use of process evaluation instruments and mixed methods, including assessment of the capacity and efficiency of front-line health workers. Measurement of gender-disaggregated impact indicators; these indicators should be carefully selected on the basis of nutrition goals and interventions included in programmes and could include: anthropometry, micronutrient status biomarkers, child development outcomes, child morbidity. Measurement of intermediary outcomes along the impact pathway (eg, household consumption; food security and dietary diversity; dimensions of womens empowerment, maternal physical, and mental health; detailed dietary intake or simpler measures such as dietary diversity for target individuals). Detailed costing for assessment of cost-effectiveness. Development of methods to allow comparison of the social benefits of complex programmes with many objectives and joint outcomes, with the benefits of single-outcome programmes. Formative research and focused ethnographic studies to guide selection, design, and implementation of nutrition interventions to be integrated in nutrition-sensitive programmes, and for overall design of nutrition-sensitive programmes. Qualitative research to understand barriers to participation, adoption and use of programme inputs and services (eg, agricultural inputs, compliance with conditions in conditional transfers, recommended feeding or caregiving practices). Research to rigorously test the feasibility and desirability of integration of interventions from several sectors versus co-location. Such research would establish whether programme implementers should develop new instruments and methods for joint planning, implementation, monitoring, and assessment, or whether investments should focus on effective programme co-location and implementation. Research to test and document the scalability of newly released biofortified crops. Assessment of effectiveness large-scale programmes combining early child development and nutrition interventions in different contexts and assessment of synergies both in programming and outcomes. Research to test different delivery platforms for programmes to reduce maternal depression. Research to test different delivery systems for reaching adolescent girls (eg, school programmes, social safety nets with conditions to keep girls in school, agricultural programmes targeting adolescent girls at home). Assessments of school nutrition programmes and their short term effect on knowledge of school children and long-term effect on parenting skills.

design flaws reduce the ability to detect an effect even if one exists. In view of the complexity and diversity of agricultural programmes, their many goals, and their long impact pathways, some argue that impact assessments should focus on outcomes such as food security and diet quality, rather than child nutritional status during the 1000 days window. Although complexity is a valid concern, the many pathways by which agriculture can improve nutritionand evidence that effects on several indicators along these pathways are achievedsupport giving these programmes a fair chance to deliver on child nutrition
76

outcomes. Future work should include testing of pro grammes with stronger designs, nutrition goals, and interventions; use of rigorous programme-theory based impact and impact pathway assessments; and assessment of cost and cost-effectiveness (panel 6). Biofortification has made substantial progress by establishing proof of concept for orange sweet potato and showing effects, through rigorous assessments, on maternal and child intake of vitamin A and child vitamin A status. The challenges for biofortification now rest in showing effectiveness of new crops, refining delivery and marketing strategies, scaling up successfully, and integrating new varieties into national agricultural research systems. Social safety nets are a powerful way to reduce poverty, and currently provide income support to a billion chronically poor individuals and to shock victims. They have been shown to improve household food availability and dietary quality and to foster certain aspects of womens empowerment, and for conditional cash transfers in particular, to stimulate demand for health and education services. Despite these many benefits, pooled evidence shows little effect on nutritional outcomes. By contrast with agriculture programmes, several effectiveness assess ments of social safety nets have used rigorous random ised controlled trial designs, although most of these studies were done in middle-income countries and might under estimate the magnitude of effect that could be achieved in poorer settings. Additionally, some aspects of programme design or implementation might have diluted their nutri tional effect, including poor timing and short duration of maternal and child exposure, absence of clear nutrition goals, and poor selection or implementation of nutrition interventions in some programmes. Also, the gap between increased use of health and nutrition services and nutrition benefits has been attributed, at least partly, to the poor quality of services provided.78 Conditional cash transfers are designed to increase health awareness and service demand, but ultimately their nutritional effect rests on the quality of public health services. Our review of early child development interventions provides little evidence that stimulation alone has a direct effect on nutrition outcomes, but it suggests that combined early child development and nutrition inter ventions can have additive or synergistic effects on development outcomes, and in some cases on nutrition. The examples of successful joint delivery of these services point to an area in which programmatic synergies, cost savings, and potential benefits for both child development and nutrition might be identified. Girls schooling is increasing in many countries, largely as a result of government interventions to change incen tives and reduce barriers to girls school enrolment and participation. Increases in parental schooling have con tributed to reductions in stunting, but larger effects could probably be achieved if effective nutrition education programmes were incorporated into school curricula.
www.thelancet.com

Series

Our review shows the potential of programmes in the four sectors reviewed to improve the lives of poor households and individuals, both in the short-term and the long-term. It also shows, however, that more needs to be done to increase the nutrition sensitivity of programmes so that their potential to deliver on maternal and child nutrition outcomes is unleashed. The nutrition sensitivity of promising programmes can be enhanced in several ways. First, targeting on the basis of nutritional vulnerability (eg, age, physiological status) in addition to geographic targeting on the basis of poverty, food insecurity, or location can help reach households and individuals most likely to benefit from the programme. Alternatively, targeting of nutri tionally vulnerable indi viduals could be used as a second level of targeting for subgroups of programme beneficiaries who meet precultural established criteria. For example, targeting agri programmes to households with pregnant or lactating women or children younger than 2 years might be neither logistically feasible nor optimum for community develop ment; however, geographic community-level targeting could be used as a first targeting criteria, with a second level focusing on reaching mothers and young children with a specific package of preventive nutrition and health interventions. Another key target group for nutritionsensitive programmes is adolescent girls; conditions or other incentives can be used to keep girls in school, help delay first pregnancy, address HIV risk factors,135 and improve adolescent girls' nutrition knowledge and micro nutrient status to prepare them for motherhood. Second, evidence shows that nutrition improvements are not automatic even with programmes that are success ful at reducing poverty, food insecurity, and sex inequalities. To reach their full potential, programmes such as those reviewed need careful identification of nutrition goals and appropriate design and effective implementation of interventions to achieve them. A third way to enhance the nutrition sensitivity of programmes is to engage women and include interventions to protect and promote their nutritional wellbeing, physical and mental health, social status, decision making, and their overall empowerment and ability to manage their time, resources, and assets. A fourth promising, yet underused approach to enhance the nutrition sensitivity of programmes is to use them as delivery platforms for various nutrition-specific inter ventions. Nutrition behaviour-change com muni cations, which are incor porated in several agriculture, social safety nets, early child development, and school health pro grammes are one example of such use. Other opportunities include addition of the distribution of micronutrientfortified products to nutritionally vulnerable adolescent girls, mothers, and young children,101,136 or of preventive health inputs to agriculture, social safety net, early child development, or school programmes. Finally, a crucial question that remains to be addressed when designing nutrition-sensitive programmes is the degree to which programmes should indeed integrate
www.thelancet.com

actions from several sectors, or co-locate programmes managed by different sectors so that they reach and saturate the same communities, households, and indi viduals. In view of the complexity of integration,33,137,138 especially across many sectors, it is important to carefully assess whether investments should focus on joint planning, implementation, monitoring, and assessment, or on effective programme co-location (panel 6).

Conclusions
Nutrition-sensitive programmes hold great promise for supporting nutrition improvements and boosting the scale, coverage, and benefits of nutrition-specific actions. New incentives are needed to support innovations in nutrition-sensitive programmes and unleash their potential to tackle nutrition while also achieving their own goals. New nutrition-sensitive agriculture44 and social safety net programme designs, methods, and packages of interventions are being tested and are strengthening links with health services. Rigorous impact evaluations, many of which are based on strong programme-theory and impact pathway analysis, are addressing key weaknesses encountered in previous evaluations and are assessing impacts on a range of nutrition and child development outcomes and several household and gender outcomes along the impact pathway. Evidence generated by these enhanced programmes and assessments in the next 510 years will be of crucial importance to inform future investments in agriculture and social safety net pro grammes to improve nutrition. The potential benefits of integration of early child development and nutrition programming include cost savings and gains in both child development and nutrition outcomes. Leveraging health, agriculture, or social safety net platforms for joint early child development and nutrition programming during the first 1000 days of life would help focus on the crucial period of peak vulnerability for both nutrition and development. Current work is exploring such approaches. Benefits from psychosocial interventions on cognition, however, extend well beyond the first 2 years, and therefore, continued child develop ment support is required thoughout the entire preschool period. Early child development programmes, possibly linked to conditions in transfer programmes or delivered through preschool or com munity settings, could offer psychosocial stimulation and parenting inter ventions, vant nutrition interventions for while also providing rele children 2 years and older, focusing on micronutrients, healthy diets, and obesity prevention. With improved guidance and curricula for nutrition education in schools, a new emphasis on using schools to improve nutrition knowledge and practices and preparing school children for their future parenting roles should also emerge. The immense potential of programmes addressing the underlying determinants of undernutrition to complement and enhance the effectiveness of nutritionspecific interventions is real, but is yet to be unleashed.
77

Series

Investments in nutrition-sensitive programmes can have a pivotal role in prevention of the excess stunting, wasting, and impaired child development that the scaleup of nutrition-specific interventions cannot resolve on its own.
Contributors MTR and HA conceptualised the report, reviewed the literature, and wrote the all drafts of the report. Maternal and Child Nutrition Study Group Robert E Black (Johns Hopkins Bloomberg School of Public Health, USA), Harold Alderman (International Food Policy Research Institute, USA), Zulfiqar A Bhutta (Aga Khan University, Pakistan), Stuart Gillespie (International Food Policy Research Institute, USA), Lawrence Haddad (Institute of Development Studies, UK), Susan Horton (University of Waterloo, Canada), Anna Lartey (University of Ghana, Ghana), Venkatesh Mannar (The Micronutrient Initiative, Canada), Marie Ruel (International Food Policy Research Institute, USA), Cesar Victora (Universidad de Federal de Pelotas, Brazil), Susan Walker (The University of the West Indies, Jamaica), Patrick Webb (Tufts University, USA). Series Advisory Committee Marc Van Ameringen (Gain Health Organization, Switzerland), Mandana Arabi (New York Academy of Sciences, USA), Shawn Baker (Helen Keller International, USA), Martin Bloem (United Nations World Food Programme, Italy), Francesco Branca (WHO, Switzerland), Leslie Elder (The World Bank, USA), Erin McLean (Canadian International Development Agency, Canada), Carlos Monteiro (University of So Paulo, Brazil), Robert Mwadime (Makerere School of Public Health, Uganda), Ellen Piwoz (Bill & Melinda Gates Foundation, USA), Werner Schultink (UNICEF, USA), Lucy Sullivan (1000 Days, USA), Anna Taylor (Department for International Development, UK), Derek Yach (The Vitality Group, USA). The Advisory Committee provided advice in a meeting with Series Coordinators for each paper at the beginning of the process to prepare the Series and in a meeting to review and critique the draft reports. Other contributors Derek Heady, research fellow at IFPRI, did the analysis and prepared the panel on the relation between income and nutrition. Joe Green, independent consultant, did the analysis of the role of education. Mara van den Bold, research analyst at IFPRI, reviewed the literature on womens empowerment, provided background text, and wrote the panel on womens empowerment and nutrition. Jef Leroy, research fellow at IFPRI, did the analysis and wrote the panel on the cash transfer programme in Mexico. Erick Boy, senior research fellow at IFPRI, provided background text and other material for the section on biofortification. Sivan Yosef, programme manager at IFPRI, provided research and editing assistance during all steps of the process. Conflicts of interest REB serves on the Boards of the Micronutrient Initiative, Vitamin Angels, the Child Health and Nutrition Research Initiative, and the Nestle Creating Shared Value Advisory Committee. VM serves on the Nestle Creating Shared Value Advisory Committee. All other authors delcare that they have no conflicts of interest. As corresponding author, Marie Ruel states that she has full access to all data and final responsibility for the decision to submit for publication. Acknowledgments Funding for the preparation of the Series was provided to the Johns Hopkins Bloomberg School of Public Health through a grant from the Bill & Melinda Gates Foundation. The sponsor has no role in the analysis and interpretation of the evidence or in writing the paper and the decision to submit for publication. We thank John Hoddinott, senior research fellow at IFPRI, who reviewed an early version of the paper, and to four anonymous reviewers for their insightful comments. References 1 Pinstrup-Andersen P, ed. The African food system and its interactions with human health and nutrition. Ithaca: Cornell University Press, 2010.

2 3 4 5

6 7 8 9 10

11

12 13 14 15 16 17 18 19 20

21

22

23 24

25

IFPRI. 2011 Global food policy report. Washington, DC: International Food Policy Research Institute, 2012. Nelson GC, Rosegrant MW, Koo J, et al. Climate change. Impact on agriculture and costs of adaptation. Washington, DC: International Food Policy Research Institute, 2009. Webb P. Medium-to long-run implications of high food prices for global nutrition. J Nutr 2010; 140: S14347. FAO, WFP, IFAD. The state of food insecurity in the world, 2012. Economic growth is necessary but not sufficient to accelerate reduction of hunger and malnutrition. Rome: Food and Agriculture Organization, 2012. World Bank. Global economic prospects: commodities at the crossroads. Washington, DC: World Bank, 2009. OECD-FAO. OECD-FAO agricultural outlook, 2012. DOI:10.1787/ agr_outlook-2012-en. Pinstrup-Andersen P, Watson DDI. Food policy for developing countries. The role of government in global, national and local food systems. Ithaca: Cornell University Press, 2011. Scaling Up Nutrition. Progress report from countries and their partners in the Movement to Scale Up Nutrition. New York: UN, 2011. Shekar M, Ruel-Bergeron J, Herforth A. Module A. Introduction. In: Improving nutrition through multisectoral appraoches. Washington, DC, International Bank for Reconstruction and Development, International Development Association of The World Bank, 2013. Black RE, Victora CG, Walker SP, and the Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S01406736(13)60937-X. Grosh M, Del Ninno C, Tesliuc E, Ouerghi A. For protection and promotion: the design and implementation of effective safety nets. Washington, DC: World Bank, 2008. Maxwell D, Webb P, Coates J, Wirth J. Rethinking food security in humanitarian response. Food Policy 2010; 35: 9197. Connolly MA, Gayer M, Ryan MJ, Salama P, Spiegel P, Heymann DL. Communicable diseases in complex emergencies: impact and challenges. Lancet 2004; 364: 197483. Alderman H. Safety nets can help address the risks to nutrition from increasing climate variability. J Nutr 2010; 140: S14852. Haddad L, Alderman H, Appleton S, Song L, Yohannes Y. Reducing child malnutrition: how far does income growth take us? World Bank Econ Rev 2003; 17: 10731. Alderman H, Linnemayr S. Anemia in low-income countries is unlikely to be addressed by economic development without additional programs. Food Nutr Bull 2009; 30: 26569. Demographic Health Surveys. Measure DHS Stat-Compiler. http:// www.statcompiler.com/ (accessed Nov 22, 2012). WHO. Global database on child Growth and malnutrition. World Health Organization, 2012. http://www.who.int/nutgrowthdb/ database/en/ (accessed Dec 12, 2012). Engle PL, Black MM, Behrman JR, Cabral de Mello M, Gertler PJ, Kapiriri L. Strategies to avoid the loss of developmental potential in more than 200 million children in the developing world. Lancet 2007; 369: 22942. Engle PL, Fernald LCH, Alderman H, et al. Strategies for reducing inequalities and improving developmental outcomes for young children in low-income and middle-income countries. Lancet 2011; 378: 133953. Bhutta ZA, Das JK, Rizvi A, et al, The Lancet Nutrition Interventions Review Group, and the Maternal and Child Nutrition Study Group. Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S01406736(13)60996-4. Almond D, Currie J. Killing me softly. J Econ Perspect 2011; 25: 125. Hoddinott J, Maluccio J, Behrman JR, et al. The consequences of early childhood growth failure over the life course. IFPRI Discussion Paper 01073. Washington, DC: International Food Policy Research Institute, 2011. Alderman H. The economic cost of a poor start to life. J Dev Origins Health Dis 2010; 1: 1925.

78

www.thelancet.com

Series

26 FAO. The state of food and agriculture, 20102011. Women in agriculture. Closing the gender gap for development. http://www. fao.org/docrep/013/i2050e/i2050e00.htm (accessed April 25, 2013). 27 Pinstrup-Andersen P. The food system and its interaction with human health nutrition. Leveraging agriculture for improved nutrition in health. 2020 Conference: Leveraging Agriculture for Improving Nutrition and Health. Brief 13. Washington, DC: International Food Policy Research Institute, 2011. 28 Hoddinott J, Rosegrant M, Torero M. Investments to reduce hunger and undernutrition. In: Lomborg B, ed. Copenhagen Consensus, 2012. Cambridge: Cambridge University Press, 2013. 29 World Bank. World Development Report 2008. Agriculture for development. Washington, DC: The International Bank for Reconstruction and Development. 30 Pinstrup-Andersen P. Guiding food system policies for better nutrition. Background paper for The State of Food and Agriculture, 2013. Rome: Food and Agriculture Organization, 2013. 31 Pinstrup-Andersen P. Can agriculture meet future nutrition challenge? Eur J Dev Res 2013; 25: 512. 32 Herforth A, Jones A, Pinstrup-Andersen P. Prioritizing nutrition in agricutural and rural development: guiding principles for operational investments. Health, Nutrition, and Population discussion paper. Washington, DC: The International Bank for Reconstruction and Development, World Bank, 2012. 33 World Bank. Improving nutrition through multisectoral approaches. Washington, DC: International Bank for Reconstruction and Development, International Development Association of The World Bank, 2013. 34 World Bank. From agriculture to nutrition. Pathways, synergies and outcomes. Washington, DC: The International Bank for Reconstruction and Development, World Bank, 2007. 35 Gillespie S, Harris J, Kadiyala S. The agriculture-nutrition disconnect in India: what do we know? IFPRI Discussion Paper 01187. Washington, DC: International Food Policy Research Institute, 2012. 36 World Bank. World Development Report 2012. Gender equality and development. Washington, DC: The International Bank for Reconstruction and Development, World Bank, 2011. 37 Hoddinott J. Agriculture, health, and nutrition: toward conceptualizing the linkages. 2020 Conference: Leveraging Agriculture for Improving Nutrition and Health. Paper 2. Washington, DC: International Food Policy Research Institute, 2012. 38 Leroy JL, Ruel M, Verhofstadt E, Olney D. The micronutrient impact of multisectoral programs focusing on nutrition: examples from conditional cash transfer, microcredit with education, and agricultural programs. http://www.micronutrientforum.org/innocenti/Leroy-et-alMNF-Indirect-Selected-Review_FINAL.pdf (accessed May 16, 2013). 39 Hawkes C. Identifying Innovative Interventions to Promote Healthy Eating Using Consumption-Oriented Food Supply Chain Analysis. J Hunger Environ Nutr 2009; 4: 33656. 40 Gereffi G, Lee J, Christian M. US-Based Food and Agricultural Value Chains and Their Relevance to Healthy Diets. J H Environ Nutr 2009; 4: 35774. 41 Hawkes C, Ruel M. Value chains for nutrition. 2020 Conference: Leveraging Agriculture for Improving Nutrition and Health. 2020 Conference Paper 4. Washington, DC: International Food Policy Research Institute, 2011. 42 Mazur R, Kizito Musoke H, Nakimbugwe D, Ugen M. Enhancing nutritional value and marketability of beans through research and strengthening key value-chain. 2020 Conference Note 1. Washington, DC: International Food Policy Research Institute, 2011. 43 Coote C, Tomlins K, Massingue J, Okwadi J, Westby A. Farmer, trader, and consumer decisionmaking: toward sustainable marketing of orange-fleshed sweet potato in Mozambique and Uganda. 2020 Conference Note 2. Washington, DC: International Food Policy Research Institute, 2011. 44 Hawkes C, Turner R, Waage J. Current and planned research on agriculture for improved nutrition: a mapping and a gap analysis. London: Department for International Development, 2012. 45 Sumberg J, Sabates-Wheeler R. Linking agricultural development to school feeding in sub-Saharan Africa: theoretical perspectives. Food Pol 2011; 36: 34149. 46 Gelli A, Neeser K, Drake L. Home grown school feeding: linking small holder agriculture to school food provision. PCD Working Paper 212. London: Partnership for child Development, 2010.

47 Ruel MT. Can food-based strategies help reduce vitamin A and iron deficiencies? A review of recent evidence. Food Policy Review #5. Washington, DC: International Food Policy Research Institute, 2001. 48 Berti PR, Krasevec J, Fitzgerald S. A review of the effectiveness of agriculture interventions in improving nutrition outcomes. Public Health Nutr 2004; 7: 599609. 49 Randolph TF, Schelling E, Grace D, et al. Invited review. Role of livestock in human nutrition and health for poverty reduction in developing countries. J Anim Sci 2007; 85: 2788800. 50 Arimond M, Hawkes C, Ruel MT, et al. Agricultural interventions and nutrition outcomes: lessons from the past and new evidence. In: Thompson B, Amoroso L, eds. Combating micronutrient deficiencies: food-based approaches. Rome: CAB International, Food and Agriculture Organization, 2011: 4175. 51 Masset E, Haddad L, Cornelius A, Isaza-Castro J. Effectiveness of agricultural interventions that aim to improve nutritional status of children: systematic review. BMJ 2012; 344: d8222. 52 Girard AW, Self JL, McAuliffe C, Olude O. The effects of household food production strategies on the health and nutrition outcomes of women and young children: a systematic review. Paediatr Perinat Ep 2012; 26: 20522. 53 Helen Keller International, Asian Vegetable and Research Development Centre. Home gardening in Bangladesh: evaluation report. Dhaka: HKI and AVRDC, 1993. 54 Solon F, Fernndez TL, Latham MC, Popkin BM. An evaluation of strategies to control vitamin A deficiency in the Philipines. Am J Clin Nutr 1979; 32: 144553. 55 Hoorweg J, Leegwater PH, Veeman W. Nutrition in agricultural development: intensive dairy farming by rural smallholders. Ecol Food Nutr 2000; 39: 395416. 56 Kassa H, Ayalew W, Habtegabriel Z, Gebremekel T. Enhancing the role of livestock production in improving nutritional status of farming families: lessons from a dairy goat development project in Eastern Ethiopia. Livestock Res Rural Dev 2003; 15: 6. 57 Low JW, Arimond M, Osman N, Cunguara B, Zano F, Tschirley D. A food-based approach introducing orange-fleshed sweet potatoes increased vitamin A intake and serum retinol concentrations in young children in rural Mozambique. J Nutr 2007; 137: 132027. 58 Hotz C, Loechl C, Lubowa A, et al. Introduction of carotene-rich orange sweet potato in rural Uganda resulted in increased vitamin A intakes among children and women and improved vitamin A status among children. J Nutr 2012; 142: 187180. 59 Hotz C, Loechl C, De Brauw A, et al. A large-scale intervention to introduce orange sweet potato in rural Mozambique increases vitamin A intakes among children and women. Br J Nutr 2012; 108: 16376. 60 Olney DK, Talukder A, Iannotti LL, Ruel MT, Quinn V. Assessing impact and impact pathways of a homestead food production program on household and child nutrition in Cambodia. Food Nutr Bull 2009; 30: 35569. 61 Bushamuka V, De Pee S, Talukder A, et al. Impact of a homestead gardening program on household food security and empowerment of women in Bangladesh. Food Nutr Bull; 26: 1725. 62 Hellen Keller International. Homestead food production improves household food and nutrition security (Bangladesh). Homestead Food Prod Bull 2004; 2: 14. 63 Hellen Keller International. Homestead food production program in Central and Far-Western Nepal increases food and nutrition security. Nepal Nutrition Bulletin 2004; 2: 18. 64 Kagenimana V, Anyango Oyunga M, Low J, Njoroge SM, Gichuki ST, Kabira J. Testing the effects of women farmers adoption and production of orange-fleshed sweet potatoes on dietary vitamin A intake in Kenya. Research report 3. Washington, DC: International Center for Research on Women, Opportunities for Micronutrients Interventions, 1999. 65 Mullins G, Wahome L, Tsangari P, Maarse L. Impacts of intensive dairy production on smallholder farm women in coastal Kenya. Hum Ecol 1996; 24: 23153. 66 Nielsen H. The socio-economic impact of a smallholder livestock development project, Bangladesh. In Dolberg F, Petersen PH, eds. Integrated farming in human development. Proceedings of a Workshop in Tune Landboskole, March 2529, 1996. Copenhagen: DSR-Forlag, 1997.

www.thelancet.com

79

Series

67 Begum JM. The impact of dairy development on protein and calorie intake of pre-school children. Indian J Med Sci 1994; 48: 6164. 68 Bouis HE, Hotz C, McClafferty B, Meenakshi JV, Pfeiffer WH. Biofortification: a new tool to reduce micronutrient malnutrition. Food Nutr Bull 2011; 32: S3140. 69 Saltzman A, Birol E, Bouis H, et al. Biofortification: progress toward a more nourishing future. Global Food Secur 2013; 2: 917. 70 Haas J, Beard J, Murray-Kolb LE, Del Mundo AM, Felix A, Gregoria GB. Iron-biofortified rice improves the iron stores of nonanemic Filipino women. J Nutr 2005; 135: 282330. 71 Carvalho LM, Correa MM, Pereira EJ, et al. Iron and zinc retention in common beans (Phaseolus vulgaris L) after home cooking. Food Nutr Res 2012; 56: 1561823. 72 Petry N, Egli I, Gahutu JB, Tugirimana PL, Boy E, Hurrell R. Stable iron isotope studies in Rwandese women indicate that the common bean has limited potential as a vehicle for iron biofortification. J Nutr 2012; 142: 49297. 73 DFID. Cash transfer evidence paper. London: Department for International Development Policy Division, 2011. 74 World Bank. Managing risk, promoting growth: developing systems for social protection in Africa. The World Banks Africa Social Protection Strategy 20122022. Washington, DC: World Bank, 2012. 75 Fiszbein A, Schady N, Ferreira FH, et al. Conditional cash transfers for attacking present and future poverty. World Bank Policy Research Report. Washington, DC: World Bank, 2009. 76 Leroy JL, Ruel M, Verhofstadt E. The impact of conditional cash transfer programmes on child nutrition: a review of evidence using a programme theory framework. J Deve Effect 2009; 1: 10329. 77 Manley J, Gitter S, Slavchevska V. How effective are cash transfer programmes at improving nutritional status? A rapid evidence assessment of programmes effects on anthropometric outcomes. London EPPI Centre. Social Research Science Unit. Institute of Education. London: University of London, 2012. 78 Gaarder MM, Glassman A, Todd JE. Conditional cash transfers and health: unpacking the causal chain. J Dev Effect 2010; 2: 650. 79 Lagarde M, Haines A, Palmer N. Conditional cash transfers for improving uptake of health interventions in low- and middle-income countries: a systematic review. JAMA 2007; 298: 190010. 80 Adato M, De la Brire B, Mindek D, Quisumbing A. The impact of PROGRESA on womens status and intrahousehold relations. Final Report. Washington, DC: International Food Policy Research Institute, 2000. 81 Adato M, Hoddinott J. Conditional cash transfers in Latin America. Baltimore: Johns Hopkins University Press, 2010. 82 Skoufias E, McClafferty B. Is PROGRESA working? Summary of the results of an evaluation by IFPRI. Food Consumption and Nutrition Division Discussion Paper 118. Washington, DC: International Food Policy Research Institute, 2001. 83 De Brauw A, Gilligan DO, Hoddinott J, Roy S. The impact of Bolsa Famlia on womens decision-making power. World Dev 2013; published online April 15. http://dx.doi.org/10.1016/ j.worlddev.2013.02.003. 84 Behrman J, Hoddinott J. Programme evaluation with unobserved heterogeneity and selective implementation: the Mexican PROGRESA impact on child nutrition. Oxford Bull Econ Stat 2005; 67: 54769. 85 Barber SL, Gertler PJ. Empowering women: how Mexicos conditional cash transfer programme raised prenatal care quality and birth weight. J Dev Effect 2010; 2: 5173. 86 Fernald LCH, Gertler P, Hidrobo M. Conditional cash transfer programs: effects on growth, health, and development in young children. In: Maholmes V, Kings RB, eds. The Oxford handbook of poverty and child development. New York: Oxford University Press, 2012. 87 Alderman H, Bundy D. School feeding programs and development: are we framing the question correctly? World Bank Res Obser 2012; 27: 20421. 88 Bundy D, Drake L, Burbano C. School food, politics, and child health. Public Health Nutr 2012; published online Nov 1. http:// dx.doi.org/10.1017/S1368980012004661. 89 Kristjansson EA, Robinson V, Petticrew M, et al. School feeding for improving the physical and psychosocial health of disadvantaged elementary school children. Cochrane Database Syst Rev 2007; 1: CD004676.

90 Doak C. Large-scale interventions and programmes addressing nutrition related chronic diseases and obesity: examples from 14 countries. Public Health Nutr 2002; 5: 27577. 91 Afridi F. Child welfare programs and child nutrition: evidence from a mandated school meal program in India. J Dev Econ 2010; 92: 15265. 92 Adelman S, Gilligan DO, Lehrer K. How effective are food for education programs? A critical assessment of the evidence from developing countries. IFPRI Food Policy Review 9. Washington, DC: International Food Policy Research Institute, 2008. 93 Kazianga H, De Walque D, Alderman H. Educational and health impact of two school feeding schemes: evidence from a randomized trial in rural Burkina Faso. World Bank Policy Research Working Paper 4976. Washington, DC: World Bank, 2009. 94 Osei AK, Rosenberg IH, Houser RF, Bulusu S, Mathews M, Hamer DH. Community level micronutrient fortification of school lunch meals improved vitamin A, folate, and iron status of schoolchildren in Himalayan villages of India. J Nutr 2010; 140: 114654. 95 Schady N, Rosero J. Are cash transfers made to women spent like other sources of income? Econ Lett 2008; 101: 24648. 96 Attanasio O, Battistin E, Mesnard A. Food and cash transfers: evidence from Colombia. Econ J 2012; 122: 92124. 97 Davis B, Gaarder M, Handa S, Yablonski J. Evaluating the impact of cash transfer programmes in sub-Saharan Africa: an introduction to the special issue. J Dev Effect 2012; 4: 18. 98 Akresh R, De Walque D, Kazianga H. Alternative cash transfer delivery mechanisms: impacts on routine preventative health clinic visits in Burkina Faso. World Bank Policy Research Working Paper #5958. Washington, DC: World Bank, 2012. 99 Leroy JL, Gadsden P, Rodrguez-Ramrez S, Gonzlez de Cossio T. Cash and in-kind transfers in poor rural communities in Mexico increase household fruit, vegetable, and micronutrient consumption but also lead to excess energy consumption. J Nutr 2010; 140: 61217. 100 Leroy JL, Gadsden P, De Cosso TG, Gertler P. Cash and in-kind transfers lead to excess weight gain in a population of women with a high prevalence of overweight in rural Mexico. J Nutr 2013; 143: 37883. 101 Ruel MT, Menon P, Habicht J-P, et al. Age-based preventive targeting of food assistance and behaviour change communication for reduction of childhood undernutrition in Haiti: a cluster randomized trial. Lancet 2008; 371: 58895. 102 Menon P, Ruel MT, Loechl CU, et al. Micronutrient sprinkles reduce anemia among 9- to 24-mo-old children when delivered through an integrated health and nutrition program in rural Haiti. J Nutr 2007; 137: 102330. 103 World Food Programme. Global food aid flows, 2011. Rome, International Food Aid Information System: 2012. 104 Giles J, Satriawan E. Protecting child nutritional status in the aftermath of a financial crisis: evidence from Indonesia. Policy Research Working Paper 5471. Washington, DC: World Bank, 2010. 105 Yamano T, Alderman H, Christiaensen L. Child growth, shocks, and food aid in rural Ethiopia. Am J Agric Econ 2005; 87: 27388. 106 Chaparro CM, Dewey KG. Use of lipid-based nutrient supplements (LNS) to improve the nutrient adequacy of general food distribution rations for vulnerable sub-groups in emergency settings. Mat Child Nutr 2010; 6: 169. 107 Huybregts L, Houngb F, Salpteur C, et al. The effect of adding ready-to-use supplementary food to a general food distribution on child nutritional status and morbidity: a cluster-randomized controlled trial. PLoS Med 2012; 9: e1001313. 108 Grellety E, Shepherd S, Roederer T, et al. Effect of mass supplementation with ready-to-use supplementary food during an anticipated nutritional emergency. PLoS One 2012; 7: e44549. 109 Walker SP, Wachs TD, Grantham-McGregor S, et al. Inequality in early childhood: risk and protective factors for early child development. Lancet 2011; 378: 132538. 110 Grantham-McGregor SM, Powell CA, Walker SP, Himes JH. Nutritional supplementation, psychosocial stimulation, and mental development of stunted children: the Jamaican Study. Lancet 1991; 338: 15.

80

www.thelancet.com

Series

111 Walker SP, Chang SM, Powell CA, Grantham-McGregor SM. Effects of early childhood psychosocial stimulation and nutritional supplementation on cognition and education in growth-stunted Jamaican children: prospective cohort study. Lancet 2005; 366: 180407. 112 Walker SP, Chang SM, Vera-Hernndez M, Grantham-McGregor SM. Early childhood stimulation benefits adult competence and reduces violent behavior. Pediatrics 2011; 127: 84957. 113 Nahar B, Hamadani JD, Ahmed T, et al. Effects of psychosocial stimulation on growth and development of severely malnourished children in a nutrition unit in Bangladesh. Eur J Clin Nutr 2009; 63: 72531. 114 Aboud FE, Akhter S. A cluster-randomized evaluation of a responsive stimulation and feeding intervention in Bangladesh. Pediatrics 2011; 127: e119197. 115 Gardner JMM, Powell CA, Baker-Henningham H, Walker SP, Cole TJ, Grantham-McGregor SM. Zinc supplementation and psychosocial stimulation: effects on the development of undernourished Jamaican children. Am J Clin Nutr 2005; 82: 399405. 116 Vazir S, Engle P, Balakrishna N, et al. Cluster-randomized trial on complementary and responsive feeding education to caregivers found improved dietary intake, growth and development among rural Indian toddlers. Mat Child Nutr 2013; 9: 99117. 117 Yousafzai AK, Rasheed MA, Rizvi A, Armstrong R, Bhutta ZA. Pakistan early child development scale up trial. Integrating nutrition and early childhood development: evaluation of efficacy and effectiveness research and programs, Sackler Institute for Nutrition Science, the Global Child Development Group. 2012. 118 Nahar B, Hossain MI, Hamadani JD, et al. Effects of a community-based approach of food and psychosocial stimulation on growth and development of severely malnourished children in Bangladesh: a randomised trial. Eur J Clin Nutr 2012; 66: 70109. 119 Bentley M, Wasser H, Creed-Kanashiro H. Responsive feeding and child undernutrition in low and middle-income countries. J Nutr 2011; 141: 50207. 120 Black M, Baqui AH, Zaman K, El Arifeen S, Black RE. Maternal depressive symptoms and infant growth in rural Bangladesh. Am J Clin Nutr 2009; 89: S95157. 121 Surkan PJ, Kennedy CE, Hurley KM, Black MM. Maternal depression and early childhood growth in developing countries: systematic review and meta-analysis. WHO Bull 2011; 89: 60815. 122 Tripathy P, Nair N, Barnett S, et al. Effect of a participatory intervention with womens groups on birth outcomes and maternal depression in Jharkhand and Orissa, India: a cluster-randomised controlled trial. Lancet 2010; 375: 118292. 123 Rahman A, Malik A, Sikander S, Roberts C, Creed F. Cognitive behavior therapy-based intervention by community health workers for mothers with depression and their infants in rural Pakistan: a cluster-randomized controlled trial. Lancet 2008; 372: 90209.

124 Semba R, De Pee S, Sun K, Sari M, Akhter N, Bloem M. Effect of parental formal education on risk of child stunting in Indonesia and Bangladesh. Lancet 2008; 371: 32228. 125 Chou S-Y, Liu J-T, Grossman M, Joyce T. Parental education and child health: evidence from a natural experiment in Taiwan. Am Econ J Applied Econ 2011; 2: 3361. 126 Barro R, Lee J-H. A new data set of educational attainment in the world 19502010. NBER Working Paper 15902. Cambridge, MA: National Bureau of Economic Research, 2010. 127 Grant M, Behrman J. Gender gaps in educational attainment in low income countries. Pop Dev Rev 2010; 36: 7186. 128 World Bank. Education statistics. http://www.worldbank.org/ education/edstats (accessed March 14, 2013). 129 Glewwe P. Why does mothers schooling raise child health in developing countries? Evidence from Morocco. J Hum Resour 1999; 34: 12436. 130 Block S. Maternal nutrition knowledge versus schooling as determinants of child micronutrient status. Oxford Econ Pap 2007; 1: 33050. 131 FAO. Nutrition education in schools curriculum. Rome: Food and Agriculture Organization, 2005. 132 Nederveen L. Global mapping of initiatives in school health and nutrition, with emphasis on health education. Unilever, Together for Child Vitality, World Food Programme. http://www. schoolsandhealth.org/Documents/Global Mapping Initiatives School Health and Nutrition.pdf (accessed Feb 11, 2013). 133 Roosmarijn V, Roberfroid D, Lachat C, et al. Effectiveness of preventive school-based obesity interventions in low- and middle-income countries: a systematic review. AmJ Clin Nutr 2012; 96: 41538. 134 Bhutta ZA, Ahmed T, Black RE, et al. What works? Interventions for maternal and child undernutrition and survival. Lancet 2008; 371: 41740. 135 Pettifor A, Macphail C, Nguyen N, Rosenberg M. Can money prevent the spread of HIV? A review of cash payments for HIV prevention. AIDS Behav 2012; 16: 172938. 136 Neufeld LM, Rivera J, Martinez Valle A, Grados R, Uriega S, Lpez VH. Evaluation for program decision making: a case study of the Oportunidades program. J Nutr 2011; 141: 207683. 137 Gillespie S, Haddad L, Mannar V, Menon P, Nisbettt N, for the Maternal and Child Nutrition Study Group. The politics of reducing malnutrition: building commitment and accelerating progress. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/ S0140-6736(13)60842-9. 138 Garrett J, Natalicchio M, eds. Working multisectorally in nutrition. Principles, practices and case studies. http://www.ifpri.org/ publication/working-multisectorally-nutrition (accessed Jan 30, 2013).

www.thelancet.com

81

Series

Maternal and Child Nutrition 4 The politics of reducing malnutrition: building commitment and accelerating progress
Stuart Gillespie,* Lawrence Haddad,* Venkatesh Mannar, Purnima Menon, Nicholas Nisbett, and the Maternal and Child Nutrition Study Group
Published Online June 6, 2012 http://dx.doi.org/10.1016/ S0140-6736(13)60842-9 This is the fourth in a Series of four papers about maternal and child nutrition *Joint lead authors International Food Policy Research Institute, Washington, DC, USA (S Gillespie PhD); Institute of Development Studies, Brighton, UK (Prof L Haddad PhD; N Nisbett PhD); Micronutrient Initiative, Ottawa, ON, Canada (V Mannar MS); and International Food Policy

In the past 5 years, political discourse about the challenge of undernutrition has increased substantially at national and international levels and has led to stated commitments from many national governments, international organisations, and donors. The Scaling Up Nutrition movement has both driven, and been driven by, this developing momentum. Harmonisation has increased among stakeholders, with regard to their understanding of the main causes of malnutrition and to the various options for addressing it. The main challenges are to enhance and expand the quality and coverage of nutrition-specific interventions, and to maximise the nutrition sensitivity of more distal interventions, such as agriculture, social protection, and water and sanitation. But a crucial third level of action exists, which relates to the environments and processes that underpin and shape political and policy processes. We focus on this neglected level. We address several fundamental questions: how can enabling environments and processes be cultivated, sustained, and ultimately translated into results on the ground? How has high-level political momentum been generated? What needs to happen to turn this momentum into results? How can we ensure that high-quality, well-resourced interventions for nutrition are available to those who need them, and that agriculture, social protection, and water and sanitation systems and programmes are proactively reoriented to support nutrition goals? We use a six-cell framework to discuss the ways in which three domains (knowledge and evidence, politics and governance, and capacity and resources) are pivotal to create and sustain political momentum, and to translate momentum into results in high-burden countries.

Introduction
Key messages Emerging country experiences show that rates of undernutrition reduction can be accelerated with deliberate action. Politicians and policy makers who want to promote broad-based growth and prevent human suffering should prioritise investment in scale-up of nutrition-specific interventions, and should maximise the nutrition sensitivity of national development processes. Findings from studies of nutrition governance and policy processes broadly concur on three factors that shape enabling environments: knowledge and evidence, politics and governance, and capacity and resources. Framing of undernutrition reduction as an apolitical issue is short sighted and self-defeating. Political calculations are at the basis of effective coordination between sectors, national and subnational levels, private sector engagement, resource mobilisation, and state accountability to its citizens. Political commitment can be developed in a short time, but commitment must not be squanderedconversion to results needs a different set of strategies and skills Leadership for nutrition, at all levels, and from various perspectives, is fundamentally important for creating and sustaining momentum and for conversion of that momentum into results on the ground. Acceleration and sustaining of progress in nutrition will not be possible without national and global support to a long-term process of strengthening systemic and organisational capacities. The private sector has substantial potential to contribute to improvements in nutrition, but efforts to realise this have to date been hindered by a scarcity of credible evidence and trust. Both these issues need substantial attention if the positive potential is to be realised. Operational research of delivery, implementation, and scale-up of interventions, and contextual analyses about how to shape and sustain enabling environments, is essential as the focus shifts toward action.

The nutrition landscape has shifted fundamentally since the first Lancet Series on Maternal and Child Under nutrition was published in January, 2008. Since then, almost every major development agency has published a policy document about undernutrition. In a very difficult fiscal climate, official development assistance to the basic nutrition category has increased from US$259 million in 2008, to $418 million in 2011a rise of more than 60% (although it was $541 million in 2009).1 Furthermore, the G8 countries reported increases of almost 50% in bilateral spending on nutrition-specific and nutritionsensitive interventions between 2009 and 2011.2 Accord ing to Google Trends, malnutrition, now matches HIV/AIDS in terms of internet interest, whereas 5 years ago, HIV/AIDS received twice as much interest as malnutrition. This shift is attributable to several factors: the food price spikes of 200708 sparked renewed media and policy interest in undernutrition, The Lancet 2008 Series provided policy makers with a set of tangible interventions that were effective in various locations, and the 2008 Copenhagen Consensus concluded that nutrition interventions were among the most cost effective in development.3 The Scaling Up Nutrition (SUN) movement, which started in September 2010, is the most important symbol of the increased interest in nutrition.4 By the middle of May, 2013, the movement had grown to include 35 countries that are committed to the scale-up of direct nutrition interventions and the advancement of nutrition-sensitive development, including 21 of the
www.thelancet.com

82

Series

34 highest burden countries where 41% of the global burden of child stunting is located (or 56% if India is omitted). As SUN nears its 1000th day, several countries have made advances in terms of building multistake holder platforms, aligning nutrition-relevant programmes within a common results framework, and mobilising national resources, but it is too soon to evaluate the effect of SUN on rates of undernutrition reduction. As interest in nutrition has changed, so too has our thinking. The large economic returns to nutritionspecific interventions (paper two in this Series5), are clear6 and we recognise the potential of nutritionsensitive interventions (paper three7) and the impor tance of an enabling environment for reduction of under nutritionthe focus of this report.8 Most of the concepts and ideas that we develop about enabling environments apply to both undernutrition and the growing problems of overweight and obesity as documented in the first paper in this Series. We focus mainly on undernutrition because as the 2010 Global Burden of Disease estimates show, undernutrition remains the number one risk factor in sub-Saharan Africa, and the fourth in south Asia.9 We use evidence generated within academic and scientific institutions and that generated in more real-world, action oriented, transdisciplinary ways that embed nutrition within wider social and political contexts.10 We used this mixture of evidence types partly because of the paucity of the first type of evidence and partly in recognition that the second type is often more appropriate because it is more practical, politically feasible, and therefore actionable. However, the second type of evidence is not as easy to independently verify or systematise with standard systematic review protocols. Beyond the nutrition-sensitive programmes and inter ventions discussed in paper three, other macro-level drivers exist that lie at the end of long causal pathways. Seemingly quite remote from the nutritional wellbeing of children, many such drivers are nonetheless crucially important to shape both national and global political landscapes for nutrition, and basic-level determinants of nutrition status. These aspects are particularly important because each of the various determinants of nutritional outcomes can be vulnerable to sudden changes within, or caused by, these drivers. Examples include climate change, trade, the rate and pattern of economic growth, food and energy prices and volatility, and land-use policies. Previous empirical work at the country level has shown that household income growth is a necessary, but not sufficient driver, of nutrition status.11 In a crosscountry study of the drivers of nutritional change over time,12 four factors emerged as the most robust predictors of reductions in undernutrition worldwide: secondary education for girls, reductions in fertility, accumulation of household assets, and increased access to health services. In view of the scarce evidence for these drivers we do not discuss the related scientific literature. Rather,
www.thelancet.com

we reiterate that through the approaches for shaping enabling environments for nutrition, described here, we might be better able to advocate for attention to nutrition within these broad development debates.

Research Institute, New Delhi, India (P Menon PhD) Correspondence to: Dr Stuart Gillespie, International Food Policy Research Institute, Washington, DC, 20006-1002, USA s.gillespie@cgiar.org

Characterisation of enabling environments


What does an enabling environment for undernutrition reduction look like? In recognition of the general con sensus that income growth is necessary but not sufficient for undernutrition reduction,7,13,14 we undertook a system atic review of the nutrition-relevant policy process and governance literature (panel 1). After a surge of activity in the late 1970s to early 1980s, a two decade gap ensued in research of nutrition policy processes, punctuated by one book in 1993, until interest re-emerged in 2003. In the past decade, several multicountry and single-country studies of such processes have been undertaken, in which conceptual and analytical frameworks have been applied.1523 These studies sought to uncover key structures, pathways, and dynamics of policy processes for nutrition, with an emphasis on challenges and constraints. In doing so, research from other specialties (eg, political science24,25 and health systems26) was drawn on to adopt and adapt analytical frameworks and research methods to study nutrition policy. We define an enabling environment as political and policy processes that build and sustain momentum for the effective implementation of actions that reduce undernutrition. Rather than wait for political will to emerge by chance, our review clearly shows that a political momentum can be developed and sustained
Panel 1: search strategy and selection criteria

See Online for appendix

For analysis of enabling environments, we searched Medline, Web of Science, and Econlit, between Nov 12 and 16, 2012, with predefined search terms (nutrition, governance and poli*, words to appear in the title of paper), with no date or language restrictions. Results were exported to a bibliographic reference manager (EndNote). We did further searches in ELDIS and Google Scholar to identify references in the grey literature. We screened results for duplicate references and for relevance to this paper. For assessment of the Scaling Up Nutrition (SUN) movement we used two sources of new data. First, monitoring data from 30 countries (submitted in September, 2012) was provided by the SUN secretariat, including detailed information about core indices being tracked by SUN, local expectations and proposed commitments of SUN focal points in these countries (appendix). The four indicators routinely tracked by SUN relate to the presence of a multistakeholder platform, a legal and political framework, a common results framework, and alignment and mobilisation of resources. Second, we undertook a closed online discussion with the Eldis Communities web platform, a service provided by the Institute of Development Studies, with 75 invited participants from six countries (Bangladesh, Ethiopia, Indonesia, Kenya, Nepal, and Nigeria) from Nov 27, 2012, to Dec 4, 2012, to explore perceived benefits and expectations of joining SUN, the main challenges and constraints faced by countries, and what needs to happen next. Participants consisted of experts from central and subnational government, multilateral and bilateral development agencies, national and international non-governmental organizations, civil society organisations, and research institutionsall of whom are working directly or indirectly in nutrition (appendix).

83

Series

Panel 2: Framework for creation of an enabling environment for accelerated undernutrition reduction Framing, generation, and communication of knowledge and evidence Issues and challenges to creation and sustaining of momentum Framing and narratives Evidence of outcomes and benefits What works and how well do nutrition interventions work relative to others? Advocacy to increase priority (civil society) Evidence of coverage, scale, and quality Issues and challenges to conversion of momentum into results Implementation research (what works, why, and how?) Programme evaluation (impact pathways) Generation of demand for evidence of effectiveness Political economy of stakeholders, ideas, and interests Issues and challenges to creation and sustaining of momentum Incentivising and delivering of horizontal coherence (multisectoral coordination) Development of accountability to citizens Enabling and incentivising of positive contributions from the private sector Issues and challenges to conversion of momentum into results Delivery of vertical coherence The role of civil society and the private sector in delivery Capacity (individual, organisational, systemic) and financial resources Issues and challenges to creation and sustaining of momentum Leadership and championing Systemic and strategic capacity Making the case for additional resource mobilisation Issues and challenges to conversion of momentum into results Delivery and operational capacity New forms of resource mobilisation Prioritisation and sequencing of nutrition action Implementation and scale-up

communicated clearly to generate pressure on politicians to act. Second, politics and governance. Various stake holders and agencies, each with different and frequently competing agendas (especially in decentralised systems of governance), need to work together to reduce under nutrition. All but the most extreme manifestations of undernutrition have no visible symptoms and are thus open to neglect, so even well-meaning governments might underinvest in nutrition. Data for nutrition trends and programme effectiveness are often out of date or scarce, allowing unsubstantiated political narratives to be sustained in an evidence vacuum. Third, capacity and resources. Human and organisational capacity need to encompass not only nutrition know-how, but also a set of soft-power skills to operate effectively across boundaries and disciplines, such as leadership for alliance build ing and networking, communication of the case for collaboration, leveraging of resources, and being able to convey evidence clearly to those in power. Strategic and operational capacities of different stakeholders at several levels are key. Additional financial resources and much better budget data are needed if undernutrition efforts are to be scaled up, with innovation from governments and donors to maximise investment. Panel 2 shows the issues and challenges for creation and conversion of momentum within these three parameters. We apply this framework to three case studies (Malawi, Peru, and Maharashtra [a state in India]) where trends from the past few years have been positive and rigorous efforts have been made to prioritise nutrition, reshape policy, and scale-up or improve nutrition-related pro gramming (appendix).

Creation and sustaining of momentum


Narratives, knowledge, and evidence
The 2008 Lancet Nutrition Series showed how effective marshalling of evidence can create momentum by identifying a set of interventions that were effective at reducing undernutrition in various contexts, identifying a window of opportunity1000 daysas a focal point, and imparting a sense of priority and feasibility by showing how undernutrition is concentrated in a small set of high-burden countries. The 2008 Series also empha sised the fragmented nature of the international nutrition community with regard to messaging, priorities, and funding,28 and contributed to birth of the SUN movement (panel 3). Undernutrition has unique features that guide the kinds of knowledge and evidence needed for progress (panel 2).

through deliberate action.10,23,27 Moreover, translation of any such momentum into effect on nutrition status is far from automatic and needs the deliberate alignment of several factors and processes.10,1822 Our review also emphasises three linked factors as being crucial for building and sustaining of momentum and for con version of that momentum into results. First, knowledge and evidence. Undernutrition is a multisectoral challenge that is open to various inter pretations (eg, as a health, economic growth, inter generational rights, or humanitarian issue). Each context needs its own enabling narrative or framing. This multisectoral nature also raises challenges for imple mentation of nutrition programmes and increases the importance of quality implementation research and impact evaluation. Undernutrition in early life is irre versible; therefore timely and reliable information about nutrition status and its determinants in pro grammatic contexts is crucial. Additionally, rigorous research is needed to capture the long-term inter generational benefits of undernutrition prevention, with evidence
84

The importance of framing


Reduction of undernutrition is a multisectoral activity, thus choices exist for how it is framed. In Guatemala and Bolivia, framing has been focused on hunger elimination, strongly determined by Brazils own Zero Hunger campaign.1820 In Peru, civil society developed under nutrition reduction as an electoral issue21
www.thelancet.com

Series

Panel 3: Main points from an online electronic consultation among stakeholders from six Scaling Up Nutrition (SUN) countries SUN represents an unprecedented opportunity for coordination, collaboration, cross learning, and advocacy to catalyse sustainable nutrition gains at national and global levels. Membership implies a national commitment to address undernutrition. SUNs own monitoring system is centred on four key indicators (appendix). However, to track and compare progress between so many countries, monitoring systems will tend to default to quantitative data of what does or does not exist. Quality and process is not so easily measured. For this reason, and to help us to uncover local perceptions about key issues, challenges, and constraints related to translation of SUN ambitions on the ground, online discussionsorganised by the Institute of Development Studies and the International Food Policy Research Institutewere undertaken (appendix). 75 key stakeholders from different sectors in Bangladesh, Nepal, Indonesia, Ethiopia, Nigeria, and Kenya were actively involved over the 8 days of consultation, from Nov 27, 2012, to Dec 4, 2012. In brief, perceived expectations of joining SUN are that it provides a framework and platform for improved coordination and cooperation in nutrition. SUN encourages advocacy, which has increased the number of stakeholders across sectors who are working to address undernutrition. In turn, this increase is hoped to increase leveraging of resources, knowledge sharing, and institutional capacity. The SUN movement is also considered to hold stakeholders (especially the government) accountable, and secure further commitment to improve resource mobilisation and allocation. Areas of perceived progress include increased awareness and advocacy across sectors. Ambassadors and champions for nutrition at various levels, from the prime minister to the community, have pushed nutrition onto the agenda. Policy makers are increasingly aware of nutrition as a development issue, and some countries have increased nutrition-relevant budgets. The main perceived challenges and constraints to SUN within countries include little coordination and collaboration between (and within) different ministries, related scarcity of clarity and consensus vision on what scaling up means, undefined roles and responsibilities, and few or ineffective policies and political commitment. Decentralisation of SUN is a major challenge in some countries. Translation of SUN from national to community levels is restricted. The issue of weak capacity (all types and at all levels) was raised several times with particular challenges, including inadequately qualified personnel (eg, doctors and nurses) and community and extension workers (eg, front-line workers and health volunteers) in remote areas, and high employee turnover. Financial resources are often unsustainable and unpredictable with funding for nutrition interventions largely donor driven. Funding for scale-up is insufficient and issues exist about budgetary allocation (emphasis on treatment over prevention) and coordination. Poor quality of monitoring and evaluation data affects assessment of the effect of interventions, weakens advocacy strategies, and jeopardises funding. Finally, views about engagement with the private sector were mixed and suspicion around motivations was reported. Private sector involvement needs close regulation and a framework within which to engage.

(appendix). In India, nutrition has risen on the agenda through a combination of advocacy around the finding that economic growth has not generated nutritional benefits,29 a strong rights-based movement led by the Right to Food initiative,30 and a growing stakeholder consensus of the need for multisectoral action.31 In Ghana, which has achieved the fastest decline in child stunting in sub-Saharan Africa in the past 5 years (from 35% in 2003, to 28% in 2008,32 a rate of 15 percentage points per year), the agenda was one of investment in agriculture as a driver of economic growth and poverty reduction,33 together with feeding initiatives for infants and young children, all in the context of a stable political environment.34 The multisectoral nature of undernutrition reduction adds some complexity to the implementation of effec tive programmes. Even breastfeeding promotion, for example, needs action on various fronts: behavioural change from breastfeeding mothers, workplace oppor tunities to breastfeed, responsible advertising about breast-milk substitutes, and effective legislation to define and monitor unacceptable behaviour or to challenge countervailing narratives. The returns to high-quality impact evaluation in the face of such complexity are likely
www.thelancet.com

to be large. The inclusion of nutrition objectives and targets within nutrition-sensitive programmes is thought to be important to leverage resources for nutrition within those programmes; however, this hypothesis needs to be tested.

The timeliness, credibility, and persuasiveness of data


The irreversibility of undernutrition early in life makes quick and effective action crucial. The availability of timely and credible data presented in accessible ways can help governments and other stakeholders to be responsive to changing circumstances, and help civil society organ isations to hold them accountable for the effective ness of their interventions. Data from the Demographic and Health Survey and the Multiple Indicator Cluster Survey are essential for evaluation of national trends, but are only collected every 35 years and are less useful for immediate programmatic decision making. Surveillance mechanisms, for tracking of nutrition trends and to inform timely decision making, only exist in a few countries.3538 Advances in health management infor mation systems and the growing availability of new technologies could facilitate the real-time monitoring of nutrition outcomes and programme coverage and quality.
85

Series

When, where, how, and why these new technologies are practical and will lead to responsive and effective action for nutrition are important research issues.39,40

Political economy and governance


The politics of undernutrition reduction have long been neglected. The multitude of involved stakeholders at many levels, the invisibility of undernutrition, and the imbalance of power between governments and multi national organisations, generate little accountability for commitment and delivery, and fuel the political economy of undernutrition reduction.

Communication of the benefits to improved nutrition


The benefits of undernutrition reduction are lifelong, and yet their temporal distribution reduces their political appeal. The studies in Guatemala of the long-term bene fits of undernutrition prevention41,42 have been extremely influential worldwide, and the Consortium on Health Orientated Research in Transitional Societies (COHORTS) group is starting to yield multicountry evidence about the long-term implications of early childhood nutrition.43,44 The challenge is to generate contemporary political payoffs to these nutritionally driven long-term labour-market bene graphic transition that many developing fits. The demo countries are experiencing and debating at the highest policy circles presents an example of one such opportunity to communicate the importance of nutrition in ways that resonate. The so-called demographic dividend45 due to the declining ratio of adults of non-working age to those of working age will be greatly enhanced if those of working age can secure market employment. Investments in mater nal and early childhood nutrition that build human capital can be framed as one way to secure this dividend.
Panel 4: Nutrition post-2015 Despite the negligible presence of nutrition in the Millennium Development Goals (one indicator of one goal), inclusion of the underweight indicator has probably helped donors and development agencies justify increased attention to nutrition. This increased attention needs to be shown more fully in the next set of development goals to maintain the high levels of commitment and to guide action. We recommend the following approach: 1 Find a location for nutrition as an equal partner within a likely goal, such as hunger reduction or poverty or health. This location in a vertical goal will raise the profile of nutrition 2 Make sure that nutrition indicatorsnutrition-specific and nutrition-sensitive are located within an additional number of vertical goals, such as gender equity, education, and employment. All these indicators should be linked across the different goals with the framework developed in paper one of this Series to generate a horizontal nutrition goal 3 Endorse the six global targets for nutrition-specific indicators (including replacing of underweight with stunting) proposed by the World Health Assembly in 2012 Why not advocate for a separate nutrition goal? A stand-alone nutrition goal has many desirable features: it makes ignoring of malnutrition harder and is likely to galvanise stakeholders in the nutrition (and possibly development) community and in the general public. However, extensive reading of the post-2015 scientific literature47 suggests that support for a separate goal is insufficient. Building of support might still be possible, but nutrition lags behind other more high-profile disease burdens; stiff competition might come from other constituencies who think they should have a separate goal (eg, water, sanitation, population); and the case has to be made as to why nutrition would not fit better into closely related goals, such as food or health. There are risks to having a separate nutrition goal: constituencies of the other goals might find it easier to ignore nutrition, and we know that reductions in malnutrition require their engagement. We judge our recommendation as more feasible politically and if done strategically it could well leverage more resources for nutrition, especially from nutrition-sensitive programmes and interventions.

Global governance
National governments, civil society (global and national), international and regional organisations (including UN agencies, development banks, and the African Union), bilateral donors, charitable foundations, international research organisations (eg, the Consultative Group on International Agricultural Research), academia, and private-sector companies all have a role in the global institutional architecture for nutrition. 5 years ago, the stewardship or governance of this system was fragmented and dysfunctional.28 Since then, a process to reform UN institutional architecture has started and the SUN movement has emerged (panel 3), engaging more than 100 bodies within these organisations. SUN is governed by a lead group of heads of state and other key stake holders, but is focused mainly on galvanising national and country-led action (panel 3). Despite SUNs substantial convening power, some external and country-level confusion exists about the role of the SUN movement, the UN Standing Committee on Nutrition, and the UN REACH programme (the latter two focus on UN level technical support and governance coordination, respectively). Most individuals recognise the continued value of the UN Standing Committee, but it remains in a fragile position and in need of further internal reform (unpublished). Other important global initiatives include the multinational 1000 days partnership, the partnership of G8 countries, and the New Alliance for Food Security and Nutrition (consisting of several African countries and private companies). Meanwhile, the UN Secretary-Generals High Level Task Force on the Global Food Security Crisis and a revitalised Committee on Food Security have emerged as important bodies, coordinating UN and global responses to food insecurity and comple menting the role of existing UN food and agriculture bodies. The World Health Assemblys agreement on six new global undernutrition targets to be achieved by 2025 has also become an important part of the global nutrition focus; however, questions remain about the achievability of these targets46 and their incorporation within the framework of the Millennium Development Goals (panel 4). The true potential of the SUN movement will be realised through its application in each SUN country. Success of the movement will need maintenance of support and consensus amongst all SUN stakeholders, and development of a strong sense of country-level ownership, the absence of which was a major reason for
www.thelancet.com

86

Series

the failure of the multisectoral nutrition planning experiments of the 1970s.48 As SUNs global scope increases, so will demands for effective information and knowledge management. Presentation of results that correlate SUN activities with measurable reductions in nutrition indicators will become a key focus.

The need for horizontal coordination


Different agencies, each with different and frequently competing agendas, need to work together if under nutrition is to be reduced. Associations are horizontal (at the same level of government) and vertical (at central, state, and district levels) and the potential for conflicting agendas in all directions is substantial. A political analysis21,49 of the horizontal and vertical associations in Brazil, Peru, Ethiopia, Zambia, India, and Bangladesh reached several conclusions regarding the roles of the executive branch of government and well-resourced coordination bodies, the importance of narratives that link nutrition with development, and civil society pressure mechanisms. Another study of multisectoral (horizontal) coordination in Senegal and Colombia16 emphasised the importance of inclusiveness of institu tions and stake holders, incentives, and lateral (as opposed to top-down) leadership. SUN has sought to promote horizontal coherence through establishment of multisectoral plat forms to catalyse and enable comple mentary, coordinated, and integrated action. However the data from six SUN countries show that convergence and coordination continue to be a challenge (panel 3, appendix).

commitment to fighting hunger and malnutrition is needed.54 For governments and donors, the Institute of Development Studies has developed a Nutrition Commitment Index for cross-country and countryspecific comparisons over time (panel 5). For food and beverage manufacturers, a new index has been launched by the Global Alliance for Improved Nutrition to evaluate their policies, practices, and performance in contribution to the reduction of undernutrition and overweight and obesity.55 The potential of mechanisms, such as social audits and community monitoring, to promote account ability and improve the provision of direct public services is clear50,56 and has been positively appraised,57 but has not yet been empirically tested for the provision of frontline nutrition services. Empirical evidence about the effect of such accountability mechanisms on the quality of care and health facilities is weak, but encouraging.58,59 A trial of community-based monitoring of health service provision in Uganda showed a 33% reduction in mortality in children younger than 5 years and a signifi cant 014 increase in weight-for-age Z score.60

Civil society engagement


Most of the roughly 100 organisations who have signed up to the SUN movement are civil society organisations. Their role in combating undernutrition is as multi-faceted and multi-functional as the sector itself, but the effect of citizen engagement is difficult to evaluate.61 Of the many roles of these organisations, four stand out: (1) global and national advocacy to call attention to nutritional depriv ation and galvanise commitment to act, (2) ensuring of accountability for nutrition-relevant service coverage and quality, (3) generation of context-specific knowledge about
Commitment To reduce stunting faster Data make commitments transparent Commitment Index Nutrition Diagnostics

The need to strengthen accountability


Providers, governments, donors, and the private sector need strong mechanisms to incentivise and hold them accountable for the quality and effectiveness of any nutrition investment. Although the evidence base for nutrition lags behind the positive evidence base for a range of other sectors,50 investments to increase commit ment and accountability for nutrition services and measure their effects could be one of the most rewarding applications of research to macro (commitment) and micro (accountability) levels. Increases in nutrition com mit ment and accountability could be achieved through trialing and identification of various innovative new methods and mechanisms (figure), including information and communication technology monitoring systems, commit ment indices, and social accountability mechan isms. One such method is the PolicyMaker software for analysis of the political economy of nutrition.51 Indices of a countrys progress towards particular goals, such as the UN Human Development Index and the International Food Policy Research Institutes Global Hunger Index, are increasingly common in development and, if methodologically sound, can be a useful focal point for civil society advocacy.52,53 The pros and cons of such indices have been evaluated with the conclusion that a separate index that measures political will and
www.thelancet.com

Outcomes Accelerated reductions in stunting Data guide need for intensication of commitments Real Time Monitoring

Accountability To incentivise improvements

Methods for building of commitment to reduce undernutrition

Data promote feedback about performance Community score cards

Responsiveness Capacity to improve performance Data guide change in strategy Real Time Monitoring

Figure: Examples of methods to improve the commitment, accountability, and responsiveness to undernutrition reduction Reproduced from reference 8, by permission of Palgrave Macmillan.

87

Series

Panel 5: The Nutrition Commitment Index Nutrition outcomes are the result of many factors that governments do and do not have control of. Climate change and associated droughts and floods, and cross-border issues such as arms and drugs trading, mass migration, and capital flight can have enormous effects on nutrition outcomes. Conversely, the commitment to nutrition can be generated and shaped by governments, and should, if informed by evidence, be a positive force for future undernutrition reduction. If commitment can be measured, can it be used to strengthen accountability? The Nutrition Commitment Index (NCI) is the first attempt to measure government commitment to reducing rates of undernutrition. The index combines secondary data for 12 indicators across three domains (spending, policies, and legislation) at three levels (direct [nutrition-specific] interventions, indirect [nutrition-sensitive] interventions, and the fundamental drivers) to construct an overall index. The 2012 NCI results rank, in order, Guatemala (most commitment to undernutrition reduction), the Gambia, Nepal, Mozambique, Bangladesh, Malawi, Brazil, Indonesia, Madagascar, Tanzania, Peru, and the Philippines as the top 12 of 45 countries for which recent data are available. India, the country which has a third of the undernutrition burden, is in the bottom half of the 45 countries on commitment to reduce undernutrition. The appendix shows case studies for Peru and Malawi. When the NCI ranks are set against a countrys nutrition outcome indicators, we can see how the index might be used to guide resources. In countries where commitment is low and undernutrition rates are high, some resources need to be allocated towards strengthening of commitment. Where both commitment and undernutrition rates are high, most resources can be allocated to the scale-up of and capacity to deliver nutrition programmes. Although the countries that do well on the NCI do have high levels of stunting, they have some of the fastest declines in stunting rates over the past 20 years. The top 12 countries show a decline in stunting rates between the 1990s and 2000s that is twice as high as the remaining countries. Additionally, the ranks show that the commitment to hunger reduction and the commitment to malnutrition reduction are only weakly correlated: a commitment to hunger reduction does not automatically equate to a commitment to malnutrition reduction. Future econometric work will rigorously explore the associations between nutrition outcomes and nutrition commitment, with attention on other independent variables, which could explain stunting and the time lags between changes in commitment and changes in stunting. Future qualitative work will focus on whether and how the NCI helps mobilise commitment for undernutrition reduction.

For more on commitment to reduce undernutrition and on stunting trends see http://www. hancindex.org

key drivers of undernutrition and relevant remedial options, and (4) implementation of nutrition programmes and provision of delivery platforms to maximise scale-up and ensure equity by reaching the unreached. Organ isations should also be held accountable for their commitment and performance in reducing malnutrition. The table outlines key roles and principles of civil society and private sector engagement in nutrition.

Private sector engagement: maximising potential and managing risks


The scale, know-how, reach, financial resources, and existing involvement of the private sector in actions that determine nutrition status is well known. The share of food and health care purchased through the market is increasing steadily, at all levels of income. This increase has partly taken place because malnutrition exists at all income quintiles and because companies are looking to the base of the pyramidie, to the poorest socioeconomic groups62,63to expand market share64 if the initial market size is large enough.62,65 Private sector involvement in food and health-care choices goes well beyond the large multinational food and pharmaceutical companies. Agri-food businesses, medium-scale and small-scale pro cessors of staple foods, and private health networks now have an active involvement in the production, marketing, and consumer choice in the purchase of food and other nutrition-relevant goods and services.66 Other
88

develop ments increase the opportunity for the private sector to contribute to acceleration of malnutrition reduc tion. For example, new private philanthropic support for develop ment has expanded,67 logistics and information and computer technology businesses have emerged, and m-health (health services using mobile technologies) initiatives have flourished, with benefits to service delivery and care management.68 New forms of publicprivate partnerships have emerged in the health sector from which lessons can be learned about how to identify a balance of interests and incentives among partners.69 As a result of these many public and privatesector intersections, the interest of the public sector towards business involvement in undernutrition efforts has increased substantially. The SUN Business Network is one indication of this change in interest.70 The fourth paper in the 2008 Lancet Series acknow ledged the inextricable role of the private sector and its importance, but also called for additional evaluation of effectiveness and documentation of best practices.71 However, although the private sector is now even more important in the national nutrition system, too few independent and rigorous evaluations have been done of the effectiveness of involvement of the commercial sector in nutrition. In the absence of such evaluations, distrust of the private sector, especially the food industry, remains high and is somewhat linked to the decadeslong tension related to the marketing of breast-milk
www.thelancet.com

Series

Civil society Framing, generation, and communication of knowledge and evidence Surveillance to generate data showing severity and distribution of undernutrition Global and national advocacy; framing and packaging of information to galvanise commitment and push nutrition up the development agenda

Private sector Generation of evidence about the positive and negative effects of private sector and market-led approaches to nutrition Building of recognition of how the private sector already strongly determines nutrition status (food, pharmaceutical sectors, health care) Assurance of (cost-shared) monitoring and evaluation and absolute transparency of any publicprivate endeavour (including open data and open-access research) The public sector (elected governments) should set a regulatory framework and policy direction; national nutrition plans are needed Need to positively shape the substantial and existing effect of the private sector, to harness innovation (eg, mobile health and other information and communication technologies in nutrition), and to explore any comparative advantage in goods and service delivery Harness extensive private sector resources (including consumer spending) by creation of demand-side and supply-side incentives for nutritious foods and provision of health care and sanitation services (eg, publicprivate partnerships for new nutritious products, the potential for cobranding, and price guarantees) Improved public sector and private sector capacity to understand the potential contributions, opportunities, weaknesses, and threats

Political economy of stakeholders, ideas, and interests

Assurance of accountability of different stakeholders (including civil society organisations themselves) for coverage, quality, and equity of actions to reduce undernutrition Contribute to multistakeholder platforms for decision making (eg, in-country support of the Scaling Up Nutrition movement) Strengthening of the voice of communities, women, and children Extra layer of capacity to deliver services and reach marginalised communities Ability to raise financial resources through effective public campaigning

Capacity (individual, organisational, systemic) and financial resources

Table: Key roles and principles of civil society and private sector engagement in nutrition

substitutes in developing countries and sugar-sweetened beverages and fast foods worldwide.72 Much of the private-sector dialogue centres around the International Code of Marketing of Breast-milk Substitutes (ie, how to enforce it and the extent of its domain73,74) and around whether the Codex Alimentarius food and nutrient standards give businesses too much freedom to downgrade nutrition concerns.75 Some commentators have argued that particular inter pretations of the code have almost completely driven the private sector out of efforts to improve the nutrition of children aged 624 months.73 But caution is essential in view of the continued code violations by several large-scale privatesector enterprises.74 A troubled history combined with continued violations makes it increasingly difficult for the private sector to be a major contributor to the collective creation and sustenance of momentum for malnutrition reduction. This sector has yet to earn the trust of some groups of the nutrition community. In view of the needs and the considerable resources, effect, and convening power of the private sector, this opportunity might be missed. Additionally, opportunities exist for collaboration around advocacy, monitoring, value chains, technical and scientific collaboration, and fortification of staple foods that are uncontentious and deserve further exploration. When the interests of different participants are not perfectly aligned and when substantial information and power asymmetries exist, such as between large corporations and under-resourced governments, the search for winwin solutions for undernutrition and overweight and obesity is a matter of governance arrange ments: how rules are set, monitored, and enforced. Lessons need to be learned from the long experience of the regulation and legislation of fortified foods76 and from the experiences of publicprivate partnerships in international
www.thelancet.com

health,7779 which suggest that such solutions can be identified on the basis of sufficient trust and verification. Such experiences suggest that some urgency exists towards building of trust, especially around infant feeding. Recom men dations for building of trust in companies manu facturing infant formula feeds include establishment of a public register of meetings between companies and governments about the International Code of Marketing of Breast-milk Substitutes, strengthening of whistleblowing procedures within companies, and implementation of prevention of code violations into the job descriptions of companies senior representatives in each country.74 Governments need to play their part by enshrining the code and subsequent resolutions into national law, and putting independent, transparent, and effective monitor ing mechanisms in place.

Capacity and resources Leadership in nutrition


All the nutrition success storieseg, in Brazil, Peru, Vietnam, and Thailandhave strong and effective networks of national nutrition leaders at their core.21,80 For undernutrition reduction to be sustained, nutrition leaders at all levels should be able to forge strong alliances (across and between government, civil society, and the private sector), take timely and decisive action, and create and be subject to strong accountability. Enhancement of effective leadership needs investment and yet only a handful of courses in nutrition leadership are offered worldwide. Every year, the African Nutrition Leadership Programme, an African-led initiative, enrolls 30 participants for 10 days81less than one professional per African country per year. No nutrition leadership programme exists in south Asia; however, UNICEF Indias engagement with young political leaders through the Citizens Alliance Against
89

Series

Malnutrition seeks to strengthen political leader ship. Panel 6 summarises research done to identify what makes a champion in nutrition. Leaders and champions in nutrition need systemic and organisational capacity to create and sustain nutrition policy and institutional change. Again, civil society can play a strong part in this aspect as shown, for example, in Peru where civil society champions were linked with political and financial decision makers (appendix).

Understanding of the financial resources available to build commitment for nutrition


A focus on three areas is needed to make the case for additional resources to build and sustain momentum for undernutrition reduction: the cost, an understanding of present resource flows to nutrition, and more and better estimates than presently exist of benefit to cost ratios for nutrition investments at the country level. Answers to these questions could help convince financial analysts in the public and private sectors to invest. Estimations for the SUN movement clearly show the costs of addressing undernutrition via nutrition-specific interventions.92 More work is needed to contextualise and specify these costs for different countries and this work is ongoing. Unfortunately, investments in nutrition are hard to track because of the weak designation of donor and govern ment spending. For example, analyses of data from the
Panel 6: What makes a nutrition champion? In seeking to achieve large-scale, systemic changes to address undernutrition, several initiatives have recognised the important role of key individualsleaders, champions, catalysts, and policy entrepreneursin the development of beneficial policy changes.21,25,82,83 Because level of change does not necessarily correspond to levels of formal power, visibility, ambition, or technical knowledge, research is being done to identify and better understand the capacities and attributes of the individuals who have substantially contributed to policy advances for nutrition. This research is based on principles and concepts from complexity science and adult development. Through network and power mapping and consultations with key informants, relevant stakeholders were identified in Kenya and Bangladesh (about 75 stakeholders per country) and semi-structured interviews were done with a purposive sample of these stakeholders (30 in Kenya and 24 in Bangladesh). These interviews provided information to assess the attributes of the interviewee and other influential stakeholders (ie, self-reporting and peer-reporting) and provided further insights into network and power dynamics and case studies. In Kenya and Bangladesh, this research shows that a handful of catalytic individuals, well-connected and trusted in their formal and informal social networks,84,85 have played a crucial part in transfer of information, changing of perceptions, and resolving of conflicts; achievements that have proven essential to advance the nutrition agenda in the

Creditor Reporting System, which is maintained by the Development Assistance Committee of the Organisation ment, show that for Economic Cooperation and Develop a substantial proportion of spending designated as nutri tional, is actually being spent on non-nutrition projects. Similarly, much nutrition spending is in categories that are not nutritional.93,94 Data for donor spending on nutrition often do not match those of governments (unpublished). However, costbenefit estimates are quite favourable. With assumptions about the 11% uplift in income attributable to prevention of a third of stunting by age 3 years, and about the 5% discount rate of future benefit streams, average costbenefit estimates have been generated for 20 countries,6 with a median ratio of 18 (Bangladesh). These ratios compare extremely favour ably with other investments for which public funds compete.95 Findings from the COHORTS study96 reinforce the consensus that the first 1000 days is the key window of opportunity for investments. With data from five countries, the COHORTS investigators reported that the growth effects on human capital are largest at age 2 years. The most powerful way of building commitment to increased resource allocation to nutrition could be shown in the example set by countries that have achieved scale-up. The three case studies identified in the appendix provide examples of

context of fragmentation and competing interests between and within various groups of stakeholders.23,82,86,87 Preliminary findings show that these individuals have, in addition to extensive knowledge and experience in nutrition, relatively strongly developed stakeholder awareness and perspective awareness. They show an understanding of the stakeholders relevant to nutrition policy processes and the associations among them, and tend to view the properties of their own and others perspectives as perspectives with complex contributory causes. Patterns of sense making generally shape ones goals and activities,8890 and the catalytic individuals identified in this study tended to identify ways in which shifts in stakeholder views and associations can lead to positive outcomes; genuinely adapt behaviour, language, and framing of issues to different stakeholders; and focus on establishment of associations of mutual trust, rather than unidirectional forms of influence. This research, led by the Transform Nutrition consortium), is ongoing and will continue in Ethiopia and India. Future research will focus on the ways to move beyond identification and assessment of champions to evaluating ways of supporting them, including through training and capacity building, curricula development, public recognition and support for identified champions (eg, through awards and scholarships); and the development of competency frameworks and institutional and workplace incentives.91

For more on the Transform Nutrition consortium see http:// www.transformnutrition.org

90

www.thelancet.com

Series

what can be done to improve nutrition. A few SUN countries should show that increased commitment can be turned into real resultssuch examples will act as a spur for many others.

Monitoring of programme coverage


Inherent in the SUN process is the acknowledgment that programme coverage of nutritionally vulnerable popu lations has to increase from very low levels; how ever, routine mechanisms to monitor nutrition-related inter vention coverage worldwide are poorly designed. Research of child survival100 has shown the large gap in scale-up of evidence-based interventions for maternal, newborn, and child survival, many of which have sub stantial benefits for nutrition, but several nutrition indicators are not yet embedded in these monitoring processes. WHOs Nutrition Landscape Information System123 needs to be strengthened by generation of a consensus on, and expansion of the range of, inter ventions to be tracked.

Conversion of momentum into results


Knowledge and evidence
Building of momentum for undernutrition reduction is not an easy task, nor is it sufficient; such momentum needs to be translated into ground-level results. Again the three dimensions of an enabling environment come into play: knowledge and evidence about how to scale up interventions in an effective way, the political economy behind the interplay between national and subnational levels of government, and the capacity and resources needed to scale and expand coverage of programmes while retaining cost-effectiveness.

Programme evaluations to learn and improve


Programme evaluations play a crucial part in inform ing the scale-up, reconfiguration, or cessation of pro grammes. Solid guidance now exists to bring rigour to evaluations of nutrition programmes.52,100,124126 This gui dance is needed to create solid ground for evaluation of the progress, and pathways to progress, of nutrition interventions,110,127,128 with theory-driven and qualitative evaluations exploring the whys and hows of progress and the extent.100,129 Analyses of effectiveness and operational evaluations of innovations that are introduced into scaled-up programmes, or of the process of scale up of innovative programmes from small-scale pilots to a large scale, are essential, but challenging.

Implementation research: what works, why, and how?


Despite calls to action,97,98 and by stark contrast with the Countdown to 2015 report on maternal, newborn, and child survival,99 no systematic process is in place for collation of the implementation-related evidence base about how to scale up the vast array of nutrition-specific and nutrition-sensitive interventions with quality and equity. Development of this scientific literature needs careful attention to several factors, but perhaps most importantly, needs a relentless focus on unpacking of programme impact pathways to effects71,100,101 and docu menting of contextual factors that affect implemen tation. Comprehensive frameworks already exist to provide insights into the types of process-related and contextual factors that need to be further studied through implemen tation research. A process convened by the New York Academy of Sciences and WHO for setting of research agendas in nutrition emphasises crucial gaps and a framework to undertake implemen tation research in nutrition.102 Examples of such research have been emerging in the form of feasibility studies and formative research,103106 operations research and process evaluations,107111 and costing studies.112114 How ever, the scientific literature about implementation through delivery platforms, such as community-based or health-facility-based programmes, is more developed than is that of the use of mass media or market-based approaches to scale up interventions.115,116 Much implementation research is from small-scale interventions, as opposed to large-scale programmes or interventions, for which the challenges to ensurance of quality, intensity, equity, and coverage are different and need various factors to operate in concert.117122 Analyses of scaled-up programmes or of scale-up of small area or pilot interventions raise several challengeseg, estab lishing of counterfactuals, assurance that real-time process documentation captures nuances of organi sational changes that facilitated or hampered scale-up, and that research generated is of a publishable quality.
www.thelancet.com

Learning during crisis


Increases in the frequency of natural disasters130 and the persistence and repeated cycles of conflict131 raise humanitarian needs and stifle progress in reduction of undernutrition in fragile contexts. The need for effec tive surveillance; early warning; mitigation; and timely, appropriate, and effective responses to nutrition-related crises is greater than ever. Yet little new evidence has been generated of the effectiveness of emergency interven tions since the first Lancet Series was publishedpartly because of persisting ethical concerns and conceptual and practical difficulties posed by research in such situations.132 The time has come for increasing recognition of government accountability to lead in the provision of services that are needed to meet short-term emergencyrelated spikes in demand.133 This situation creates a growing tension between stakeholders who are driven by the humanitarian imperative to deliver timely and effective assistance, and those who seek to strengthen government systemic capacity to lead general efforts to scale up nutrition-related interventions and services. A pertinent example is the community-level treatment of wasting, which in the past decade has moved from being a programme led by non-governmental organisations, to a service integrated within national health systems, which is intended to be accessible to children in need throughout
91

Series

the year.134 How to enable such systems to protect and reinforce the resilience of populations in fragile contexts, and to create a surge in response to increases in acute needs, is still a major challenge.

Political economy and governance Subnational governance


Just as building and sustaining of commitment is a political process, so too is conversion of momentum into results. Political scientists often conclude that most policy is formulated at the front line, and the situation should be no different for efforts to reduce undernutrition. Findings from the six country nutrition governance study21,49 suggest that in addition to the key ingredients for building of momentum, a further five are crucial to generate change: (1) local government capacity to deliver effective nutrition services, (2) local politicians who care about nutrition and are empowered via decentralised budgets and knowledge that nutrition can be a vote winner, (3) timely data for undernutrition, (4) nutrition funding channelled through one funding mechan ism rather than fragmented funding streams, and (5) earmarked and protected nutrition funding commit ments and exploration of new revenue streams. The findings for local government incentives and capacity are highly relevant because many countries in Asia and sub-Saharan Africa are rapidly moving to decentralised political, administrative, and financial systems. Decentralisation necessitates building of commit ment and capacity at various political and bureaucratic levels at which decisions are made and resources allocated. Although scientific literature is emerging for decentralisation of health systems,77,135 the research base is limited to a handful of studies.20,27,135137 In Vietnam for instance, the role of provincial planning for nutrition has been identified as an important bottleneck to translation of national policy intent and frameworks into plans and actions at the provincial level.27

nutrition initiatives in the past (eg, LINKAGES) and of some health-sector initiatives (eg, Integrated Manage ment of Childhood Illness), published works of what is needed, and how to achieve integration, are scarce. Integration of nutrition into other sectors, which are less oriented to nutrition, is hampered by issues related to motivations, capacities, and clear guidance.20 Therefore, building of experiential learning and systematic evidence about processes related to intersectoral and multisectoral integration of actions is urgently needed to reduce undernutrition.

Private sector engagement


Several promising areas for private sector engagement in nutrition value chains have been summarised in the past few years.66,140 Similarly, many promising non-peer reviewed case studies exist about how food fortification not only generates sales and reputational gains for businesses, but also nutritional benefits via increased consumption of fortified foods.141,142 The potential of other types of private sector companies to contribute to nutrition scale-up is also considered important (eg, via mobile technology providers). A major constraint in realisation of this potential is the dearth of independent peer-reviewed studies of such activities and the com plete absence of any review of the available evidence, although a review is underway by the Trans form Nutrition consortium. Of peer-reviewed studies relating to the first 1000 days of life, one noted that marketing and selling of multi nutrient powders in China to the caregivers of children aged 624 months reduced the risk of anaemia by 87%.143 Another reported decreases in iron and vitamin A deficiency in children aged 635 months in western Kenya from the sale of multinutrient powders via community vendors.144 The private sector has a part to play in the provision of fortified foods that could assist in addressing undernutrition. Attention should be paid to, and guidance be given to, the appropriate marketing of complementary foods for young children older than 6 months, that both protects breastfeeding and allows for caregivers to make informed choices from available fortified complementary foods. Beyond direct private-sector support through core business operations and investments, many individuals have argued that the sector has a much broader responsibility to ensure the health, nutrition, and welfare of their workforces and the larger communities that are dependent upon them. The creation of shared value approach is intended to be achieved through creation of economic value via companys policies and operating practices, with simultaneous advancement of the economic and social conditions in the communities in which it operates.145 Results of this approach should be carefully monitored, and best practices underscored, through initiatives such as the Ethical Trading Initiative, which none of the leading food and beverage companies on the Access to Nutrition Index have signed up to.146
www.thelancet.com

Intersectoral action
The wide recognition that action from several sectors is needed to address nutrition has gained momentum, and several country governments are implementing multi sectoral and intersectoral plans. However, few examples exist of the factors and processes that should align to enable intersectoral action to generate scaled up nutrition-specific interventions and a nutrition-sensitive household and community environment in which provisions for water, sanitation, social protection, health care, and food security are ensured. Research so far has been of intersectoral planning and action at a policy level,16,138 whereas several questions remain about how best to achieve such outcomes at subnational and local levels. Even integration of nutrition actions within the health sector (which is arguably the most ready to absorb nutrition actions) often raises many challenges.139Although such integration has been the focus of several large-scale
92

Series

On the basis of the suggested guidelines in the table, and of insights from other sectors,147 several factors are key to maximising the private sectors potential con tribution to nutrition status with minimisation of the risks to vulnerable populations: (1) understand the bottle necks that the private sector could help overcome; (2) incentivise positive roles and the development of business models that support them; (3) regulate ongoing activities for potential risks to nutrition, with strong monitoring processes; (4) be transparent about the role of the private sector in the policy process and any potential conflicts of interest; and (5) independently evaluate publicprivate partnership activities and make the data and analyses publicly available.

Panel 7: Key issues and core elements of nutrition-relevant capacity Individual capacity: methods and skills Performance capacity: are the methods, money, and equipment, for example, available to do the job? Personal capacity: are staff sufficiently knowledgable, skilled, and confident to perform properly? Do they need training, experience, or motivation? Are they deficient in technical, managerial, interpersonal, or specific role-related skills? Organisational capacity: staff and infrastructure Workload capacity: do enough staff have broad enough skills to cope with the workload? Are job descriptions practicable? Is skill mix appropriate? Supervisory capacity: are reporting and monitoring systems in place? Are lines of accountability clear? Can supervisors physically monitor all staff? Are effective incentives and sanctions available? Facility capacity: are training centres, offices, and workshops big enough, with the right staff in sufficient numbers, to support the workload? Support service capacity: are there training institutions, supply organisations, building services, administrative staff, research facilities, quality control services? Systemic capacity: structure, systems, and roles Structural capacity: are there decision-making forums or multistakeholder platforms at which intersectoral discussion of nutrition could take place, consensus is generated, collective decisions are made and recorded, and individuals called to account for non-performance? Systems capacity: do flows of information, money, and managerial decisions happen in a timely and effective manner? Are proper filing and information systems in use? Can private sector services be contracted as needed? Is there good communication with the community? Are links with non-governmental organisations sufficient? Role capacity: have individuals, teams, and committees been empowered to make decisions to ensure effective performanceeg, regarding schedules, money, and staff appointments?

Capacity and resources Sequencing and prioritisation of nutrition actions


Ideally, all the links in the nutrition chain would be addressed at the same time; if this is not possible for resource, capacity, or political reasons, priorities need to be set. A frequently heard complaint from ministries of finance in high-burden countries is what to do first when it comes to stimulation of economic growth? In response to this question, a group of researchers at Harvard University and elsewhere have developed an economic growth diagnostics process.148 The process combines evidence about the technical (what works here?), capacity (can we scale-up?) and, importantly, the political (are there any windows of opportunity for change?) aspects. The rationale is that sequencing matters and some issues can be highly rate-limiting. With nutrition, specific factors need to be in place for specific processes to take place. Similar nutrition diagnostic methods need to be developed to help prioritise nutrition plans of action.

Capacity for scale up


Several types of capacity are needed for effective scale up of priority nutrition interventions (panel 7149). Insights can also be gained from the wider scientific literature of human resources in health systems research,119 including the need to agree on exactly what should be scaled up, consider lessons on scale-up from related areas, honestly document experiences, and understand that scaling up of interventions requires a scaling down of certainties, and inclusiveness and building of relations to sustain momentum. Finally, we suggest that the existence of poor quality training programmes and academic curricula in nutrition in regions of poor quality service delivery is not a coincidence.150153 Many of these studies are from high-burden regions and they find the training and curricula to be outdated, impractical, and misaligned with local nutrition priorities. We reiterate the conclusion of the 2008 Series that much more needs to be done to strengthen strategic and operational capacity.71,154 Govern ments and donors should allocate more resources to establish a more sustainable foundation for nutrition implementation by training the next generation of
www.thelancet.com

implementers who in turn will be mentors for the generation after that.

Financial resources to support scale up


The second paper in this Series estimates that at least Int$96 billion per year will be needed to scale up the 11 proven nutrition-specific interventions for the 34 countries that account for 90% of the burden of stunting.5 If this scale-up could be achieved, at least a quarter of present stunting cases could be addressed.7 Paper two suggests that roughly $3 billion to $4 billion of this total could come from external donors and, as SUN requires, would work together with established guidelines for aid effectiveness, including the importance of country ownership and the avoidance of aid dependency. Scaling up of nutrition programmes continues to be the place to start to reduce malnutrition; however, we need estimates of what it would take to make agriculture, social protection, education, and womens empowerment policies and pro grammes, for example, sufficiently nutrition-sensitive to have a further substantial effect on malnutrition rates. Paper three provides some suggestions about how to reallocate nutrition-sensitive programme resources to achieve winwin solutions. The extra resources needed to
93

Series

Panel 8: Research priorities to build commitment and accelerate progress Framing, generation, and communication of knowledge and evidence Creation and sustaining of momentum for undernutrition reduction What types of issue framing approaches and narratives yield attention to nutrition in different contexts? What advocacy and policy engagement strategies are most effective at galvanising political attention to nutrition? What types of evidence are most powerful for creation versus sustaining of national and subnational attention to nutrition? Can real-time monitoring of nutrition outcomes and coverage lead to more responsive nutrition actions and improved nutrition outcomes? Conversion of momentum to effect on nutrition status How can nutrition interventions be mainstreamed and integrated into other sectors What types of programme evaluations and operations research are crucial to enabling programmatic actions at different stages in the life of nutrition investments? What types of learning mechanisms best enable inclusive stakeholder engagement with evidence? What types of stakeholder engagement approaches can enhance the demand for evidence of effect? Political economy and governance of stakeholders, ideas, and interests Creation and sustaining of momentum for undernutrition reduction What strategies are most effective at enabling multisectoral coordination and strategic coherence for nutrition? Which accountability strategies are most effective at mobilising commitment at different levels of government and society (eg, indices, scorecards, social audits, community monitoring)? In what ways can the private sector be regulated to protect and support exclusive breastfeeding? Conversion of momentum to effect on nutrition status What aspects of decentralisation are most crucial for enabling vertical translation of national guidance to programmatic action? What types of roles can (and should) the private sector and civil society have in supporting service delivery and scaling up? When has private sector involvement enhanced nutrition status and how? Do effective accountability mechanisms contribute to improved nutrition outcomes? What are effective incentives to help mainstream nutrition into potentially nutrition-sensitive sectors? Capacity (individual, organisational, systemic) and financial resources Creation and sustaining of momentum for undernutrition reduction What are the characteristics of nutrition policy champions? What effect do university curricula and leadership training investments have in creation of nutrition leaders? What types of institutional investments and capacity building activities yield the best systemic and strategic capacity for nutrition within national and subnational organisations? How should the resources allocated to nutrition-sensitive programmes be assigned to nutrition improvement? To what extent can research on the costing of interventions and the tracing of financial flows mobilise additional resources for nutrition and improve the effectiveness of resource allocation? What methods are effective in helping to prioritise and sequence nutrition actions? Conversion of momentum to effect on nutrition status What institutional and front line capacities are most important to enable scale-up of different types of direct nutrition interventions through community-based progammes and the health sector? How can nutrition-sensitive sectors operationalise their interventions to achieve nutrition results for women and children? Which new forms of resource mobilisation show the greatest promise for improvement of nutrition status? Does prioritising and sequencing of simple nutrition actions (eg, vitamin A supplementation, micronutrients, treatment of severe acute malnutrition) create enabling conditions for closing gaps on more complex interventions (behaviour change interventions for infant and young-child feeding?).

incentivise such reallocations might well be modest, but more experience and evidence are needed to identify the surplus requirement. The allocation of scarce public resources between nutrition and other activities (and indeed among nutrition activities) will be guided by political and technical considerations. Nutrition tends to have no institutional champion, hence the emphasis within SUN on institu tional mechanisms to address this issue, both formally (via multisector platforms) and informally via the framing around movements. Other mechanisms for promotion and protection of nutrition spending exist, such as the example in Peru of embedding of nutrition within electoral commitments (appendix),
94

and more analysis is needed about the variety and effectiveness of these mechanisms. We previously discussed private sector possibilities for additional resources. For public sources, highburden countries together with donors and multilateral organisations have a responsibility to increase allo cations to nutrition-specific and nutrition-sensitive programmes. To do this within an official development assis tance budget that has peaked, albeit with increasing tax revenues from high-burden countries,155 will be politically challenging, hence the need to build leadership, commit ment, and accountability at national and international levels.
www.thelancet.com

Series

However, the gap is unlikely to be closed from these sources. Innovation is needed across all sectors to leverage private-sector and public-sector resources and to generate additional public funding. The nutrition sector can draw on several innovative ideas from other sectors,156 including advance market contracts to ment, market levies, and taxes, on either promote invest unhealthy externalities or external sectors, as in the airline ticket levy by UNITAID157 or the mining levy funding health in Zambia. Nutritional impact bonds are another option, entailing the creation of a social impact partnership fund by private investors, which receives public funds if key service delivery targets are met. In this way, public funds catalyse and leverage private investment for which the service providers bear the risk, but also stand to generate additional revenue. Key to the success of these schemes is the collection of cred ible metrics. More research and experimentation is desperately needed in this area.

Looking ahead
In the past 5 years, the nutrition community has made major progress, but it should be judged against the effect emerging in the next 5 years and beyond. Momentum needs to be sustained and converted into lasting effects. SUN will reach its 1000th day when this Lancet Series is launched in June, 2013. Since SUNs own launch in September 2010, the movement has substantially elevated and energised the discourse on nutrition and has changed institutional arrangements. In some coun tries, the movement is beginning to catalyse resource mobilisation and programme alignment. Emphasis should now escalate to action, translating commitment into results on the ground. SUN needs to build on its commitment to be country led and results driven. To enable this development, SUN should harness and catalyse national leadership, capacity and resources, politics, and knowledge generation. Documented SUN proof-of-concept success stories are also needed to galvanise further action. One clear overarching priority is the need to strengthen strategic and operational capacity to scale up nutrition interventions and embed nutrition considerations in other sectoral actions. This point was emphasised in the 2008 Lancet series and remains the case today.156 National and global resources need to be invested in the long term to support capacity development, at individual, organ isational, and systemic levels. Leadership is needed to galvanise and spearhead action, and this again will need to be seeded, funded, and nurtured. For too long the issue of capacity has been recognised but overlookeda convenient excuse for failed plans. It is easy to neglect such issues when constructing business plans to support nutrition strategies and yet without sufficient capacity of the right type at the right level, plans become hollow wish lists. Fairly silent to date, the nutrition community needs to be a lot more engaged in the post-2015 process to ensure
www.thelancet.com

that interest in nutrition is locked into the post-2015 development settlement (panel 4). If nutrition is to be embedded into broader development processes, the nutrition community needs to actively forge alliances with those for whom malnutrition reduction is not a top priority and to do this in a politically aware manner. We have drawn on a range of evidence in this report, both academic and from the field. The academic evidence we used is valuable, but much of it is from areas outside of nutrition. We call for more research of what defines enabling environments for nutrition. We also call for more systematic ways to capture and share the learning from policy and programme operations. Panel 8 shows priority areas for research. Finally, the core problem itself is changing as the burden of disease caused by poor nutrition continues to shift from undernutrition to a double burden of undernutrition and overweight and obesity.7 Future Lancet series on nutrition will have to pay much greater attention to this double burden than we have. But the disease burden attributable to child underweight remains substantial in many countries, in other words, there is an enormous unfinished agenda.
Contributors All authors contributed to the conceptualisation and structuring of the paper, and reviewed all drafts. SG led overall development and finalisation of the paper and appendix, and coordinated inputs of coauthors and contributors. He led on the enabling environments section, Scaling Up Nutrition (SUN) electronic consultation, and case studies; prepared the first draft of the conceptual framework; wrote the sections about capacity, civil society, the abstract, and introductory and closing sections; contributed to other sections of the paper; and prepared and submitted the final draft. LH first proposed to focus the paper on the politics of nutrition and led on drafting the sections about the wider context, financial resources, horizontal and vertical coordination, the private sector, the nutrition commitment index, and the post-2015 discussions; and had responsibility for construction of the first complete draft of the paper. VM prepared background sections on private sector, SUN, civil society, and financing, which were used in the development of those sections, and made contributions to each draft. PM wrote sections about implementation research and translation of commitment to action, and provided written inputs to the private sector and enabling environment sections. NN prepared the accountability and global governments sections and the Peru case study, contributed to the enabling environments section and coedited one of the first full drafts. Four of the authors are researchers and one (VM) is a practitioner. The researchers combine expertise in nutrition, economics, politics, sociology, and anthropology. All authors are embedded in the global policy community and all are very familiar with one or more country policy communities. All authors operate at the interface between policy, practice, and research and three have done so since the 1980s. This is the basis for their collective knowledge. Maternal and Child Nutrition Study Group Robert E Black (Johns Hopkins Bloomberg School of Public Health, USA), Harold Alderman (International Food Policy Research Institute, USA), Zulfiqar A Bhutta (Aga Khan University, Pakistan), Stuart Gillespie (International Food Policy Research Institute, USA), Lawrence Haddad (Institute of Development Studies, UK), Susan Horton (University of Waterloo, Canada), Anna Lartey (University of Ghana, Ghana), Venkatesh Mannar (The Micronutrient Initiative, Canada), Marie Ruel (International Food Policy Research Institute, USA), Cesar Victora (Universidade Federal de Pelotas, Brazil), Susan Walker (The University of the West Indies, Jamaica), Patrick Webb (Tufts University, USA).

95

Series

Series Advisory Committee Marc Van Ameringen (Global Alliance for Improved Nutrition, Switzerland), Mandana Arabi (New York Academy of Sciences, USA), Shawn Baker (Helen Keller International, USA), Martin Bloem (United Nations World Food Programme, Italy), Francesco Branca (WHO, Switzerland), Leslie Elder (The World Bank, USA), Erin McLean (Canadian International Development Agency, Canada), Carlos Monteiro (University of So Paulo, Brazil), Robert Mwadime (Makerere School of Public Health, Uganda), Ellen Piwoz (Bill & Melinda Gates Foundation, USA), Werner Schultink (UNICEF, USA), Lucy Sullivan (1000 Days, USA), Anna Taylor (Department for International Development, UK), Derek Yach (The Vitality Group, USA). The Advisory Committee provided advice in a meeting with Series Coordinators for each paper at the beginning of the process to prepare the Series and in a meeting to review and critique the draft reports. Conflicts of interest REB serves on the Boards of the Micronutrient Initiative, Vitamin Angels, the Child Health and Nutrition Research Initiative, and the Nestl Creating Shared Value Advisory Committee. VM serves on the Nestl Creating Shared Value Advisory Committee. The other authors declare that they have no conflicts of interest. As corresponding author, Stuart Gillespie states that he had full access to all data and final responsibility for the decision to submit for publication. Acknowledgments Funding for the preparation of the Series was provided to the Johns Hopkins Bloomberg School of Public Health through a grant from the Bill & Melinda Gates Foundation. The sponsor had no role in the analysis and interpretation of the evidence or in writing the paper and the decision to submit for publication. We thank the contributors for their specified contributions, and the Bill & Melinda Gates Foundation for financial support for the preparation of this paper. SG received some funding from the Transform Nutrition Research Programme Consortium with UK aid from the UK Government. The development of the Nutrition Commitment Index was made possible by support from UK Aid and Irish Aid. The views expressed in this paper do not necessarily reflect the UK Governments official policies. Representatives of the Bill & Melinda Gates Foundation and the UK Department for International Development took part in series preparation meetings. Neither donor had a role in conceptualisation, data collection, analysis, or drafting of the paper. Many individuals have contributed to this paper beyond the coauthors. We thank Jody Harris for the literature review of nutrition policy processes and Jody Harris and Andrew Kennedy (both International Food Policy Research Institute [IFPRI]) for support in the production of the report; Tom Barker, Jessica Meeker, Karine Gatellier, and Adrian Bannister (Institute of Development Studies [IDS]) for organising the online SUN discussions in six countries; Mara van den Bold and Andrew Kennedy (IFPRI) for preparation of daily summaries of these discussions, and for other background work on SUN (appendix); Elise Wach and Dolf te Lintelo (IDS) for panels on nutrition champions and Nutrition Commitment Index; Victor Aguayo (UNICEF India) and Kavita Singh (IFPRI New Delhi) for the Maharashtra case study, Andrew Kennedy for the Malawi case study, and Andres Mejia Acosta (IDS) for the Peru case study; and Victoria Sibson (Save the Children UK) for contributions on nutrition crises. References 1 Di Ciommo M. The aid financing landscape for nutrition. Bristol, UK: Development Initiatives, 2013. 2 G-8. Camp David accountability report: actions, approach and results. Camp David, MD: USAID, 2012. 3 Horton S, Alderman H, Rivera J. Hunger and malnutrition. Copenhagen: Copenhagen Consensus Center, 2008. 4 Scaling Up Nutrition (SUN). Scaling up nutrition: a framework for action. Washington, DC: UNSCN, 2010. 5 Bhutta ZA, Das JK, Rizvi A, et al, The Lancet Nutrition Interventions Review Group, and the Maternal and Child Nutrition Study Group. Evidence based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet 2013; published online June 6. http://dx.doi.org/10.1016/ S0140-6736(13)60996-4.

6 7

8 9

10 11 12 13 14 15 16 17 18 19 20 21

22

23

24 25 26 27

28 29

Hoddinott J, Alderman H, Behrman JR, Haddad L, Horton S. The economic rationale for investing in stunting reduction. Washington, DC: IFPRI, 2013. Ruel M, Alderman H, and the Maternal and Child Nutrition Study Group. Nutrition-sensitive interventions and programmes. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S01406736(13)60843-0. Haddad L. How can we build an enabling political environment to fight undernutrition? Eur J Dev Res 2012; 25: 1320. Lim SS, Vos T, Flaxman AD, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 19902010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012; 471: 222460. Pelletier DL, Porter CM, Aarons GA, Wuehler SE, Neufeld LM. Expanding the frontiers of nutrition research: new questions, new methods, and new approaches. Adv Nutr 2012; 4: 92114. Haddad L, Alderman H, Appleton S, Song L, Yohannes Y. Reducing child malnutrition: how far does income growth take us? World Bank Econ Rev 2003; 17: 10731. Headey D. Developmental drivers of nutritional change: a cross-country analysis. World Dev 2013; 42: 7688. Masset E, Haddad L. Income growth and nutrition status: a critical review of the estimated relationship. Brighton, UK: Institute of Development Studies, 2013. Headey D. Pro-nutrition economic growth: what is it, and how do I achieve it? Washington, DC: IFPRI, 2011. Benson T. Improving nutrition as a development priority: addressing undernutrition in national policy processes in sub-saharan africa. Washington, DC: IFPRI, 2008. Garrett, JL. Natalicchio M. Working multisectorally in nutrition: principles, practices, and case studies. Washington, DC: IFPRI, 2010. Gillespie S, McLachlan M, Shrimpton R, eds. Combating undernutrion: time to act. Washington, DC: World Bank, 2003. Hill R, Gonzalez W, Pelletier DL. The formulation of consensus on nutrition policy: policy actors perspectives on good process. Food Nutr Bull 2011; 32: 92S104S. Hoey L, Pelletier DL. The management of conflict in nutrition policy formulation: choosing growth-monitoring indicators in the context of dual burden. Food Nutr Bull 2011; 32: S8291. Hoey L, Pelletier DL. Bolivias multisectoral Zero Malnutrition program: insights on commitment, collaboration, and capacities. Food Nutr Bull 2011; 32: 70S81S. Mejia Acosta A, Fanzo J. Fighting maternal and child malnutrition: analysing the political and institutional determinants of delivering a national multisectoral response in six countries. A synthesis paper. Brighton, UK: Institute of Development Studies, 2012 Natalicchio M, Garrett J, Mulder-Sibanda M, Ndegwa S, Voorbraak D. Carrots and sticks: the political economy of nutrition policy reforms. HNP Discussion Paper. Washington, DC: IFPRI and The World Bank, 2009. Pelletier DL, Frongillo EA, Gervais S, et al. Nutrition agenda setting, policy formulation and implementation: lessons from the Mainstreaming Nutrition Initiative. Health Policy Plan 2012; 27: 1931. Clark TW. The policy process: a practical guide for natural resources professionals. New Haven, CT: Yale University Press, 2002. Kingdon JW. Agendas, alternatives, and public policies. New York, NY: Harper Collins, 1995. Shiffman J, Smith S. Generation of political priority for global health initiatives: a framework and case study of maternal mortality. Lancet 2007; 370: 137079. Lapping K, Frongillo EA, Studdert LJ, Menon P, Coates J, Webb P. Prospective analysis of the development of the national nutrition agenda in Vietnam from 2006 to 2008. Health Policy Plan 2012; 27: 3241. Morris SS. Effective international action against undernutrition: why has it proven so difficult and what can be done to accelerate progress? Lancet 2008; 371: 60821. Subramanyam MA, Kawachi I, Berkman LF, Subramanian SV. Is economic growth associated with reduction in child undernutrition in India? PLOS Med 2011; 8: e1000424.

96

www.thelancet.com

Series

30 Mander H. Food from the courts: the Indian experience. IDS Bull 2012; 43: 1524. 31 Swaminathan M. Undernutrition in infants and young children in India: a leadership agenda for action. IDS Bull 2009; 40: 10310. 32 Measure DHS, USAID. StatCompiler. http://www.statcompiler.com (accessed March 30, 2013). 33 The World Food Prize. 2011 World Food Prize honors former Presidents of Ghana, Brazil. June 21, 2011. http://www. worldfoodprize.org/en/press/news/index.cfm?action=display& newsID=13371 (accessed March 30, 2013). 34 National Development Planning Commission, Government of Ghana, UNDP Ghana. Ghana millenium development goals report. Ghana: UNDP, 2010. 35 Khan NC, Tuyen lD, Ngoc TX, Duong PH, Khoi HH. Reduction in childhood malnutrition in Vietnam from 1990 to 2004. Asia Pac J Clin Nutr 2007; 16: 274278. 36 Ha do TP, Feskens EJ, Deurenberg P, Mai LB, Khan NC, Kok FJ. Nationwide shifts in the double burden of overweight and underweight in Vietnamese adults in 2000 and 2005: two national nutrition surveys. BMC Public Health 2011; 11: 62. 37 Campbell AA, Akhter N, Sun K, et al. Relationship of homestead food production with night blindness among children below 5 years of age in Bangladesh. Public Health Nutr 2011; 14: 162731. 38 Thorne-Lyman AL, Valpiani N, Sun K, et al. Household dietary diversity and food expenditures are closely linked in rural Bangladesh, increasing the risk of malnutrition due to the financial crisis. J Nutr 2010; 140: 182S88S. 39 Berg M, Wariero J, Modi V. Every child counts: the use of SMS in Kenya to support the community based management of acute malnutrition and malaria in children under five. New York, NY: Columbia UniversityEarth Institute, 2009. 40 Blaschke S, Bokenkamp K, Cosmaciuc R, Denby M, Hailu B, Short, R. Using mobile phones to improve child nutrition surveillance in Malawi. Brooklyn, NY: UNICEF Malawi, 2009. 41 Hoddinott J, Maluccio JA, Behrman JR, Flores R, Martorell R. Effect of a nutrition intervention during early childhood on economic productivity in Guatemalan adults. Lancet 2008; 371: 41116. 42 Maluccio JA, Hoddinott J, Behrman JR, Martorell R, Quisumbing AR, Stein AD. The impact of improving nutrition during early childhood on education among Guatemalan adults. Econ J 2009; 119: 73463. 43 Victora CG, Adair L, Fall C, et al. Maternal and child undernutrition: consequences for adult health and human capital. Lancet 2008; 371: 34057. 44 Martorell R, Horta BL, Adair LS, et al. Weight gain in the first two years of life is an important predictor of schooling outcomes in pooled analyses from five birth cohorts from low- and middle-income countries. J Nutr 2010; 140: 34854. 45 Bloom DE, Canning D. Demographics and development policy. Cambridge, MA: Harvard University, 2011. 46 Black RE, Victora CG, Walker SP, et al, and the Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S01406736(13)60937-X. 47 Haddad L. How should nutrition be positioned in the post-2015 agenda? Food Policy (in press). 48 Field JO. Multisectoral nutrition planning: a post-mortem. Food Policy 1987; 12: 1528. 49 Haddad L, Mejia Acosta A, Fanzo J. Accelerating reductions in undernutrition: what can nutrition governance tell us? Brighton, UK: IDS, 2012. 50 World Bank. World development report 2004: making services work for poor people. Washington, DC: World Bank, 2003. 51 Reich MR, Balarajan Y. Political economy analysis for food and nutrition security. HNP Discussion Paper. Washington, DC: World Bank and SAFANSI, 2012. 52 Ravallion M. Mashup indices of development. Washington, DC: World Bank, 2010. 53 IFPRI, Welthungerhilfe, Concern Worldwide. The challenge of hunger: ensuring sustainable food security under land, water, and energy stresses. Washington, DC: Global Hunger Index, 2012. 54 Masset E. A review of hunger indices and methods to monitor country commitment to fighting hunger. Food Policy 2011; 36: S10208. www.thelancet.com

55 Access to Nutrition Index. http://www.accesstonutrition.org/ (accessed March 30, 2013). 56 Haddad L, Lindstrom J, Pinto Y. The sorry state of M&E in agriculture: can people-centred approaches help? IDS Bulletin 2010; 41: 625. 57 Swain B, Sen PD. Bridging the malnutrition gap with social audits and community participation. IDS Bulletin 2009; 40: 95102. 58 Berlan D, Shiffman J. Holding health providers in developing countries accountable to consumers: a synthesis of relevant scholarship. Health Policy Plan 2012; 27: 27180. 59 Molyneux S, Atela M, Angwenyi V, Goodman C. Community accountability at peripheral health facilities: a review of the empirical literature and development of a conceptual framework. Health Policy Plan 2012; 27: 54154. 60 Bjorkman M, Svensson J. Power to the people: evidence from a randomized field experiment on community-based monitoring in Uganda. Q J Econ 2009; 124: 73569. 61 Gaventa J, Barrett G. Mapping the outcomes of citizen engagement. World Dev 2012; 40: 2399410. 62 Segr J, Winnard K, Abrha TH, Abebe Y, Shilane D, Lapping K. Willingness to pay for lipid-based nutrient supplements for young children in four urban sites of Ethiopia. Mat Child Nutr 2013; published online Dec 13, 2012. DOI:10.1111/mcn.12022. 63 Prahalad CK. The fortune at the bottom of the pyramid: eradicating poverty through profits. New Jersey: Wharton School Publishing, 2010. 64 Nelson J. Business as a partner in overcoming malnutrition: an agenda for action. Washington, DC: GAIN, 2006. 65 Reardon T, Henson S, Gulati A. Links between supermarkets and food prices, diet diversity and food safety in developing countries. In: Hawkes C, Blouin C, Henson S, Drager N, Dube L, eds. Trade, food, diet and health: perspectives and policy options. Oxford, UK: Wiley-Blackwell, 2010: 11130. 66 Dangour AD, Diaz Z, Sullivan LM. Building global advocacy for nutrition: a review of the European and US landscapes. Food Nutr Bull 2012; 33: 9298. 67 McGregor A, Burns D, Waldman L, Watson N, Williamson R. The Bellagio Initiative: the future of philanthropy and development in the pursuit of human wellbeing. Brighton, UK: Institute of Development Studies, 2012. 68 Free C, Phillips G, Watson L, et al. The effectiveness of mobile-health technologies to improve health care service delivery processes: a systematic review and meta-analysis. PLOS Med 2013; 10: e1001362. 69 Brinkerhoff DW, Brinkerhoff JM. Publicprivate partnerships: perspectives on purposes, publicness, and good governance. Public Admin Dev 2011; 31: 214. 70 Scaling Up Nutrition. Business Network. 2012. http:// scalingupnutrition.org/about/the-global-movement/businessnetwork (accessed March 30, 2013). 71 Bryce J, Coitinho D, Darnton-Hill I, Pelletier D, Pinstrup-Andersen P. Maternal and child undernutrition: effective action at national level. Lancet 2008; 371: 51026. 72 Moodie R, Stuckler D, Monteiro C, et al. Profits and pandemics: prevention of harmful effects of tobacco, alcohol, and ultraprocessed food and drink industries. Lancet 2013; 381: 67079. 73 Rao V. Law on infant foods inhibits the marketing of complementary foods for infants, furthering undernutrition in India. BMJ 2012; 345: e8131. 74 Save the Children UK. Superfood for babies: how overcoming barriers to breastfeeding will save childrens lives. London, UK: Save the Children,2013. 75 International Baby Food Action Network. The scaling up nutrition initiative: IBFANs concern about the role of businesses. IBFAN, 2012. 76 Sablah M, Klopp J, Steinberg D, Baker S. Private-public partnerships drive one solution to vitamin and mineral deficiencies: fortify west Africa. SCN News 2011; 39: 4044. 77 Bloom G, Kanjilal B, Lucas H, Peters DH. Transforming health markets in Asia and Africa: Improving quality and access for the poor. New York, NY: Routledge, 2013. 78 Harding A. Partnerships with the private sector in health: what the international community can do to strengthen health systems in developing countries. Washington, DC: Center for Global Development, 2009.

97

Series

79 Patouillard E, Goodman CA, Hanson KG, Mills AJ. Can working with the private for-profit sector improve utilization of quality health services by the poor? a systematic review of the literature. Int J Equity Health 2007; 6: 17. 80 Gillespie S, Mason J, Martorell, R. How nutrition improves. Geneva: World Health Organization, 1996. 81 African Nutrition Leadership Programme. http://www. africanutritionleadership.org/ (accessed March 30, 2013). 82 Heaver R. Strengthening country commitment to human development, lessons from nutrition. Directions in development. Washington, DC: The World Bank, 2005. 83 Mintrom M. Policy entrepreneurs and the diffusion of innovation. Am J Polit Sci 1997; 41: 73870. 84 Foster-Fishman P, Nowell B, Yang H. Putting the system back into systems change: a framework for understanding and changing organizational and community system. Am J Community Psych 2007; 39: 196215. 85 Brown BC. Conscious leadership for sustainability: how leaders with a late-stage action logic design and engage in sustainability initiatives. Santa Barbara, CA: Fielding Graduate University, 2011. 86 Pelletier D. Research and policy directions. In: Semba RD, Bloem W, eds. Nutrition and health in developing countries. Totowa, NJ: Humana Press, 2001: 52350. 87 Pelletier D. A framework for improved strategies. In: Gillespie S, McLachlan M, Shrimpton R, eds. Combating undernutrion: time to act. Washington, DC: World Bank, 2003. 88 Williams B, Hummelbrunner R. Systems concepts in action: a practitioners toolkit. Stanford, CA: Stanford University Press, 2010. 89 Benford R, Snow D. Framing processes and social movements: an overview and assessment. Ann Rev Soc 2000; 26: 61139. 90 Jordan T. Skillful engagement with wicked Issues: a framework for analyzing the meaning-making structures of societal change agents. Integral Rev 2011; 7: 4791. 91 Hughes R, Shrimpton R, Recine E, Margetts B. A competency framework for global public health nutrition workforce development: a background paper. World Public Health Nutrition Association, 2011. 92 Horton S, Shekar M. Scaling up nutrition: what will it cost? Washington, DC: World Bank, 2010. 93 Mutuma S, Fremont E, Adebayo A. Aid for nutrition: can investments to scale up nutrition actions be accurately tracked? London: ACF International, 2012. 94 Development Initiatives. Nutrition advocacy landscaping in Europe: an analysis of donor commitments. London: Development Initiatives, 2011. 95 Hoddinott J, Rosegrant M, Torero M. Hunger and malnutrition: investments to reduce hunger and undernutrition. Washington, DC: Copenhagen Consensus, 2012. 96 Adair LS, Fall CHD, Osmond C, et al, for the COHORTS group. Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: findings from five birth cohort studies. Lancet 2013; published online March 28. http://dx.doi.org/10.1016/S01406736(13)60103-8. 97 Shekar M. Delivery sciences in nutrition. Lancet 2008; 371: 1751. 98 Leroy JL, Menon P. From efficacy to public health impact: a call for research on program delivery and utilization in nutrition. J Nutr 2008; 138: 62829. 99 Countdown to 2015: maternal, newborn, and child survival. Building a future for women and children: the 2012 report. Washington, DC: World Health Organization and UNICEF, 2012. 100 Habicht JP, Pelto GH, Lapp J. Methodologies to evaluate the impact of large scale nutrition programs. Washington, DC: World Bank, 2009. 101 Marsh DR, Alegre JC, Waltensperger KZ. A results framework serves both program design and delivery science. J Nutr 2008; 138: 63033. 102 The New York Academy of Sciences. Global research agenda for nutrition sciences. March 8, 2013. http://www.nyas.org/ Publications/Ebriefings/Detail.aspx?cid=38eb8a81-ed18-49f4-b9d10a5d887e712c (accessed March 30, 2013). 103 Rasheed S, Haider R, Hassan N, et al. Why does nutrition deteriorate rapidly among children under 2 years of age? Using qualitative methods to understand community perspectives on complementary feeding practices in Bangladesh. Food Nutr Bull 2011; 32: 192200.

104 Paul KH, Muti M, Chasekwa B, Mbuya MN, Madzima RC, Humphrey JH, Stoltzfus RJ. Complementary feeding messages that target cultural barriers enhance both the use of lipid-based nutrient supplements and underlying feeding practices to improve infant diets in rural Zimbabwe. Mat Child Nutr 2012; 8: 22538. 105 Pelto GH, Armar-Klemesu M. Balancing nurturance, cost and time: complementary feeding in Accra, Ghana. Mat Child Nutr 2011; 7: 6681. 106 Pelto GH, Armar-Klemesu M, Siekmann J, Schofield D. The focused ethnographic study assessing the behavioral and local market environment for improving the diets of infants and young children 6 to 23 months old and its use in three countries. Mat Child Nutr 2013; 9: 3546. 107 Suchdev PS, Ruth L, Obure A, et al. Monitoring the marketing, distribution, and use of Sprinkles micronutrient powders in rural western Kenya. Food Nutr Bull 2010; 31: S16878. 108 Loechl CU, Menon P, Arimond M, et al. Using programme theory to assess the feasibility of delivering micronutrient Sprinkles through a food-assisted maternal and child health and nutrition programme in rural Haiti. Mat Child Nutr 2009; 5: 3348. 109 Bonvecchio A, Pelto GH, Escalante E, et al. Maternal knowledge and use of a micronutrient supplement was improved with a programmatically feasible intervention in Mexico. J Nutr 2007; 137: 44046. 110 Olney DK, Talukder A, Iannotti LL, Ruel MT, Quinn V. Assessing impact and impact pathways of a homestead food production program on household and child nutrition in Cambodia. Food Nutr Bull 2009; 30: 35569. 111 Pelto GH, Santos I, Goncalves H, Victora C, Martines J, Habicht JP. Nutrition counseling training changes physician behavior and improves caregiver knowledge acquisition. J Nutr 2004; 134: 35762. 112 Puett C, Sadler K, Alderman H, Coates J, Fiedler JL, Myatt M. Cost-effectiveness of the community-based management of severe acute malnutrition by community health workers in southern Bangladesh. Health Policy Plan 2012; published online Aug 9. DOI:10.1093/heapol/czs070. 113 Fiedler JL, Mubanga F, Siamusantu W, Musonda M, Kabwe KF, Zul C. Child health week in Zambia: costs, efficiency, coverage and a reassessment of need. Health Policy Plan 2012; published online Dec 14. DOI:10.1093/heapol/czs129. 114 Fiedler JL, Villalobos CA, De Mattos AC. An activity-based cost analysis of the Honduras community-based, integrated child care (AIN-C) programme. Health Policy Plan 2008; 23: 40827. 115 Frongillo EA, Rubinstein-Gillis S, Horan C, Frongillo D, Menon P. Delivering for impact: a systematic review of strategies used for delivering critical public health interventions. Report prepared for the World Health Organization, Department of Adolescent and Child Health. Geneva, 2007. 116 Olney DK, Rawat R, Ruel MT. Identifying potential programs and platforms to deliver multiple micronutrient interventions. J Nutr 2012; 142: 178S85S. 117 Perez-Escamilla R, Curry L, Minhas D, Taylor L, Bradley E. Scaling up of breastfeeding promotion programs in low- and middle-income countries: the breastfeeding gear model. Adv Nutr 2012; 3: 790800. 118 Cooley L, Kohl R. Scaling upfrom vision to large-scale change: a management framework for practitioners. Washington, DC: Management Systems International, 2006. 119 Cash RA, Chowdhury MR, Smith GB, Ahmed RA. From one to many: scaling up health programs in low income countries. Dhaka: The University Press, 2011. 120 McCannon CJ, Berwick DM, Massoud MR. The science of large-scale change in global health. JAMA 2007; 298: 193739. 121 Mills A, Hanson K. Expanding access to health interventions in low and middle-income countries: constraints and opportunities for scaling-up. J Int Dev 2003; 15: 1131. 122 ExpandNet: scaling up health innovations. Scaling-up literature. http://www.expandnet.net/literature.htm (accessed March 30, 2013). 123 WHO. Nutrition Landscape Information System (NLiS). http:// www.who.int/nutrition/nlis/en/ (accessed March 30, 2013). 124 Gilligan D, Hoddinott J, Taffesse AS. The impact of Ethiopias productive safety net programme and its linkages. J Dev Stud 2009; 45: 1684706.

98

www.thelancet.com

Series

125 Bryce J, Victora CG. Ten methodological lessons from the multicountry evaluation of integrated Management of Childhood Illness. Health Policy Plan 2005; 20 (suppl 1): i94105. 126 Victora CG, Black RE, Boerma JT, Bryce J. Measuring impact in the Millennium Development Goal era and beyond: a new approach to large-scale effectiveness evaluations. Lancet 2011; 377: 8595. 127 Masset E, Haddad L, Cornelius A, Izara-Castro J. Effectiveness of agricultural interventions that aim to improve nutritional status of children: systematic review. BMJ 2012; 344: d8222. 128 Stoltzfus RJ. Research needed to strengthen science and programs for the control of iron deficiency and its consequences in young children. J Nutr 2008; 138: 254246. 129 White H. Theory-based impact evaluation: principles and practice. J Dev Effect 2009; 1: 27184. 130 Jennings S. Times bitter flood: trends in the numbers of reported natural disasters. Oxford, UK: Oxfam, 2011. 131 World Bank. World development report 2011: conflict, security and development. Washington, DC: World Bank 2011. 132 Hall A, Blankson B, Shoham J. The impact and effectiveness of emergency nutrition and nutrition-related interventions: a review of published evidence 20042010. Oxford, UK: Emergency Nutrition Network, 2011. 133 De Savigny D, Adam T. Systems thinking for health systems strengthening. Alliance for Health Policy and Systems Research. Geneva: World Heath Organization, 2009. 134 Dolan C, Mejia Acosta A, Shoham J. A synthesis of lessons: government experiences of scale-up of community-based management of acute malnutrition (CMAM). Oxford, UK: Emergency Nutrition Network, 2012. 135 Bossert TJ, Mitchell AD. Health sector decentralization and local decision-making: decision space, institutional capacities and accountability in Pakistan. Soc Sci Med 2011; 72: 3948. 136 Harriss J, Kohli N. Notes on the differing states of child undernutrition in rural India. IDS Bull 2009; 40: 915. 137 Fixsen DJ, Naoom SF, Blase KA., Friedman RF, Wallace F, Wallace P. Implementation research: a synthesis of the literature. Tampa, FL: Florida Mental Health Institue, National Implementation Research Network, 2005. 138 Levitt EJ, Pelletier D, Pell AN. Revisiting the UNICEF malnutrition framework to foster agriculture and health sector collaboration to reduce malnutrition: A comparison of stakeholder priorities for action in Afghanistan. Food Policy 2009; 34: 15665. 139 Ved R, Menon P. Analyzing inter-sectoral convergence to improve child undernutrition in India: development and application of a framework to examine policies in agriculture, health and nutrition. IFPRI Discussion Paper 1208. Washington, DC: IFPRI, 2012. 140 Hawkes C, Ruel MT. Value chains for nutrition. In: Fan S, Pandya-Lorch R, eds. Reshaping agriculture for nutrition and health. Washington, DC: International Food Policy Research Institute, 2012: 7381.

141 GAIN: Global Alliance for Improved Nutrition. http://www. gainhealth.org/ (accessed March 30, 2013). 142 Micronutrient Initiative. www.micronutrient.org/ (accessed March 30, 2013). 143 Sun J, Yaohua D, Shuaiming Z, et al. Implementation of a programme to market a complementary food supplement (Ying Yang Bao) and impacts on anaemia and feeding practices in Shanxi, China. Mat Child Nutr 2011; 7: 96111. 144 Suchdev PS, Ruth LJ, Woodruff BA, et al. Selling Sprinkles micronutrient powder reduces anemia, iron deficiency, and vitamin A deficiency in young children in Western Kenya: a cluster-randomized controlled trial. Am J Clin Nutr 2012; 95: 122330. 145 Porter ME, Kramer MR. Creating shared value. Harvard Bus Rev 2011 http://hbr.org/2011/01/the-big-idea-creating-shared-value (accessed May 16, 2013). 146 Ethical Trading Initiative. www.ethicaltrade.org/about-eti/ourmembers (accessed March 30, 2013). 147 STEPS Centre. Innovation, sustainability, development: a new manifesto. Brighton, UK: STEPS, 2010. 148 Hausmann R, Klinger B, Wagner R. Doing growth diagnostics in practice: a mindbook. CID Working Paper no 177. Cambridge, MA: Harvard University, 2008. 149 Potter C, Brough R. Systemic capacity building: a hierarchy of needs. Health Policy Plan 2004; 19: 33645. 150 Hampshire RD, Aguayo VM, Harouna H, Roley JA, Tarini A, Baker S. Delivery of nutrition services in health systems in subSaharan Africa: opportunities in Burkina Faso, Mozambique and Niger. Public Health Nutr 2004; 7: 104753. 151 Brown KH, McLachlan M, Cardos P, Tchibindat F, Baker S. Strengthening public health nutrition research and training capacities inWest Africa: report of a planning workshop convened in Dakar, Senegal, 2628 March 2009. Glob Public Health 2010; 5 (suppl 1): S119. 152 Khandelwal S, Dayal R, Jha M, Zodpey S, Reddy KS. Mapping of nutrition teaching and training initiatives in India: the need for Public Health Nutrition. Public Health Nutr 2012; 15: 202025. 153 SUNRAY. Sustainable nutrition research for Africa in the years to come. 2012. http://www.unscn.org/files/Announcements/ Other_announcements/SUNRAY_for_SCN__2_.pdf (accessed March 30, 2013). 154 Heikens GT, Amadi BC, Manary M, Rollins N, Tomkins A. Nutrition interventions need improved operatonal capacity. Lancet 2008; 371: 34. 155 Mubiru A. Domestic resource mobilisation across Africa: trends, challenges and policy options. Washington, DC: African Development Bank Group, 2010. 156 Spratt S. Aid for nutrition: using innovative financing to end undernutrition. Brighton, UK: Action Against Hunger, 2012. 157 UNITAID. http://www.unitaid.eu/ (accessed March 30, 2013).

www.thelancet.com

99

Articles

Mortality risk in preterm and small-for-gestational-age infants in low-income and middle-income countries: a pooled country analysis
Joanne Katz, Anne CC Lee, Naoko Kozuki, Joy E Lawn, Simon Cousens, Hannah Blencowe, Majid Ezzati, Zulfiqar A Bhutta, Tanya Marchant, Barbara A Willey, Linda Adair, Fernando Barros, Abdullah H Baqui, Parul Christian, Wafaie Fawzi, Rogelio Gonzalez, Jean Humphrey, Lieven Huybregts, Patrick Kolsteren, Aroonsri Mongkolchati, Luke C Mullany, Richard Ndyomugyenyi, Jyh Kae Nien, David Osrin, Dominique Roberfroid, Ayesha Sania, Christentze Schmiegelow, Mariangela F Silveira, James Tielsch, Anjana Vaidya, Sithembiso C Velaphi, Cesar G Victora, Deborah Watson-Jones, Robert E Black, and the CHERG Small-for-Gestational-Age-Preterm Birth Working Group

Summary
Published Online June 6, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)60993-9 Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD USA (Prof J Katz ScD, A C Lee MD, N Kozuki MSPH, Prof A H Baqui DrPH, Prof P Christian DrPH, Prof J Humphrey ScD, L C Mullany PhD, Prof J Tielsch PhD, Prof R E Black MD); Brigham and Womens Hospital, Boston, MA, USA (A C Lee); Saving Newborn Lives and Save the Children USA, Washington, DC, USA (Prof J E Lawn PhD); Maternal Reproductive and Child Health Centre, London School of Hygiene and Tropical Medicine, London, UK (Prof J E Lawn, T Marchant PhD, B A Willey PhD); Faculty of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, London, UK (Prof S Cousens DipMathStat, H Blencowe MRCPCH, B A Willey); MRC-HPA Centre for Environment and Health, Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, London, UK (Prof M Ezzati PhD); Division of Women and Child Health, Aga Khan University, Karachi, Pakistan (Prof Z A Bhutta PhD); Faculty of Infectious Disease and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, UK (Tanya Marchant); Malaria Centre, London School of Hygiene and Tropical Medicine, London, UK (T Marchant, B A Willey, D Watson-Jones PhD); University of North Carolina School of Public Health, NC, USA (Prof L Adair PhD);

Background Babies with low birthweight (<2500 g) are at increased risk of early mortality. However, low birthweight includes babies born preterm and with fetal growth restriction, and not all these infants have a birthweight less than 2500 g. We estimated the neonatal and infant mortality associated with these two characteristics in low-income and middle-income countries. Methods For this pooled analysis, we searched all available studies and identified 20 cohorts (providing data for 2015019 livebirths) from Asia, Africa, and Latin America that recorded data for birthweight, gestational age, and vital statistics through 28 days of life. Study dates ranged from 1982 through to 2010. We calculated relative risks (RR) and risk differences (RD) for mortality associated with preterm birth (<32 weeks, 32 weeks to <34 weeks, 34 weeks to <37 weeks), small-for-gestational-age (SGA; babies with birthweight in the lowest third percentile and between the third and tenth percentile of a US reference population), and preterm and SGA combinations. Findings Pooled overall RRs for preterm were 682 (95% CI 3561307) for neonatal mortality and 250 (148422) for post-neonatal mortality. Pooled RRs for babies who were SGA (with birthweight in the lowest tenth percentile of the reference population) were 183 (95% CI 134250) for neonatal mortality and 190 (132273) for post-neonatal mortality. The neonatal mortality risk of babies who were both preterm and SGA was higher than that of babies with either characteristic alone (1542; 9112612). Interpretation Many babies in low-income and middle-income countries are SGA. Preterm birth affects a smaller number of neonates than does SGA, but is associated with a higher mortality risk. The mortality risks associated with both characteristics extend beyond the neonatal period. Differentiation of the burden and risk of babies born preterm and SGA rather than with low birthweight could guide prevention and management strategies to speed progress towards Millennium Development Goal 4the reduction of child mortality. Funding Bill & Melinda Gates Foundation.

Introduction
An estimated 20 million infants every year are born with low birthweight (LBW; <2500 g),1 and these infants have an increased risk mortality in the first year of life. The pri mary causes of LBW are preterm birth, intrauterine growth restriction (IUGR), or a combination of the two. Of 135 million children born in low-income and middle-income countries (LMICs) in 2010, an estimated 297 million were born both term and small-for-gestationalage (SGA), 109 million were born preterm and appropriatefor-gestational-age, and 28 million were born preterm and SGA.2 Risk factors and interventions to reduce the number of babies born SGA might differ from those to reduce the number of babies born preterm. The survival and growth patterns of preterm or growth-restricted newborn babies are not well described in LMICs and the contribution to mortality of non-LBW babies (2500 g) who are preterm or those with IUGR in such settings is unknown.

Few studies in LMICs have investigated differences in mortality by extent of prematurity, IUGR, or the two in combination,3,4 or mortality risk in infants who are SGA stratified by gestational age.510 Examination of the mortality risk by degree of prematurity and SGA as a proxy for IUGR might be crucial in understanding the attributable disease burden, especially because regions such as south Asia have a reported SGA prevalence of about 40%.11,12 Such mortality risk estimates and attributable burden could enable the specific targeting of these disorders with appropriate interventions to more effectively save lives. The Child Health Epidemiology Reference Group (CHERG) previously examined the risk of infant mortality associated with term-LBW as a proxy for IUGR.13 However, term-LBW excludes growth-restricted infants weighing more than 2500 g and high risk infants born both preterm and SGA, and such
www.thelancet.com

100

Articles

associations between mortality and SGA-non-LBW or SGA-preterm have not been well described in LMICs. With more population-based studies in LMICs now collecting data for gestational age in addition to birth weight, the CHERG identified an opportunity to assess the mortality risk of SGA and preterm on early neonatal, late neonatal, neonatal, post-neonatal, and infant mortality.

Methods

Dataset identification
We searched Medline, WHO regional databases (African Index Medicus, LILAS, EMRO), bibliographies of sentinel articles and reviews, and grey literature to identify potential datasets from low-income and middleincome countries that recorded data for gestational age and birthweight, and systematically recorded vital status from delivery through at least 28 days of life. The most recent search was done on Feb 22, 2010. We applied no no date or language restrictions. Search terms included preterm or SGA, neonatal or infant mortality, and developing country (see appen dix for detailed search terms). CHERG investi gators also identified additional datasets that were not retrieved in our search. We excluded datasets for the following reasons: greater than 25% missing data for birthweight or gestational age, or loss to follow-up; measured weight only after the first week of life; did not have systematic follow-up of vital status in the first month of life; determined gestational age in months or by fundal height; or when we deemed gestational age deter mination inaccurate or poorly linked to birthweight (appendix). We aimed to include only datasets that were population-based, representing all deliveries arising from specific geographical or catchment areas, whether home-based or facility-based. We excluded stillbirths from the analyses of these cohorts. We approached principal investigators to do a set of standard analyses themselves, or to share their data with the CHERG working group to do the analyses. Datasets that were shared with the working group did not contain personal identifiers and were therefore deemed exempt by the Johns Hopkins Bloomberg School of Public Health institutional review board. Datasets analysed by the original study investigators were covered under existing institutional review board approvals.

(moderate preterm), and 34 weeks to less than 37 weeks (late preterm).14 Weight was analysed if measured within 72 h. SGA was defined as birthweight below the tenth percentile of a standard optimal reference population for a given gestational age and sex. Appropriate-for-gestational-age (AGA) was defined as above the tenth percentile. WHO previously recommended data in a study by Williams and colleagues (collected in California from 1972 to 1976) as the reference.15 We chose the more recent and commonly cited Alexander reference,16 which includes 3 134 879 nationally representative, multi-ethnic births in the USA in 1991. Because that dataset provided data at only the tenth percentile, we identified another dataset (Oken and colleagues17) based on US data from 19992000 690 717 births) that provided Z scores. We created two (6 categories of SGA using a combination of these two reference populations:16,17 the tenth percentile of Alexander16 to the third percentile of Oken and colleagues,17 and the lowest third percentile of Oken and colleagues. AGA included large-for-gestational-age (LGA) infants (90% birth weight for gestational age). However, in these datasets, the prevalence of LGA was very small and their inclusion in the reference population likely had very little effect on mortality risk. We created four mutually exclusive exposures to capture interaction between preterm (<37 weeks) and SGA (<10%): term and appropriate-for-gestational-age (as reference), term-SGA, preterm and appropriate-for-gestational-age, and preterm and SGA. We defined mortality as early neonatal (birth to 7 days), late neonatal (828 days), neonatal (birth to 28 days), post-neonatal (29365 days), and infant (birth to 365 days) mortality. For late neonatal and post-neonatal infant mortality, the denominators were infants alive at the start of the interval of interest with vital status available through the end of the interval.

Analysis of individual datasets


The same analysis was done on each dataset. We used an algorithm developed by Alexander and colleagues16 to exclude infants with incompatible birthweight-gesta tional age combinations. Gestational ages of less than 24 weeks and more than 48 weeks were also excluded. We calculated the prevalence of LBW, preterm, SGA, and the overlap of these disorders for each dataset. We calculated relative risks (RRs) and risk differences (RDs) for early, late, neonatal, post-neonatal, and infant mortality asso ciated with preterm and SGA categories. RRs were adjusted for all confounders available in each dataset (including occupation of head of house hold, land ownership, years of maternal and paternal education or literacy, and maternal age and parity). We did adjusted analyses on 13 of the 20 datasets. Regression coefficients for the primary associations of interest were attenuated at less than 10% in all datasets except for one,32 for which the parameters were not statistically significant.

Exposure definitions
If more than one gestational age measurement was available we used estimates in the following order of preference: ultrasound, best obstetric estimate (com bin ation of ultrasound, date of last menstrual period, and neonatal clinical exam), last menstrual period, or clinical examination of the infant within 72 h of birth. Prematurity was defined as a gestational age of less than 37 completed weeks. We categorised prematurity into less than 32 weeks (early preterm), 32 weeks to less than 34 weeks
www.thelancet.com

Programa de Ps-graduacao em Epidemiologia, Universidade Federal de Pelotas, Pelotas, RS, Brazil (Prof F Barros PhD, M F Silveira PhD, Prof C G Victora PhD); Programa de Ps-graduao em Sade e Comportamento, Univertsidade Catlica de Pelotas, Centro, Pelotas, RS, Brazil (Prof F Barros); Department of Nutrition, Harvard School of Public Health, Boston, MA, USA (Prof W Fawzi DrPH); Department of Epidemiology, Harvard School of Public Health, Boston, MA, USA (Prof W Fawzi, A Sania PhD); Department of Global Health and Population, Harvard School of Public Health, Boston, MA, USA (Prof W Fawzi); Pontificia Universidad Catlica de Chile, School of Medicine, Santiago, Chile (Prof R Gonzalez MD); Clnica Santa Mara, Santiago, Chile (Prof R Gonzalez MD); Zvitambo, Borrowdale, Harare, Zimbabwe (Prof J Humphrey); Department of Food Safety and Food Quality, Ghent University, Ghent, Belgium (L Huybregts PhD, Prof P Kolsteren PhD); Woman and Child Health Research Center, Department of Public Health, Institute of Tropical Medicine, Antwerpen, Belgium (L Huybregts, Prof P Kolsteren, D Roberfroid PhD); ASEAN Institute for Health Development, Mahidol University, Nakhon Pathom, Thailand (A Mongkolchati PhD); Vector Control Division, Ministry of Health, Kampala Uganda (R Ndyomugyenyi PhD); Fetal Maternal Medicine Unit, Clinica Davila, Santiago, Chile (J K Nien MD); Faculty of Medicine, Universidad de Los Andes, Santiago, Chile (J K Nien); Institute for Global Health, UCL Institute of Child Health, London, UK (D Osrin PhD, A Vaidya PhD); Department of Global Health, George Washington School of Public Health and Health Services, George Washington University, Washington, DC, USA (Prof J Tielsch); Centre for Medical Parasitology, Institute of International Health, Immunology, and Microbiology, University of Copenhagen and Department of Infectious Diseases, Copenhagen University Hospital, Copenhagen, Denmark (C Schmiegelow PhD); Department of Paediatrics,

101

Articles

Division of Neonatology, Chris Hani Baragwaneth Hospital, University of Witwatersrand, Soweto, South Africa (S C Velaphi Mmed); Mwanza Intervention Trial Unit, National Institutes of Medical Research, Mwanza, Tanzania (D Watson-Jones) Correspondence to: Prof Joanne Katz, Department of International Health, Johns Hopkins School of Public Health, 615 N Wolfe Street, W5009, Baltimore, MD 21205, USA jkatz@jhsph.edu See Online for appendix

Missing exposure data


All studies had few missing gestational age data (<10%). Analyses of mortality risk by preterm categories used all available gestational age data, irrespective of missing birthweight (full gestational age cohort). There were two main reasons for missing birthweight data (most of which were from home deliveries): some infants who died soon after delivery were not weighed, and infants who were weighed were measured at different times after delivery. For risk estimates associated with SGA, we included infants with weights taken within 72 h of birth (weighed cohort). For four Asian and two African studies with missing data for birthweight or that measured some weights after 72 h, we used multiple imputation18 to impute birthweights (appendix). The imputed datasets were used to estimate the RRs and RDs.

all statistical analyses. Random effects models used the DerSimonian-Laird pooled RRs and 95% CIs, given between-study heterogeneity.19

Role of the funding source


The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to most of the datasets, had access to all summary estimates from each dataset for metaanalysis, and had final responsibility for the decision to submit for publication.

Results
We included 20 datasets with 2 015 019 livebirths from sub-Saharan Africa, Latin America, and south and southeast Asia, with gestational age available for 2 008 675 babies and both gestational age and birth weight available for 1 996 763 babies (table and appendix).4,2040 Study dates ranged from 1982 through to 2010. The Chilean national birth registry35 provided much of these data. The prevalence of preterm birth (<37 weeks) in the study datasets ranged from 27% to 28%, with varying methods of gestational age determination (see appendix for specific study
IMR (deaths per 1000 livebirths) % % Systematic % follow-up LBW preterm SGA period % facility delivery

Pooled analysis
Because multivariate adjustment did not substantially modify estimates of associations, we used the crude RRs and 95% CIs, and pooled data for the major UN Millennium Development Goal regions (Asia, Africa, and Latin America). We estimated overall and regional RRs of mortality for each exposure variable (preterm, SGA, and combinations). We used Stata (version 11) for
Setting Primary study design Study population N (original cohort)

N (analysed cohort NMR (deaths for preterm/for per 1000 SGA) livebirths) 10585/10550* 31

Asia Bangladesh (2005)21 Rural Sylhet Cluster RCT of community sepsis treatment RCT of newborn vitamin A supplementation Population-based recruitment of all pregnant women in study area Population-based recruitment of all pregnant women in study area 10585 1 month 30% 19% 50% 6%

India (2000)26

Rural Tamil Nadu

13294

12693/12693*

38

6 months

33%

14%

62%

63%

Nepal (1999)22 Nepal (2003)25

Rural Sarlahi

Cluster RCT of multiple Recruitment of all pregnant women in micronutrient study area supplementation Antenatal clinicbased recruitment of pregnant women in study area Population-based recruitment of all pregnant women in study area Population-based recruitment of all pregnant women in study villages

4130

4122/4094*

42

67

1 year

39%

22%

56%

6%

Peri-urban/rural RCT of antenatal Dhanusha micronutrient supplementation Rural Sarlahi Cluster RCT of newborn skin-umbilical cord cleansing with chlorhexidine Cluster RCT of maternal micronutrient supplementation

1106

1106/1052

25

1 month

22%

7%

53%

53%

Nepal (2004)29

23662

23650/22723*

32

6 months

30%

18%

52%

10%

Pakistan (2003)39

Rural Sindh

1548

1464/1434

18

1 month

19%

28%

28%

100%

Philippines (1983)20 Thailand (2001)24

Urban Cebu

Population-based, Longitudinal healthrandom cluster nutritional survey of infant feeding patterns sample of census Prospective follow-up of birth cohort Longitudinal birth cohort of all births in four districts

3080

3050/2785

14

36

2 years

11%

18%

25%

34%

Urban Bangkok

4245

4032/3860

1 year

8%

9%

22%

99%

(Continues on next page)

102

www.thelancet.com

Articles

Setting

Primary study design

Study population

N (original cohort)

N (analysed cohort NMR (deaths for preterm/for per 1000 SGA) livebirths)

% % Systematic % IMR follow-up LBW preterm SGA (deaths per 1000 period livebirths)

% facility delivery

(Continued from previous page) Sub-Saharan Africa Burkina Faso Rural Hounde (2004)27 Burkina Faso Rural Hounde (2006)36 South Africa (2004)34 Tanzania (1998)32 Tanzania (2001)23 Tanzania (2008)38 Urban Soweto RCT of multiple micronutrient supplementation RCT of maternal fortified food supplementation Prospective, community-based cohort Prospective, community-based cohort 1373 1311/1212* 21 67 1 year 17% 16% 35% 77%

1316

1261/1067

20

1 month

16%

18%

29%

84%

RCT of birth canal and Facility-based newborn skin cleansing delivery, tertiary-care hospital with chlorhexidine Maternal syphilis treatment, observational cohort RCT of mutivitamin supplementation Observational malaria study Facility-based recruitment; urban, antenatal clinics Facility-based, antenatal clinics Facility-based recruitment, antenatal clinics, community follow-up. Facility-based recruitment ANC clinics; only include facility births Facility-based recruitment, 14 maternity clinics and hospitals Population-based, all births in Pelotas hospitals (100% facility delivery) Population-based, all births in Pelotas hospitals (100% facility delivery) Population-based, all births in Pelotas hospitals (100% facility delivery) Population-based registry

8113

8113/8098

1 month

8%

4%

16%

100%

Rural Mwanza

1496

1425/1172

16

3 months

10%

3%

25%

98%

Urban Dar es Salaam Rural Korogwe

7752 915

7740/7557 820/731

28 33

6 weeks 28 days

8% 11%

17% 5%

20% 22%

97% 88%

Uganda (2005)37

Rural Kabale district

RCT intermittent preventive malaria therapy and insecticide-treated nets RCT of maternalneonatal vitamin A supplementation

1561

1553/1477

17

1 month

7%

6%

10%

100%

Zimbabwe (1997)33,40

Urban Harare

14110

13960/13914

12

93

1 year

14%

8%

33%

100%

Latin America Brazil (1982)30 Urban Pelotas city, Rio Grande do Sul, southern Brazil Urban Pelotas city, Rio Grande do Sul, southern Brazil Urban Pelotas city, Rio Grande do Sul, southern Brazil Chilean national birth registry Longitudinal birth cohort survey 5914 4675/4670 11 28 1 year 7% 6% 17% 100%

Brazil (1993)31

Longitudinal birth cohort survey

5279

4707/4632

14

1 year

9%

11%

19%

100%

Brazil (2004)28

Longitudinal birth cohort survey

4287

3903/3837

10

17

1 year

11%

16%

15%

100%

Chile (2000)35

Birth registry

1901611

1898250/ 1898250

1 month

6%

7%

7%

98%

RCT=randomised controlled trial. SGA=small for gestational age. NMR=neonatal mortality rate. IMR=infant mortality rate. LBW=low birthweight. *Weights imputed for datasets that met criteria described in appendix. Enrolment of newborn babies occurred up to 96 h after birth, and the study might have missed neonatal deaths that happened before enrolment.

Table: Study characteristics of 20 included datasets

methods). The proportions of preterm births before 32 weeks of gestation ranged from 0% to 4% (appendix). Comparison of preterm prevalence by method of assessment was not possible because few studies used more than one method. The prevalence of SGA was generally higher than that of preterm births, ranging from 7% in the Chile national registry to 62% in a community-based trial in south India (appendix). The proportion of LBW infants who were SGA or preterm varied by region (figure 1). In the Asian cohorts, 83% of
www.thelancet.com

LBW infants were SGA and 33% were preterm (67% of LBW were term and SGA). In the African cohorts, 79% of LBW infants were SGA and 38% were preterm (62% of LBW were term and SGA). In the Latin American cohorts, 53% of LBW infants were SGA and 71% were preterm (29% of LBW were term and SGA). South Asia had the highest prevalence of preterm births and number of LBW and SGA infants. A substantial proportion of SGA infants did not have LBW in Asia (54%), Africa (65%), and Latin America (59%). The
103

Articles

proportion of children who were both SGA and preterm was small in these datasets (4% in Asia, 1% in Africa, and 2% in Latin America). In Africa and Asia, term
A

2500 g

N=5711 (7%)

N=38 297 (49%)


al age ation r gest fo e t ia pr al age Appro ation r gest fo ll a Sm

2500 g

N=16 656 (21%) N=3020 (4%)

Low N=2748 (4%) N=11 616 (15%)

2500 g

N=1949 (5%)

N=26 567 (68%)

e nal ag statio for ge e e t g ia a r p nal Appro statio for ge Small

2500 g Low

N=6230 (16%) N=885 (2%)

N=707 (2%)

N=2610 (7%)

2500 g Birthweight

N=52 671 (3%)

N=1 663 506 (87%)

2500 g Low

N=83 297 (4%) N=52 819 (3%) N=26 397 (2%) Preterm 37 weeks

e gestational ag Appropriate for e ag l na tio Small for gesta

N=32 044 (2%) Full-term Gestational age

Figure 1: The relation between birthweight and gestational age in Asia (A), Africa (B), and Latin America (C) For Asian cohorts (A), percentages are for 78048 infants. For African cohorts (B), percentages are for 38948 infants. For Latin American cohorts (C), percentages are for 1910734 infants.

infants were more often SGA than were preterm infants, but the opposite was true in Latin America (figure 1). Neonatal mortality rates and relative risks increased with decreasing gestational age across studies and regions (figure 2). Although the highest relative risks (RRs) were seen in Latin America (mainly due to the very low mortality in the reference group), the risk differences were similar across regions. The overall pooled RRs across all regions were 6.82 (3561307) for neonatal mortality and 250 (148422) for postneonatal mortality. The overall pooled RRs across all regions for late preterm (34 weeks to <37 weeks), when most preterm births occur, were 305 (202460; figure 2 and appendix). The RDs per 1000 livebirths for late preterm ranged from nine (95% CI 018) in Africa to 18 (928) in Asia, with an overall RD per 1000 livebirths of 13 (918; appendix). The overall pooled RRs for early preterm (<32 weeks) were 2882 (15515356; figure 2 and appendix), although the RDs per 1000 livebirths ranged from 196 (95297) in Asia to 350 (189511) in Latin America, with an overall RD of 245 per 1000 livebirths (192297; appendix). Early and late neonatal and post-neonatal infant mortality followed similar patterns, but the RRs were more attenuated the longer the follow-up period extended (appendix), except for early preterm in Latin America, when mortality risk decreased substantially across the early-post-neonatal to late-post-neonatal periods. How ever, a statistically significantly increased mortality risk persisted into the post-neonatal infant period in all regions for preterm compared to term. Imputation of missing birthweights did not alter the prevalence of SGA, which was already high, although it did increase the risk of mortality associated with SGA (figure 3 and appendix). Neonatal mortality rates increased with severity of SGA (figure 3). The overall pooled RRs for SGA across all regions were 183 (134250) for neonatal mortality and 190 (132273) for post-neonatal mortality. The highest RR for SGA in neonates below the third percentile was in Latin America, compared with 191 (140260) in Asia, although the CIs overlapped. The overall RR was 241 (166350). Regional pooled RDs per 1000 livebirths for SGA below the third percentile ranged from 12 (1015) in Asia to 23 (1729) in Latin America, with an overall RD per 1000 livebirths of 14 (919; appendix). The RRs associated with SGA were of similar magnitude in the early-neonatal, late-neonatal, and post-neonatal periods across regions (appendix). The RRs associated with SGA were overall of smaller magnitude than preterm, and the association with increased mortality did not attenuate beyond the neonatal period and persisted through the first year of life. Relative to term and AGA, the increased risk of neo natal mortality was lowest in those born term-SGA (overall RR 244, 167357) and highest in preterm-SGA infants (1542, 9112612). RRs were similar for Asia and Africa, but higher in Latin America (figure 4). These
www.thelancet.com

104

Birthweight

Birthweight

Articles

higher RRs in Latin America were probably driven by the low mortality rate (24 per 1000) in the reference group. Pooled RDs were similar across regions (appendix). The RRs for term-SGA babies did not vary by timing of mortality (figure 5). For preterm-AGA and preterm-SGA babies, RRs were highest in the early neonatal period and lowest in the post-neonatal period, although the CIs largely overlapped (ie, no statistically significant betweengroup differences; figure 5 and appendix). This pattern of attenuated risk by length of follow-up was probably driven by prematurity and not SGA.

1000

100 1898 10 535 264 1 Asia Africa Latin America Asia Africa Latin America Asia Africa Latin America Asia 274 544 1102

10010 2579

Relative risk

1660

Africa <32 weeks 223 Africa 209

Discussion
We identified a large percentage of infants who were SGA but not LBW or preterm (21% in Asia, 16% in Africa, and 4% in Latin America), although their mortality rates were lower than preterm infants or SGALBW infants. Although most LBW was in term and SGA babies in Asia and Africa, the majority of babies with LBW in Latin America were preterm. Preterm mortality risk associations were generally higher at all gestational age categories (late, moderate, and early preterm) than SGA (3% to <10% or <3%). However, the attributable mortality risk for SGA infants is substantial, because many infants in LMICs are born SGA, especially in south Asia. The predominant increased mortality risk asso ciated with preterm birth occurred in the first week of life, with statistically significant but attenuated risk remaining in the late-neonatal and post-neonatal periods. Although CIs overlapped, mortality risks asso ciated with SGA were slightly higher in the late than early neonatal period, with persistent risk in the post-neonatal period. The highest neonatal and infant mortality rates were detected in the Asian and African studies. The highest RRs for all exposures were seen in Latin America, although absolute risk differences were more comparable across regions. This higher RR in Latin America is due to very low mortality in the reference group in Latin America compared with Africa and Asia. In particular, the Latin America analysis was dominated by national registry data from Chile in which term neonatal mortality was 17 per 1000 livebirths. By comparison with these data, the average term neonatal mortality was 11 per 1000 live births in the African datasets and 19 per 1000 livebirths in the Asian ones. Alternately, preterm infants might be more severely preterm and survive delivery in Latin America but have delayed mortality. Preterm birth was associated with an increased mortality risk compared with term birth. The proportion of infants born within 32 weeks of gestation was low but these infants had substantially higher mortality risks than did term infants in the first week of life. Late preterm birth comprised 5096% of preterm births, and was associated with a smaller but statistically significant neonatal mor tality risk ranging from 252 in Asia to 558 in Latin America (risk differences per thousand livebirths were nine in Africa, 11 in Latin America, and 18 in Asia).
www.thelancet.com

37 weeks (reference) Median neonatal mortality rate (per 1000 livebirths) 187 92 35

34 to <37 weeks 3233 weeks Gestational age 406 223 142 775 639 632

2748 3065 2375

Figure 2: Relative risk of neonatal mortality associated with gestational age Error bars are 95% CI.
10

Relative risk

193 120 1 Asia Latin America Asia Africa Latin America Asia Africa 127

191

Appropriate for gestational age (reference) Median neonatal mortality rate


(per 1000 livebirths)

120

216

884

Small for gestational age (310%) Size for gestational age 151 180 126

Small for gestational age (<3%) 269 276

Figure 3: Relative risk of neonatal mortality associated with small for gestational age

Evidence-based, low-cost interventions are feasible for LMICs and could reduce mortality related to preterm birth complications, such as antenatal cortico steroids for preterm labour, Kangaroo mother care, and treatment of neonatal infections.4143 Our findings suggest that simple interventions to target the improved care of late and moderate preterm infants could have a large effect on the reduction of mortality burden in preterm infants. In areas with a large proportion of facility deliveries and much post-delivery care, clinical inter ventions such as surfactant administration and continuous positive airway pressure might also improve survival.
105

Latin America

Latin America 401

Articles

100 3948 2577 Relative risk 1673 10 534 343 201 1 Latin America Asia Africa Latin America Asia Africa Latin America Asia Africa Latin America Asia Africa 696 621 1005

Term and appropriate for gestational age Median neonatal mortality rate
(per 1000 livebirths)

79

84

24

Term and Preterm and small for appropriate for gestational age gestational age Characteristics at birth 176 119 101 235 642 530

Preterm and small for gestational age 1065 702 755

Figure 4: Relative risk of neonatal mortality associated with preterm and size for gestational age
100

Relative risk

1597 10 906 574 312 309 285 22

1865

577

1 Early-neonatal Early-neonatal Early-neonatal Late-neonatal Late-neonatal Post-neonatal Post-neonatal Late-neonatal Post-neonatal

Term and small for gestational age

Preterm and appropriate for gestational age Characteristics at birth

Preterm and small for gestational age

Figure 5: Relative risk of early-neonatal, late-neonatal, and post-neonatal infant mortality associated with preterm and size for gestational age

For The Intergrowth Study see http://www.intergrowth21.org.uk

Using a common US birthweight reference population, the prevalence of SGA and term-SGA babies was very high (higher than 50%) in many of the community-based South Asian cohorts.11,12 In statistical modelling to esti mate national and regional prevalences of SGA, we estimated that 42% of babies were term-SGA and that 3% of babies were preterm-SGA.2 These findings contrast with a much lower prevalence of SGA in Africa, despite the high prevalence of risk factors such as malaria and HIV infection in pregnancy. SGA neonates had roughly double the mortality risk of appropriate-for-gestationalage infants, with slightly higher risk in those with more severe SGA. A large proportion of SGA infants weighed

2500 g or more, although their mortality risk was similar to that of term-AGA babies. The very high rates of SGA in South Asia might be explained by higher rates of adolescent preg nancy, chronic maternal malnutrition, low pre-pregnancy body-mass index, low weight gain in pregnancy, and low maternal height.4446 The mortality risk associated with being preterm-SGA was substantially higher than for either alone. An estimated 28 million infants are born both preterm and SGA in LMICs annually2 and these infants are at a 1040 times increased risk of dying in the first month of life compared with term and appropriate-for-gestationalage infants. These children are key targets for public health interventions. A strength of this analysis was the large number of representative livebirths analysed from a wide range of studies and geographical regions, producing internal validity and reduced variance of estimates. We adjusted for several covariates but noted that they did not alter the risk estimates. Although the cohort sizes varied sub stantially, with the Chilean national registry data domin ating in absolute numbers, the use of random effects models to do meta-analyses downweights the effect of such large datasets. Although it constitutes 92% of the tality risk data, its contribution to the Latin American mor estimate is about 50% and its contribution to the global estimate is 97%. While Chile is proably a good representation of a middle-income country in Latin America, it does not represent other low-income coun tries in the region. The earliest Brazilian cohort30 might be more representative of such countries because it was done in a low-income setting. We re-analysed individual data to estimate SGA preva lence and risk, applying a common SGA reference popu lation. This reduced the variation associated with the use of different SGA reference populations commonly seen across studies. Although the use of a common reference population is a controversial issue, with argu ments made about whether it is appropriate to apply one standard to all populations, we selected one reference population to be able to better compare our analyses across populations. Our use of a birthweight rather than fetal growth standard might have system atically under-represen ted SGA in pre term infants because preterm infants might have more pathological IUGR than fetuses that remain in the womb for longer. The Intergrowth Study is collecting data for a common fetal growth reference using healthy populations from many countries. The use of this standard will help resolve this limitation in the future. The large sample size allowed us to stratify preterm and SGA into severity categories and to pool mortality risk associated with each category by re-analysing primary data. Available covariates varied by study and residual con founding could have occurred. However, these findings might not be generalisable to countries or regions as a whole because they are derived from cohorts or randomised trials in which the geographical
www.thelancet.com

106

Articles

areas might have been selected for specific risk characteristics as evidenced by the range of preterm prevalence across studies (table). However, the use of meta-analyses with random effects of 20 cohorts probably reduced the bias and increased the precision of regional and global risk estimates. Although the prevalence of preterm and SGA births seemed quite variable across these datasets, the mortality differences associated with these characteristics were stable within regions. Hence these estimates are likely to be valid for use in attributable risk calculations if representative estimates of preterm and SGA prevalence are used and residual confounding was minimal. A limitation of our community-based datasets was the exclusion of early deaths in infants who were not weighed. We used multiple imputation to address the potential effect on mortality associations. Nevertheless, we might have underestimated the mortality risks associated with SGA given that we were missing some covariate data with which to impute birthweight. Another limitation is the variability and accuracy of gestational age measurement between datasets. Nine studies included ultrasound dating, whereas the remainder relied on maternal recall of their last menstrual period or clinical exam. We used the best gestational age deter mination, excluding datapoints with improbable gesta ations and datasets tional age and birthweight combin with poor gestational age measures. Gestational age categories might have been subject to some mis mised classification although these were probably mini by the active, frequent pregnancy surveillance in many studies. Stillbirths were excluded from these analyses, but it is possible that there was misclassification of stillbirths and livebirths, especially in home deliveries that relied on maternal report, but the direction of this misclassification is unclear. Although assessment of cause of death by adverse pregnancy outcome is of interest, very few studies had cause of death, and most population-based studies that did, had cause assigned by physician via verbal autopsy. Assessing causes of death associated with SGA or preterm in these contexts would be a valuable future contribution to the design of appro priate mortality reduction interventions. Many babies are born SGA in LMICs, especially in south Asia and sub-Saharan Africa, and such babies have increased risk of mortality and poor growth.47 Preterm birth affects a smaller number of neonates but is associated with a higher mortality risk and seems to be a greater contributor to attributable risk in Latin America than in Asia. In both these regions, these risks extend beyond the neonatal period. Although few effective interventions to prevent preterm exist,48 several inter ventions exist to improve preterm survival. Inter ventions shown to reduce SGA have focused mainly on increasing calories and protein during pregnancy, and more recently (during the past decade) maternal micronutrient supplementation.48 As measurement of gesta tional age
www.thelancet.com

improves in LMICs, targeting inter ventions and tracking outcomes that reduce the incidence and improve the survival of babies born preterm and SGA rather than with a LBW might more clearly guide progress towards the reduction of child mortality.
Contributors JK, ACL, and NK did the literature review, collected the data, analysed the data, and drafted the paper. JEL, SC, HB, ME, ZAB, and REB helped draft the analysis plan and reviewed the paper. JKN, AS, MFS, and AV did analyses of their individual studies, contributed data, and reviewed the paper. TM, BAW, LA, FB, AHB, PC, WF, RG, JH, LH, PK, AM, LCM, RN, DO, DR, CS, JT, SCV, CGV, and DW-J contributed data and reviewed the paper. Conflicts of interest We declare that we have no conflict of interest. Acknowledgments Funding was provided by the Bill & Melinda Gates Foundation (8102054) by a grant to the US Fund for UNICEF to support the activities of the Child Health Epidemiology Reference Group. Financial support for analysis was offered to investigators through a subcontract mechanism administered by the US Fund for UNICEF. For funding information for the individuals studies please see appendix. We thank the additional members of the CHERG SGA-Preterm Birth working group: Sin Clarke, Simon Kariuki, John Lusingu, James Ndirangu, Marie-Louise Newell, Robert Ntozini, Heather Rosen, and Feiko O Ter Kuile. The individual studies would like to acknowledge the role played by following people in those studies: Ramesh Adhikari, Christian Coles, Anthony Costello, Gary Darmstadt, Sheela Devi, Subarna Khatry, Hermann Lanou, Steven LeClerq, Dharma Manandhar, Daniel Minja, Gernard I Msamanga, R D Thulasiraj, Willy Urassa, Helen Weiss, and Keith West. We would like to thank Joanna Schellenberg for her support of this work. References 1 UNICEF. WHO. Low birthweight: country, regional and global estimates. New York 2004. 2 Lee ACC, Katz J, Blencowe H, et al. Born too small: national and regional estimates of term and preterm small-for-gestational-age in 138 low-income and middle-income countries in 2010. Lancet Glob Health (in press). 3 Kramer MS, Demissie K, Yang H, Platt RW, Sauv R, Liston R, and the Fetal and Infant Health Study Group of the Canadian Perinatal Surveillance System. The contribution of mild and moderate preterm birth to infant mortality. JAMA 2000; 284: 84349. 4 Marchant T, Willey B, Katz J, et al. Neonatal mortality risk associated with preterm birth in East Africa, adjusted by weight for gestational age: individual participant level meta-analysis. PLoS Med 2012; 9: e1001292. 5 Narchi H, Skinner A, Williams B. Small for gestational age neonatesare we missing some by only using standard population growth standards and does it matter? J Matern Fetal Neonatal Med 2010; 23: 4854. 6 Clausson B, Cnattingius S, Axelsson O. Preterm and term births of small for gestational age infants: a population-based study of risk factors among nulliparous women. Br J Obstet Gynaecol 1998; 105: 101117. 7 Kristensen S, Salihu HM, Keith LG, Kirby RS, Fowler KB, Pass MA. SGA subtypes and mortality risk among singleton births. Early Hum Dev 2007; 83: 99105. 8 Pulver LS, Guest-Warnick G, Stoddard GJ, Byington CL, Young PC. Weight for gestational age affects the mortality of late preterm infants. Pediatrics 2009; 123: e107277. 9 Regev RH, Reichman B. Prematurity and intrauterine growth retardationdouble jeopardy? Clin Perinatol 2004; 31: 45373. 10 Grisaru-Granovsky S, Reichman B, Lerner-Geva L, et al. Mortality and morbidity in preterm small-for-gestational-age infants: a population-based study. Am J Obstet Gynecol 2012; 206: 150 e17. 11 Wu LA, Katz J, Mullany LC, et al. The association of preterm birth and small birthweight for gestational age on childhood disability screening using the Ten Questions Plus tool in rural Sarlahi district, southern Nepal. Child Care Health Dev 2012; 38: 33240.

107

Articles

12 Osendarp SJ, van Raaij JM, Arifeen SE, Wahed M, Baqui AH, Fuchs GJ. A randomized, placebo-controlled trial of the effect of zinc supplementation during pregnancy on pregnancy outcome in Bangladeshi urban poor. Am J Clin Nutr 2000; 71: 11419. 13 Black RE, Allen LH, Bhutta ZA, et al, and the Maternal and Child Undernutrition Study Group. Maternal and child undernutrition: global and regional exposures and health consequences. Lancet 2008; 371: 24360. 14 Barfield WD, Lee KG. Late preterm infants. 2012; http://www. uptodate.com/contents/late-preterm-infants (accessed July 5, 2012). 15 Williams RL, Creasy RK, Cunningham GC, Hawes WE, Norris FD, Tashiro M. Fetal growth and perinatal viability in California. Obstet Gynecol 1982; 59: 62432. 16 Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national reference for fetal growth. Obstet Gynecol 1996; 87: 16368. 17 Oken E, Kleinman KP, Rich-Edwards J, Gillman MW. A nearly continuous measure of birth weight for gestational age using a United States national reference. BMC Pediatr 2003; 3: 6. 18 Rubin DB. Multiple imputation after 18+ years. J Am Stat Assoc 1996; 91: 47389. 19 DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 17788. 20 Adair LS. Low birth weight and intrauterine growth retardation in Filipino infants. Pediatrics 1989; 84: 61322. 21 Baqui AH, El-Arifeen S, Darmstadt GL, et al, and the Projahnmo Study Group. Effect of community-based newborn-care intervention package implemented through two service-delivery strategies in Sylhet district, Bangladesh: a cluster-randomised controlled trial. Lancet 2008; 371: 193644. 22 Christian P, West KP, Khatry SK, et al. Effects of maternal micronutrient supplementation on fetal loss and infant mortality: a cluster-randomized trial in Nepal. Am J Clin Nutr 2003; 78: 1194202. 23 Fawzi WW, Msamanga GI, Urassa W, et al. Vitamins and perinatal outcomes among HIV-negative women in Tanzania. N Engl J Med 2007; 356: 142331. 24 Isaranurug S, Mo-suwan L, Choprapawon C. A population-based cohort study of effect of maternal risk factors on low birthweight in Thailand. J Med Assoc Thai 2007; 90: 255964. 25 Osrin D, Vaidya A, Shrestha Y, et al. Effects of antenatal multiple micronutrient supplementation on birthweight and gestational duration in Nepal: double-blind, randomised controlled trial. Lancet 2005; 365: 95562. 26 Rahmathullah L, Tielsch JM, Thulasiraj RD, et al. Impact of supplementing newborn infants with vitamin A on early infant mortality: community based randomised trial in southern India. BMJ 2003; 327: 254. 27 Roberfroid D, Huybregts L, Lanou H, et al, and the MISAME Study Group. Effects of maternal multiple micronutrient supplementation on fetal growth: a double-blind randomized controlled trial in rural Burkina Faso. Am J Clin Nutr 2008; 88: 133040. 28 Santos IS, Barros AJ, Matijasevich A, Domingues MR, Barros FC, Victora CG. Cohort profile: the 2004 Pelotas (Brazil) birth cohort study. Int J Epidemiol 2011; 40: 146168. 29 Tielsch JM, Darmstadt GL, Mullany LC, et al. Impact of newborn skin-cleansing with chlorhexidine on neonatal mortality in southern Nepal: a community-based, cluster-randomized trial. Pediatrics 2007; 119: e33040. 30 Victora CG, Barros FC. Cohort profile: the 1982 Pelotas (Brazil) birth cohort study. Int J Epidemiol 2006; 35: 23742. 31 Victora CG, Hallal PC, Arajo CL, Menezes AM, Wells JC, Barros FC. Cohort profile: the 1993 Pelotas (Brazil) birth cohort study. Int J Epidemiol 2008; 37: 70409.

32 Watson-Jones D, Changalucha J, Gumodoka B, et al. Syphilis in pregnancy in Tanzania. I. Impact of maternal syphilis on outcome of pregnancy. J Infect Dis 2002; 186: 94047. 33 Malaba LC, Iliff PJ, Nathoo KJ, et al, and the ZVITAMBO Study Group. Effect of postpartum maternal or neonatal vitamin A supplementation on infant mortality among infants born to HIV-negative mothers in Zimbabwe. Am J Clin Nutr 2005; 81: 45460. 34 Cutland CL, Madhi SA, Zell ER, et al, and the PoPS Trial Team. Chlorhexidine maternal-vaginal and neonate body wipes in sepsis and vertical transmission of pathogenic bacteria in South Africa: a randomised, controlled trial. Lancet 2009; 374: 190916. 35 Gonzalez R, Merialdi M, Lincetto O, et al. Reduction in neonatal mortality in Chile between 1990 and 2000. Pediatrics 2006; 117: e94954. 36 Huybregts L, Roberfroid D, Lanou H, et al. Prenatal food supplementation fortified with multiple micronutrients increases birth length: a randomized controlled trial in rural Burkina Faso. Am J Clin Nutr 2009; 90: 1593600. 37 Ndyomugyenyi R, Clarke SE, Hutchison CL, Hansen KS, Magnussen P. Efficacy of malaria prevention during pregnancy in an area of low and unstable transmission: an individually-randomised placebo-controlled trial using intermittent preventive treatment and insecticide-treated nets in the Kabale Highlands, southwestern Uganda. Trans R Soc Trop Med Hyg 2011; 105: 60716. 38 Schmiegelow C, Minja D, Oesterholt M, et al. Factors associated with and causes of perinatal mortality in northeastern Tanzania. Acta Obstet Gynecol Scand 2012; 91: 106168. 39 Bhutta ZA, Rizvi A, Raza F, et al. A comparative evaluation of multiple micronutrient and iron-folic acid supplementation during pregnancy in Pakistan: impact on pregnancy outcomes. Food Nutr Bull 2009; 30 (suppl): S496505. 40 Humphrey JH, Hargrove JW, Malaba LC, et al, and the ZVITAMBO Study Group. HIV incidence among post-partum women in Zimbabwe: risk factors and the effect of vitamin A supplementation. AIDS 2006; 20: 143746. 41 Lawn JE, Mwansa-Kambafwile J, Barros FC, Horta BL, Cousens S. Kangaroo mother care to prevent neonatal deaths due to pre-term birth complications. Int J Epidemiol 2010; 39 (suppl 1): 14454. 42 Mwansa-Kambafwile J, Cousens S, Hansen T, Lawn JE. Antenatal steroids in preterm labour for the prevention of neonatal deaths due to complications of preterm birth. Int J Epidemiol 2010; 39 (suppl 1): i12233. 43 March of Dimes, The Partnership of Maternal, Newborn, and Child Health, Save the Children, World Health Organization. Born too soon: the global action report on preterm birth. Geneva, 2012. 44 Mehra S, Agrawal D. Adolescent health determinants for pregnancy and child health outcomes among the urban poor. Indian Pediatr 2004; 41: 13745. 45 Bott SJS, Shah I, Puri C, eds. Towards adulthood: exploring the sexual and reproductive health of adolescents in South Asia. Geneva: World Health Organization, 2000. 46 Monden CW, Smits J. Maternal height and child mortality in 42 developing countries. Am J Hum Biol 2009; 21: 30511. 47 Black RE, Victora CG, Walker SP, and the Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S01406736(13)60937-X. 48 Bhutta ZA, Das JK, Rizvi A, et al, The Lancet Interventions Review Group, and the Maternal and Child Nutrition Study Group. Evidence based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet 2013; published online June 6. http://dx.doi.org/10.1016/S0140-6736(13)60996-4.

108

www.thelancet.com

Articles

Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: findings from five birth cohort studies
Linda S Adair, Caroline H D Fall, Clive Osmond, Aryeh D Stein, Reynaldo Martorell, Manuel Ramirez-Zea, Harshpal Singh Sachdev, Darren L Dahly, Isabelita Bas, Shane A Norris, Lisa Micklesfield, Pedro Hallal, Cesar G Victora, for the COHORTS group

Summary

Background Fast weight gain and linear growth in children in low-income and middle-income countries are associated with enhanced survival and improved cognitive development, but might increase risk of obesity and related adult cardiometabolic diseases. We investigated how linear growth and relative weight gain during infancy and childhood are related to health and human capital outcomes in young adults. Methods We used data from five prospective birth cohort studies from Brazil, Guatemala, India, the Philippines, and South Africa. We investigated body-mass index, systolic and diastolic blood pressure, plasma glucose concentration, height, years of attained schooling, and related categorical indicators of adverse outcomes in young adults. With linear and logistic regression models, we assessed how these outcomes relate to birthweight and to statistically independent measures representing linear growth and weight gain independent of linear growth (relative weight gain) in three age periods: 02 years, 2 years to mid-childhood, and mid-childhood to adulthood. Findings We obtained data for 8362 participants who had at least one adult outcome of interest. A higher birthweight was consistently associated with an adult body-mass index of greater than 25 kg/m (odds ratio 128, 95% CI 121135) and a reduced likelihood of short adult stature (049, 044054) and of not completing secondary school (082, 078087). Faster linear growth was strongly associated with a reduced risk of short adult stature (age 2 years: 023, 020052; midchildhood: 039, 036043) and of not completing secondary school (age 2 years: 074, 067078; mid-childhood: 087, 083092), but did raise the likelihood of overweight (age 2 years: 124, 117131; mid-childhood: 112, 106118) and elevated blood pressure (age 2 years: 112, 106119; mid-childhood: 107, 101113). Faster relative weight gain was associated with an increased risk of adult overweight (age 2 years: 151, 143160; mid-childhood: 176, 169191) and elevated blood pressure (age 2 years: 107, 101113; mid-childhood: 122, 115130). Linear growth and relative weight gain were not associated with dysglycaemia, but a higher birthweight was associated with decreased risk of the disorder (089, 081098). Interpretation Interventions in countries of low and middle income to increase birthweight and linear growth during the first 2 years of life are likely to result in substantial gains in height and schooling and give some protection from adult chronic disease risk factors, with few adverse trade-offs. Funding Wellcome Trust and Bill & Melinda Gates Foundation.

Published Online March 28, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)60103-8 See Online/Comments http://dx.doi.org/10.1016/ S0140-6736(13)60716-3 Department of Nutrition, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA (Prof L S Adair PhD); MRC Lifecourse Epidemiology Unit, University of Southampton, Southampton General Hospital, Southampton, UK (Prof C H D Fall DM, Prof C Osmond PhD); Hubert Department of Global Health, Rollins School of Public Health, Emory University, Atlanta, GA, USA (Prof A D Stein PhD, Prof R Martorell PhD); Unit of Nutrition and Chronic Diseases, Institute of Nutrition of Central America and Panama, Guatemala City, Guatemala (M Ramirez-Zea MD); Sitaram Bhartia Institute of Science and Research, New Delhi, India (Prof H S Sachdev MD); Centre for Epidemiology and Biostatistics, University of Leeds, Leeds, UK (D L Dahly PhD); Office of Population Studies Foundation, University of San Carlos, Cebu, Philippines (I Bas MA); Medical Research Council/Wits Developmental Pathways for Health Research Unit, Department of Pediatrics, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa (S A Norris PhD, L Micklesfield PhD); and Universidade Federal de Pelotas, Pelotas, Brazil (P Hallal PhD, Prof C G Victora MD) Correspondence to: Prof Linda S Adair, Carolina Population Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 275162524, USA linda_adair@unc.edu

Introduction
Promotion of faster weight gain in children with growth failure is an important goal of paediatric care in countries of low and middle income, in view of the well known associations between poor growth and impaired cognitive development and increased morbidity and mortality. However, some studies suggest that rapid weight gain in the first 2 years of life is related to an increased risk of obesity1,2 and insulin resistance3,4 later in life. Reports about this so-called catch-up dilemma57 draw attention to the potential risks and benefits of faster early growth. Much of the concern is based on studies of weight gain in children in high-income countries, without a concomitant consideration of how linear growth and weight gain relative to linear growth affect outcomes later in life.
www.thelancet.com

Separation of their effects is important because, although early linear growth strongly predicts adult height,8 lean body mass,9,10 attained schooling,11 employment,12 and earnings,13,14 excess adiposity is a well recognised risk factor for cardiometabolic diseases. The key questions for paediatricians and policy makers are: what is the optimum age for promotion of growth for enhanced survival and human capital, and will this promotion necessarily lead to an increase in cardiometabolic disease? We explored how birthweight, linear growth, and weight gain relative to linear growth in childhood and adulthood are related to body size and composition, cardiometabolic risk factors (blood pressure and plasma glucose concentrations), and human capital outcomes (height and attained schooling) in young adults. We

109

Articles

For more on the first 1000 days see http://www.thousanddays.org

compared results across these diverse outcomes to explore possible trade-offs. We focused on birth and the period from 0 to 2 years to provide additional insights into the long-term importance of the first 1000 days widely recognised as a particularly sensitive period for child health and development.

Methods
For details of each cohort study see http://www.cohortsgroup.org

Study design and participants


We used data from five prospective birth cohort studies included in the Consortium for Health Orientated Research in Transitional Societies (COHORTS):15 the 1982 Pelotas Birth Cohort (Brazil);16 the Institute of Nutrition of Central America and Panama Nutrition Trial Cohort (Guatemala);17 the New Delhi Birth Cohort (India);10 the Cebu Longitudinal Health and Nutrition Survey (Philippines);18 and the Birth to Twenty Cohort (South Africa).19 All studies of these cohorts were reviewed and approved by appropriate ethics review boards.

Procedures
We identified common ages of childhood measurements across the five sites, and focused on birthweight, birth length (available for Guatemala, India, and the Philippines), weight and length at age 2 years, and weight and height mid-childhood (aged 4 years for Brazil, Guatemala, India, and South Africa; 8 years for the Philippines). Birthweight was measured in hospitals at delivery in South Africa and Brazil, in the community by research staff within 72 h of birth in India, in hospitals or at home by birth attendants provided with project scales in the Philippines, and by a project nurse at home or in a health-care centre in Guatemala. Gestational age estimates were based on the mothers report of the date of her last menstrual period. Birthlength was measured by trained research staff within 72 h of birth in India, within 6 days in the Philippines, and 15 days after birth in Guatemala. Statistical methods are needed to separate the effects of linear growth and weight gain because they are strongly correlated, as are repeat measurements taken in the same individual. We derived standardised residuals by regressing current size (represented as Fisher-Yates transformed weight-for-age and length-for-age Z scores) on all previous size measures to produce conditional size measures.20,21 Conditional height is present length or height accounting for previous length or height, and weight (but not present weight). Conditional relative weight is present weight accounting for present height and all previous weight and height measures. For example, adult conditional relative weight is derived from a regression of adult weight on adult height, weight and height at mid-childhood, weight and length when aged 2 years, and birthweight. Conditional variables represent childrens deviation from the expected size on the basis of their own previous measures and the growth of the other children in each cohort, and can be interpreted as representing faster or
110

slower relative weight gain or linear growth. For example, a child with a positive value for mid-childhood conditional height is taller than expected in view of previous size and thus had a faster rate of linear growth than would be expected from age 2 years to mid-childhood. Our expo sure variables represent birthweight and conditional height and conditional relative weight when aged 2 years, during mid-childhood, and during adulthood. Equiva lents of the conditional variables (g or kg, and cm) are shown in the appendix (p 20). We calculated conditional variables separately for each site and sex from the largest sample with complete nonpregnant anthropometric data at all selected ages to avoid bias as a result of including only individuals with adult outcomes. Because birthlength was not available for Brazil and South Africa, we included only birthweight in the regression models for all sites. In separate analyses of the three sites for which data for birthlength were available, we compared results with and without the inclusion of birthlength. Associations of the resulting conditional relative weight and conditional height variables with later outcomes were similar with each method. Height, weight, and waist circumference were measured with standard techniques. Binary variables represent over weight and obesity (body-mass index [BMI] 25 kg/m or the International Obesity Task Force equivalent for individuals aged <18 years22) and short stature (height-forage Z score23 <2 when aged <19 years; or <1501 cm for women and <1619 cm for men if aged 19 years). Fat mass, fat-free mass, and percentage body fat were calculated with site-specific methods (bioelectrical impe dance in Brazil; validated equations including weight, height, and abdominal circumference in Guatemala, or skinfold thickness in India and the Philippines; and dual x-ray absorptiometry [Delphi, Hologic] in South Africa).9 Within each site and sex, we used a Fisher-Yates trans formation to express fat mass and fat-free mass in SD units to enable comparisons within and between sites. Blood pressure was measured by aneroid sphyg momanometer in Brazil, mercury sphygmo manometer in the Philippines, and digital devices elsewhere (UA-767 [A and D Medical] in Guatemala; Omron M6 [Omron] in South Africa; and Omron 711 [Omron] in India). Participants were measured seated after 510 min of rest, with appropriate cuff sizes. We used the mean of three measurements (Philippines), two measurements (Brazil and India), or the second and third of three measurements (Guatemala and South Africa). In accordance with inter national recommendations, elevated blood pressure was defined as a systolic blood pressure of 130 mm Hg or more or a diastolic blood pressure of 85 mm Hg or more24 in view of the young ages and low prevalence of hyper tension in several cohorts. Fasting blood glucose concentration was measured in all cohorts except in Brazil, where a random finger-prick capillary whole-blood sample was obtained and adjusted for time since previous meal.25 Adjustments were made
www.thelancet.com

Articles

for differences in plasma, whole venous blood, and capillary values.26,27 Participants with glucose concen trations of 61 mmol/L or more, or those taking drugs for diabetes were considered to have dysglycaemia (ie, impaired fasting glucose or diabetes).28 The number of completed years of schooling was recorded in all sites.11 As an adverse outcome, a binary variable indicates failure to complete secondary school. Socioeconomic status at birth was represented by mothers education and a site-specific indicator of household wealth, which was a summary measure of housing type and ownership and type of dwelling (India), or an asset index based on household possessions (all other sites). Additionally, we tested whether inclusion of gestational age or premature birth

(gestational age <37 weeks) affected the coefficients of the conditional variables.

Statistical analysis
We used linear regression for continuous outcomes and mate logistic regression for categorical outcomes to esti associations with conditional height and con ditional relative weight at different ages. Models for each outcome included the largest sample with com plete data. We excluded women who were pregnant at the time of measure ment from all analyses. We compared our main analysis sample with samples with adult data but incom plete anthropometry, and with those with birth measures but without adult outcome data by use of Bonferroniadjusted ANOVA to account for multiple comparisons. To

Brazil Men (n=1889) Birthweight (kg) Birthlength (cm) Gestational age (weeks) Adult age (years) Adult height (cm) Adult weight (kg) Adult body-mass index (kg/m2) Adult waist circumference (cm) Adult body fat (%) Adult fat mass (kg) Adult fat-free mass (kg) Adult systolic blood pressure (mm Hg) Adult diastolic blood pressure (mm Hg) Adult plasma glucose concentration (mmol/L) Number of completed years of schooling Mothers height (cm) Number of years mother spent in school Body-mass index >25 kg/m2 Elevated blood pressure Dysglycaemia Short stature Did not complete secondary school 329 (052) NA 393 (19) 227 (04) 1738 (69) 721 (140) 238 (41) 809 (102) 163 (38) 122 (50) 604 (96) 1235 (144) 756 (117) 51 (07) 89 (32) 1567 (61) 65 (41) 584/1882 (310%) 666/1884 (354%) 148/1590 (93%) 74/1883 (39%) 952/1770 (538%) Women (n=1702) 317 (051) NA 391 (30) 227 (04) 1609 (62) 606 (126) 234 (47) 748 (105) NA* NA* NA* 1112 (130) 716 (107) 49 (07) 99 (31) 1563 (58) 66 (43) 437/1701 (257%) 224/1702 (132%) 98/1503 (65%) 57/1701 (34%) 651/1619 (402%)

Guatemala Men (n=171) 309 (052) 498 (24) 395 (30) 314 (14) 1630 (61) 643 (101) 241 (33) 853 (83) 194 (59) 130 (59) 513 (52) 1170 (104) 726 (92) 52 (06) 50 (35) 1485 (48) 13 (16) 54/157 (344%) 29/161 (180%) 4/114 (35%) 69/157 (439%) 154/168 (917%) Women (n=155) 302 (043) 488 (21) 387 (26) 311 (13) 1511 (52) 613 (115) 268 (46) 923 (112) 349 (69) 221 (85) 391 (33) 1077 (112) 693 (90) 51 (11) 48 (35) 1485 (53) 12 (15) 86/141 (610%) 10/152 (66%) 5/133 (37%) 59/141 (418%) 144/154 (935%)

India Men (n=775) 285 (044) 486 (21) 391 (25) 291 (13) 1696 (62) 718 (138) 249 (43) 903 (119) 242 (58) 180 (69) 538 (78) 1184 (114) 780 (103) 55 (11) 132 (34) 1520 (56) 57 (45) 370/775 (477%) 209/769 (272%) 161/762 (211%) 78/775 (101%) 114/775 (147%) Women (n=553) 276 (039) 481 (20) 386 (22) 292 (14) 1549 (56) 592 (131) 246 (50) 796 (122) 342 (69) 209 (79) 381 (57) 1068 (111) 736 (92) 53 (08) 139 (31) 1519 (50) 58 (45) 253/551 (459%) 67/546 (123%) 75/540 (139%) 105/551 (191%) 42/553 (76%)

Philippines Men (n=1006) 303 (042) 494 (20) 389 (21) 213 (08) 1630 (59) 560 (93) 210 (31) 721 (75) 167 (51) 97 (45) 463 (58) 1118 (109) 758 (96) 47 (06) 103 (34) 1506 (49) 70 (33) 102/1004 (102%) 174/1004 (173%) 8/870 (09%) 432/1004 (430%) 370/1006 (368%) Women (n=889) 298 (041) 489 (20) 382 (19) 211 (10) 1511 (54) 463 (79) 203 (31) 677 (73) 327 (48) 154 (46) 310 (42) 994 (100) 678 (86) 45 (05) 115 (28) 1505 (49) 69 (32) 68/883 (77%) 20/885 (23%) 7/709 (10%) 374/888 (421%) 185/889 (208%)

South Africa Men (n=584) 314 (053) NA 381 (19) 181 (06) 1708 (78) 601 (117) 206 (37) NA 143 (57) 89 (53) 509 (65) 1206 (107) 709 (88) 47 (05) 108 (15) 1585 (66) 97 (25) 54/573 (94%) 118/555 (213%) 4/358 (11%) 41/573 (71%) 271/568 (477%) Women (n=638) 302 (048) NA 393 (19) 181 (06) 15965 (621) 578 (105) 227 (39) NA 324 (64) 194 (74) 389 (55) 1147 (98) 719 (85) 45 (04) 114 (13) 1581 (65) 96 (25) 162/630 (257%) 60/617 (97%) 1/367 (03%) 36/631 (57%) 171/621 (275%)

Data are mean (SD) or n/N (%). Participants had at least one adult outcome of interest and complete anthropometric data. NA=not available. *Female body composition not measured in Brazil because examination occurred at an army recruitment centre. Systolic blood pressure 130 mm Hg or diastolic blood pressure 85 mm Hg.24 Plasma glucose concentration >61 mmol/L or taking drugs for diabetes.24 Height-for-age Z score <2 when aged <19 years; or <1501 cm for women or <1619 cm for men if aged 19 years.

Table 1: Characteristics of participants in the five cohorts

www.thelancet.com

111

Articles

A
20 Change in body-mass index (kg/m2) per SD change in conditional growth variable

Men Women Pooled data for all men with 95% CI Brazil Guatemala India

Pooled data for all women with 95% CI Philippines South Africa

15

10

05

05

Birthweight*

Age 2 years*

Mid-childhood*

Age 2 years*

Mid-childhood

B
06 05 Change in fat mass (SD) per SD change in conditional growth variable 04 03 02 01 0 01

Conditional relative weight

Conditional height

Birthweight*

Age 2 years*

Mid-childhood*

Age 2 years*

Mid-childhood*

Conditional relative weight

Conditional height

C
06 05 04 03 02 01 0 01

Change in fat-free mass (SD) per SD change in conditional growth variable

assess potential bias related to attrition and missing data, we compared results estimated by Heckman29 selection models with standard regression models. We assessed site and sex heterogeneity with F tests, allowing other predictors (eg, age and socioeconomic status) to vary across sites. To measure the variation of effect sizes between combinations of sex and site, we used restricted maximum likelihood to estimate the SD of the effect size for each site and sex around their pooled value. This SD value is close to the SD of estimates specific to each site and sex for each growth variable with each outcome. When it is zero, the effect sizes for each sex and site are compatible with a fixed-effects model. Glucose and blood-pressure models included birth weight, and conditional height and conditional relative weight when aged 2 years, at mid-childhood, and during adulthood. Because attained schooling cannot be affected by subsequent growth in people who dropped out as children or adolescents, schooling models exclude adult conditional relative weight and conditional height. Adult size outcomes (height, BMI, and body composition measures) also exclude adult conditional height and con ditional relative weight. Because conditional height and conditional relative weight variables are not correlated, they can be included together in models without concerns about colinearity. Because we noted heterogeneity by site and sex for at least one conditional relative weight or conditional height variable in most models, we present results stratified by sex and site; stratified by sex and adjusted for site; and pooled, adjusted for site and sex. Because of the large number of outcomes, we created detailed tables (appendix) that include all estimated coefficients, but summarised results graphically to present site and sex-specific coefficients, with sex-stratified pooled results. We adjusted all models for adult age. Adjustment for gestational age or socioeconomic status at birth did not meaningfully alter coefficients or p values in the models for body composition or cardiometabolic outcomes, so for these models we present results unadjusted for these factors. We adjusted height and schooling models for socioeconomic status at birth, and additionally adjusted height for maternal height to account for differences in genetic growth potential. Information about paternal height was not available. We used Stata (version 12) for analyses.

Role of the funding source


The sponsors of the study provided financial support for data management and data analysis, but had no other role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and all authors made the decision to submit for publication.

Birthweight

Age 2 years

Mid-childhood*

Age 2 years

Mid-childhood*

Conditional relative weight

Conditional height

Figure 1: Association of birthweight, conditional relative weight and conditional height with (A) body-mass index, (B) fat mass, and (C) fat-free mass Site-specific datapoints represent coefficients from linear regression models done separately for each site and sex. *Significant heterogeneity between sexes and sites.

Results
We obtained data for 8362 participants who had provided information about at least one adult outcome of interest
www.thelancet.com

112

Articles

Body-mass index >25 kg/m2 Overall number affected Birthweight Conditional relative weight at age 2 years Conditional relative weight mid-childhood Conditional height at age 2 years Conditional height mid-childhood 2170/8297 (262%) 128 (121135) 151 (143160) 176 (166186) 124 (117131) 112 (106118)

Elevated blood pressure* 1571/8233 (191%) 093 (088099) 107 (101113) 122 (115130) 112 (106119) 107 (101113)

Dysglycaemia 511/6947 (74%) 089 (081098) 095 (086104) 108 (098118) 098 (089118) 094 (086103)

Short stature 1325/8302 (160%) 049 (044054) 094 (086103) 113 (104123) 023 (020025) 039 (036043)

Did not complete secondary school 3054/8121 (376%) 082 (078087) 095 (090099) 104 (099110) 074 (067078) 087 (083092)

Data are n/N (%) or odds ratio (95% CI). Odds ratios calculated with logistic regression models of the pooled sample and indicate how a difference of 1 SD in each measure affects the likelihood of the adverse outcome. All models were adjusted for adult age. The models for short stature and completion of secondary school were further adjusted for mothers education and household wealth at birth; the model for short stature was also adjusted for mothers height. *Systolic blood pressure 130 mm Hg or diastolic blood pressure 85 mm Hg. Plasma glucose concentration 61 mmol/L or taking drugs for diabetes. Height-for-age Z score <2 when aged <19 years; <1501 cm for women or <1619 cm for men when aged 19 years.

Table 2: Association of birthweight, conditional relative weight gain, and conditional height with adverse outcomes

(table 1). Participants who were lost to follow-up or excluded because of missing data rarely differed signifi cantly from those who were included (appendix p 2). Mean adult age at which measurements were taken ranged from about 18 years in South Africa to more than 31 years in Guatemala (table 1). The prevalence of adverse outcomes generally reflects the age and relative economic development of the five samples. Adult height and fatfree mass were highest in Brazilian men (table 1). Dysglycaemia was too rare to analyse in sex-stratified models in the South African cohort (table 1). Higher birthweight and conditional relative weight at age 2 years and mid-childhood were consistently asso ciated with higher adult BMI in men and women (figure 1, appendix p 5). The association between conditional relative weight and BMI strengthened with age at measure ment: overall, BMI increased by 05 kg/m per SD increase in birthweight, by 10 kg/m at 2 years, and 13 kg/m in mid-childhood. Higher conditional height at 2 years and mid-childhood was also associated with higher adult BMI, but coefficients were smaller than were those for conditional relative weight and diminished with age (figure 1). The findings were similar for the binary outcome of obesity (table 2, appendix p 6). Heterogeneity between sexes and sites was primarily a result of size rather than direction of coefficients. All conditional relative weight and conditional height variables were associated with higher adult fat mass and fat-free mass (figure 1, appendix pp 78). Coefficients for birthweight and relative weight increased with the age at which size was measured, while conditional height coefficients decreased (figure 1). Birthweight and conditional relative weight at age 2 years were more strongly associated with fat-free mass than with fat mass, contrasting with stronger associations of con ditional relative weight in mid-childhood with fat mass than with fat-free mass (appendix pp 78). Conditional height at age 2 years and mid-childhood were more strongly associated with fat-free mass than with fat mass, and both associations were stronger at 2 years than at mid-childhood. The findings for percentage body fat and waist circumference (appendix pp 910, 2122)
www.thelancet.com

were similar to those for fat mass and BMI. Heterogeneity between sexes and sites was greater for fat mass than for fat-free mass. Birthweight was unrelated to systolic blood pressure (figure 2, appendix p 11). Higher conditional relative weight at all ages was associated with higher systolic blood pressure, and the associations strengthened with increasing age at measurement (figure 2). Conditional height at age 2 years and mid-childhood was also positively related to systolic blood pressure (figure 2). The association between birthweight and systolic blood pressure was not the same in Guatemala as in other sites (figure 2, appendix p 11). The coefficients for asso ciations between conditional relative weight at midchildhood and adulthood and conditional height at 2 years were larger in men than in women (appendix p 11). The findings were similar for diastolic blood pressure and elevated blood pressure (table 2, appendix pp 1213, 23), except that higher birthweight was associated with a reduced likelihood of elevated blood pressure in adulthood (table 2). Conditional relative weight in adulthood and, in women only, in mid-childhood were associated with higher plasma glucose concentration in adulthood (figure 2, appendix p 14). Conditional relative weight at age 2 years was unrelated to glucose, and birthweight was inversely associated with plasma glucose concentration in women (figure 2, appendix p 14). Conditional height at any age was unrelated to adult plasma glucose concentration (figure 2, appendix p 14). The findings were similar for dysglycaemia (table 2, appendix p 15). Heterogeneity between sexes and sites was significant only for conditional height at age 2 years (appendix). Each SD of birthweight predicted 15 cm higher adult height in the pooled sample (figure 3). Conditional relative weight at age 2 years was unrelated to adult height, but conditional relative weight at mid-childhood was inversely related (figure 3, appendix p 16). By contrast, increased conditional height at age 2 years and midchildhood strongly predicted an increase in adult height in all sites (32 cm per SD increase in conditional height at age 2 years and 19 cm per SD mid-childhood; figure 3,

See Online for appendix

113

Articles

Men Women Pooled data for all men with 95% CI Brazil Guatemala India

Pooled data for all women with 95% CI Philippines South Africa

60 50 40 30 20 10 0 10 20

Change in systolic blood pressure (mm Hg) per SD change in conditional growth variable

inversely related to attained schooling in both sexes (appendix p 19); by contrast, higher conditional height at 2 years was strongly related to higher attained schooling (except in South Africa), with one SD in conditional height at 2 years relating to a half year increase in the pooled sample (figure 3). Lower birthweight and conditional height at age 2 years were associated with increased odds of failing to complete high school (table 2, appendix p 19). Overall, heterogeneity reflected differences between sites rather than between sexes. Estimations produced by Heckman selection models did not produce coefficients that were substantially different from those estimated with standard linear and logistic regression models.

Discussion
Birthweight Age 2 years Midchildhood Adult* Age 2 years* Midchildhood Adult

Conditional relative weight

Conditional height

B
03 Change in log plasma glucose concentration (SD) per SD change in conditional growth variable

02

01

01

02

03

Birthweight

Age 2 years

Midchildhood

Adult

Age 2 years*

Midchildhood

Adult

Conditional relative weight

Conditional height

Figure 2: Association of birthweight, conditional relative weight, and conditional height with (A) systolic blood pressure and (B) log plasma glucose concentrations Site-specific datapoints represent coefficients from linear regression models done separately for each site and sex. *Significant heterogeneity between sexes and sites.

appendix p 16). Heterogeneity was a result of site and sex differences in the size rather than direction of coefficients (coefficients in men more than coefficients in women). Birthweight and conditional height at age 2 years and mid-childhood were strongly inversely related to the chance of short adult stature (table 2, appendix p 17). One SD higher birthweight predicted about 02 years more schooling (figure 3, appendix p 18). One SD higher conditional relative weight at 2 years was associated with 014 years more schooling in men, but there was no association in women. Conditional relative weight at midchildhood was unrelated to attained schooling (figure 3). In Guatemala, conditional relative weight at 2 years was
114

We have shown that higher birthweight is related to increased liklihood of adult overweight in five populations in countries of low and middle income, but to lower likelihood of other adverse outcomes. Although higher conditional relative weight at age 2 years was associated with higher risk of adult overweight and a slightly increased risk of elevated blood pressure, it was unrelated to dysglycaemia, adult stature, or educational attainment. By contrast, conditional relative weight in mid-childhood was associated with higher likelihood of adult overweight and elevated blood pressure, but was unrelated to schooling. Although associated with slightly increased likelihood of adult overweight (mostly related to lean mass) and elevated blood pressure, higher conditional heights at age 2 years and at mid-childhood were related to lower risk of short stature and poor educational attainment. Few studies from low-income and middle-income settings have followed up cohorts to adulthood. Individuals in the cohorts that we studied were born when poor early life nutrition was more common than it is now, which is shown by the high prevalence of stunting during infancy.30 However, the individuals grew up in rapidly changing environments that fostered development of obesity and chronic disease risk. Countries that are challenged by the dual burden of persistent undernutrition and emerging obesity need information about the many effects of early child growth, particularly during the important first 1000 days. Although the studies included in our analysis cannot be fully representative of the countries in which they were done, the many similarities of results across settings suggest that results may be generalised to other low-income and middle-income settings. Another strength of our study is that we were able to separate linear growth from relative weight gain. Weight gain is a result of linear growth and soft tissue gain (fat mass and fat-free mass); our conditional relative weight variables represent weight change that is separated from change in height. Conditional relative weight and conditional height variables are uncorrelated, and we expressed them in SD units to allow direct comparison of coefficients within regression models. Our variables
www.thelancet.com

Articles

therefore have advantages when compared with other representations of growth, and give more nuanced results than do those that are based on weight gain alone. In our study, one SD in conditional relative weight at 2 years corresponds to change in weight-for-age Z score from birth to 2 years that is slightly less than the 067 units typically used to define rapid weight gain.5 In a previous COHORTS report,11 one SD of conditional weight at age 2 years related to 04 years more schooling. Here, we showed that attained schooling was pre dominantly associated with conditional height in the first 2 years, and an increase in conditional relative weight in this period contributed to a rise of only 009 years of schooling. These findings are important, because an extra half year of schooling is associated with a 5% annual return to income.11 Each estimate for conditional relative weight or con ditional height represents an independent age-specific association. We noted that associations of conditional relative weight with adult adiposity and cardiometabolic risk factors strengthened substantially with age. Thus, the most important period for adverse associations of higher relative weight gain with adult cardiometabolic risk was after the age of 2 years. This finding is consistent with the ALSPAC study31 in the UK, which also showed that associations of conditional weight for length after infancy with systolic blood pressure and BMI were strongest at age 10 years, and with previous studies3234 that showed increased risks related to fast weight gain after infancy. By contrast, for cardiometabolic risk factors, conditional height coefficients were smaller than conditional relative weight coefficients and decreased with age, which is consistent with a Brazilian study.35 In view of the nature of conditional growth variables, we could assess independent associations of birthweight with subsequent outcomes, without concern about statistical issues such as the reversal paradox.36 Furthermore, in our analysis, gestational age was not a confounder of the relation between conditional variables and adult outcomes. A separate analysis of COHORTS data (unpublished) established that a premature birth or being small for gestational age were related to persistent deficits in adult height and schooling, but did not affect blood pressure or plasma glucose concentrations. We addressed potential trade-offs of different early growth patterns by comparing human capital outcomes known to be associated with impaired child growth and development with cardiometabolic disease risk factors. Birthweight and faster conditional relative weight in the first 2 years of life had little relation with adult cardio metabolic risk factors, but were important for fat-free mass. Indeed, early weight gain was more strongly associated with adult fat-free mass than with adult fat mass. This finding could explain the negative associations of birthweight and the small or non-significant associations of early relative weight gain with elevated blood pressure and dysglycaemia. Faster mid-childhood relative weight
www.thelancet.com

Men Women Pooled data for all men with 95% CI Brazil Guatemala India

Pooled data for all women with 95% CI Philippines South Africa

40

Change in adult height (cm) per SD change in conditional growth variable

30

20

10

10

Birthweight*

Age 2 years

Mid-childhood

Age 2 years*

Mid-childhood*

Conditional relative weight

Conditional height

B
10 Change in completed years of schooling per SD changein conditional growth variable

05

05

01

Birthweight

Age 2 years*

Mid-childhood

Age 2 years*

Mid-childhood*

Conditional relative weight

Conditional height

Figure 3: Association of birthweight, conditional relative weight, and conditional height with (A) adult height and (B) years spent at school Site-specific datapoints represent coefficients from linear regression models done separately for each site and sex. *Significant heterogeneity between sexes and sites.

gain was more strongly associated with fat mass than fatfree mass, and increased risk of elevated blood pressure and dysglycaemia. The positive associations of all growth variables at all ages with adult BMI and risk of overweight and obesity are probably a result of their joint effect on fat mass and fat-free mass, and the fact that overweight adults have highly variable amounts of body fat. Despite being restricted to participants with complete data, the sample that we analysed differed little from individuals who were excluded, and estimates produced with Heckman selection models were not different from models estimated with conventional methods. fore, our estimated coefficients do not seem to There have been biased by attrition or missing data. Some
115

Articles

Panel: Research in context Systematic review A 2008 systematic review30 examined how birthweight and weight-for-age and length-for-age Z scores at age 2 years related to adult human capital and chronic disease risk. Therefore, we searched PubMed for reports published in any language since Jan 1, 2009, which related birthweight and childhood growth rates to these same adult outcomes. We focused on studies that attempted to assess both weight gain and linear growth at different ages, and that involved appropriate statistical methods to differentiate their effects on subsequent outcomes. Our search terms were weight gain, linear growth, rapid weight gain, birth cohorts, and prospective studies. Few identified studies were done in low-income and middle-income countries, and few used the statistical methods needed to model correlated life course data. Although many reports associate rapid weight gain with obesity and related outcomes, we identified only two31,35 that directly compared how linear growth and relative weight gain related to adult blood pressure, glucose metabolism, body composition, height, or schooling. No reports directly compared several outcomes representing health and human capital to enable assessment of risks and benefits of early growth patterns. Interpretation We have studied life course determinants of young adult human capital and disease risk factors in five countries of low and middle income, by using appropriate statistical methods to directly compare how faster relative weight gain and faster linear growth relate to these outcomes. We have shown that effects on the different outcomes are age specific. Fast linear growth in the first 2 years of life is associated with increased adult height and amount of schooling. Adverse associations with fast relative weight gain are largely confined to mid-childhood and adulthood. Our data support the current focus on promotion of nutrition and linear growth in the first 1000 days of life (from conception to age 2 years), and also reinforce the importance of prevention of rapid relative weight gain after age 2 years. These findings have implications for present practices in low-income and middle-income countries, particularly emphasising the need to monitor linear growth as well as weight, and to avoid promotion of excess weight gain in children older than 2 years. Optimum growth patterns in early life are likely to lead to less undernutrition, increased human capital, and reduced risks of obesity and non-communicable diseases, thus addressing both components of the double burden of nutrition.

differ ences between our cohorts were unavoidable, such as methods for measurement of birthweight, plasma glucose concentration, blood pressure, and adult body composition, and also age at mid-childhood and adult hood. Our choice of age intervals was determined by the availability of measurements in each
116

cohort rather than what we might have chosen for the greatest biological signal, such as within the first 2 years. Heterogeneity, as represented by significant site or sex interactions, or both, with key exposures, could be a result of the differences between cohorts or true effect modifi cation. Although data for 48 months were not available in the Philippines, mid-childhood estimates did not differ systematically from other sites. Although our cohorts differed in childhood size and adult cardio metabolic risk profile, significant site and sex hetero geneity mostly reflected differences in size of asso ciations, not direction. Therefore, use of pooled analyses to draw conclusions seems to be justified. Outliers were usually from Guatemala (which provided the smallest sample) or South Africa (the youngest). For example, the unique absence of an association between early height gain and schooling in South Africa is probably because participants aged 18 years had no opportunity for education after secondary school. Additionally, the absence of significant predictors of glu cose concentration in South Africa is probably a result of the cohorts young age and little variation in glucose concentration (few concentrations were >6 mmol/L). Our findings suggest that interventions to increase birthweight and linear growth during the first 2 years of life are likely to result in substantial gains in height and schooling (key aspects of human capital), and give some protection from development of adult chronic disease risk factors, with no or negligible adverse trade-offs. Consistent with a growing body of research, our results indicate that faster relative weight gain after the age of 2 years has little benefit for human capital, and weight gain after midchildhood could lead to large adverse effects on later cardiovascular risk factors. Notably, this finding is particularly true for weight gain that is not accompanied by height gain. Our data support the present focus on promotion of improved nutrition and linear growth in the first 1000 days of life, and also reinforce the importance of prevention of rapid relative weight gain after age 2 years. Although the associations we describe do not prove causation, our results challenge several programmes in countries of low and middle income (panel). Rapid weight gain should not be promoted after the age of 23 years in children who are underweight but not wasted. Growth monitoring programmes should incorporate length and height measures, not just weight measures. New interventions that specifically promote linear growth instead of weight gain should be developed, tested, and promoted; exclusive breastfeeding, high-quality protein (eg, animal), and micronutrients could be further investigated.37 Traditional school feeding programmes that increase BMI with little effect on height38 might be doing more harm than good in terms of future health. Mortality and undernutrition are falling substantially in most parts of the world, except for Sub-Saharan Africa, and new targets are being formulated to replace the present set of 2015 Millennium Development Goals.39 Our
www.thelancet.com

Articles

analyses provide strong justification for the proposal of a new goal for optimum linear growth that is expressed as a reduction in stunting. This goal should replace the present target of a reduction in underweight alone, which is one of the indicators for the first Millennium Development Goals towards the eradication of extreme poverty. Whereas promotion of linear growth in early life could build human capital in adults without increasing the burden of noncommunicable diseases, the present focus on underweight might have detrimental repercussions, particularly if interventions take place after 1000 days.
Contributors LSA, CHDF, and CO led the writing team. LSA and CO did data analysis. LSA and CHDF wrote the report and had final responsibility for the content. All authors contributed to data collection, and reviewed and commented on the report and approved its final content. COHORTS group members Universidade Federal de Pelotas, Brazil: Cesar G Victora, Fernando C Barros, Denise Gigante, Pedro C Hallal, Bernardo L Horta; Hubert Department of Global Health, Emory University, USA: Reynaldo Martorell, Aryeh Stein, Yaw Addo, Wei Hao; Instituto de Nutricin de Centra Amrica y Panam, Guatemala: Manual Ramirez-Zea; Sunderlal Jain Hospital, India: Santosh K Bhargava; Sitaram Bhartia Institute of Science and Research, India: Harshpal Singh Sachdev; MRC Lifecourse Epidemiology Unit, UK: Caroline Fall, Clive Osmond; University of North Carolina at Chapel Hill, USA: Linda Adair; Office of Population Studies, University of San Carlos, Philippines: Isabelita Bas, Nanette Lee, Judith Borja; University of Leeds, UK: Darren Dahly; Northwestern University, USA: Christopher Kuzawa; Developmental Pathways for Health Research Unit, University of the Witwatersrand, South Africa: Linda Richter, Shane Norris, Julia De Kadt. Conflicts of interest We declare that we have no conflicts of interest. Acknowledgments The COHORTS collaboration is funded by the Wellcome Trust (089257/Z/09/Z). Additional funding for data analysis for this study was provided by the Bill & Melinda Gates Foundation (OPP1020058). The cohort studies in Guatemala and the Philippines were funded by the US National Institutes of Health; that in Brazil by the Wellcome Trust; that in India by the Indian Council of Medical Research, US National Center for Health Statistics, UK Medical Research Council, and British Heart Foundation; and that in South Africa by the Wellcome Trust, Human Sciences Research Council, South African Medical Research Council, South-African Netherlands Programme on Alternative Development, Anglo American Chairmans Fund, and University of the Witwatersrand in South Africa. We thank Jane Pearce for producing graphs and formatting tables. References 1 Stettler N, Iotova V. Early growth patterns and long-term obesity risk. Curr Opin Clin Nutr Metab Care 2010; 13: 29499. 2 Demerath EW, Reed D, Choh AC, et al. Rapid postnatal weight gain and visceral adiposity in adulthood: the Fels Longitudinal Study. Obesity (Silver Spring) 2009; 17: 206066. 3 Leunissen RW, Stijnen T, Hokken-Koelega AC. Influence of birth size on body composition in early adulthood: the programming factors for growth and metabolism (PROGRAM)-study. Clin Endocrinol (Oxf) 2009; 70: 24551. 4 Dulloo AG. Thrifty energy metabolism in catch-up growth trajectories to insulin and leptin resistance. Best Pract Res Clin Endocrinol Metab 2008; 22: 15571. 5 Ong KK. Catch-up growth in small for gestational age babies: good or bad? Curr Opin Endocrinol Diabetes Obes 2007; 14: 3034. 6 Hales CN, Ozanne SE. The dangerous road of catch-up growth. J Physiol 2003; 547: 510. 7 Victora CG, Barros FC, Horta BL, Martorell R. Short-term benefits of catch-up growth for small-for-gestational-age infants. Int J Epidemiol 2001; 30: 132530.

9 10

11

12 13 14 15 16 17 18 19 20

21 22 23 24 25 26 27

28

29 30

Stein AD, Wang M, Martorell R, et al. Growth patterns in early childhood and final attained stature: data from five birth cohorts from low- and middle-income countries. Am J Hum Biol 2010; 22: 35359. Kuzawa CW, Hallal PC, Adair L, et al. Birth weight, postnatal weight gain, and adult body composition in five low and middle income countries. Am J Hum Biol 2012; 24: 513. Sachdev HS, Fall CH, Osmond C, et al. Anthropometric indicators of body composition in young adults: relation to size at birth and serial measurements of body mass index in childhood in the New Delhi birth cohort. Am J Clin Nutr 2005; 82: 45666. Martorell R, Horta BL, Adair LS, et al. Weight gain in the first two years of life is an important predictor of schooling outcomes in pooled analyses from five birth cohorts from low- and middle-income countries. J Nutr 2010; 140: 34854. Carba DB, Tan VL, Adair LS. Early childhood length-for-age is associated with the work status of Filipino young adults. Econ Hum Biol 2009; 7: 717. Thomas D, Frankenberg E. Health, nutrition and prosperity: a microeconomic perspective. Bull World Health Organ 2002; 80: 10613. Hoddinott J, Maluccio JA, Behrman JR, Flores R, Martorell R. Effect of a nutrition intervention during early childhood on economic productivity in Guatemalan adults. Lancet 2008; 371: 41116. Richter LM, Victora CG, Hallal PC, et al. Cohort profile: the Consortium of Health-Orientated Research in Transitioning Societies. Int J Epidemiol 2012; 41: 62126. Victora CG, Barros FC. Cohort profile: the 1982 Pelotas (Brazil) birth cohort study. Int J Epidemiol 2006; 35: 23742. Stein AD, Melgar P, Hoddinott J, Martorell R. Cohort profile: the Institute of Nutrition of Central America and Panama (INCAP) Nutrition Trial Cohort Study. Int J Epidemiol 2008; 37: 71620. Adair LS, Popkin BM, Akin JS, et al. Cohort profile: the Cebu longitudinal health and nutrition survey. Int J Epidemiol 2011; 40: 61925. Richter L, Norris S, Pettifor J, Yach D, Cameron N. Cohort profile: Mandelas children: the 1990 Birth to Twenty study in South Africa. Int J Epidemiol 2007; 36: 50411. Keijzer-Veen MG, Euser AM, van Montfoort N, Dekker FW, Vandenbroucke JP, Van Houwelingen HC. A regression model with unexplained residuals was preferred in the analysis of the fetal origins of adult diseases hypothesis. J Clin Epidemiol 2005; 58: 132024. Li H, Stein AD, Barnhart HX, Ramakrishnan U, Martorell R. Associations between prenatal and postnatal growth and adult body size and composition. Am J Clin Nutr 2003; 77: 1498505. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320: 124043. WHO. Child Growth Standards based on length/height, weight and age. Acta Paediatr Suppl 2006; 450: 7685. International Diabetes Federation. The IDF consensus worldwide definition of the metabolic syndrome. Brussels: International Diabetes Federation, 2006. Norris SA, Osmond C, Gigante D, et al. Size at birth, weight gain in infancy and childhood, and adult diabetes risk in five low- or middle-income country birth cohorts. Diabetes Care 2012; 35: 7279. Kumar G, Sng BL, Kumar S. Correlation of capillary and venous blood glucometry with laboratory determination. Prehosp Emerg Care 2004; 8: 37883. Petersen JR, Graves DF, Tacker DH, Okorodudu AO, Mohammad AA, Cardenas VJ Jr. Comparison of POCT and central laboratory blood glucose results using arterial, capillary, and venous samples from MICU patients on a tight glycemic protocol. Clin Chim Acta 2008; 396: 1013. WHO. Definition, diagnosis and classification of diabetes mellitus and its complications: report of a WHO Consultation, part 1: diagnosis and classification of diabetes mellitus. Geneva: World Health Organization, 1999. Heckman J. Sample selection bias as a specification error. Econometrica 1979; 47: 15361. Victora CG, Adair L, Fall C, et al. Maternal and child undernutrition: consequences for adult health and human capital. Lancet 2008; 371: 34057.

www.thelancet.com

117

Articles

31 Jones A, Charakida M, Falaschetti E, et al. Adipose and height growth through childhood and blood pressure status in a large prospective cohort study. Hypertension 2012; 59: 91925. 32 Law CM, Shiell AW, Newsome CA, et al. Fetal, infant, and childhood growth and adult blood pressure: a longitudinal study from birth to 22 years of age. Circulation 2002; 105: 108892. 33 Eriksson JG, Forsen TJ, Kajantie E, Osmond C, Barker DJ. Childhood growth and hypertension in later life. Hypertension 2007; 49: 141521. 34 Eriksson JG, Osmond C, Kajantie E, Forsen TJ, Barker DJ. Patterns of growth among children who later develop type 2 diabetes or its risk factors. Diabetologia 2006; 49: 285358. 35 Menezes AM, Hallal PC, Dumith SC, et al. Adolescent blood pressure, body mass index and skin folds: sorting out the effects of early weight and length gains. J Epidemiol Community Health 2012; 66: 14954.

36 Tu YK, Ellison GT, Gilthorpe MS. Growth, current size and the role of the reversal paradox in the foetal origins of adult disease: an illustration using vector geometry. Epidemiol Perspect Innov 2006; 3: 9. 37 Roberfroid D, Huybregts L, Lanou H, et al. Impact of prenatal multiple micronutrients on survival and growth during infancy: a randomized controlled trial. Am J Clin Nutr 2012; 95: 91624. 38 Bhutta ZA, Ahmed T, Black RE, et al. What works? Interventions for maternal and child undernutrition and survival. Lancet 2008; 371: 41740. 39 WHO. Countdown to 2015: building a future for women and children: the 2012 Report. Washington DC: WHO and UNICEF, 2012.

118

www.thelancet.com

www.thelancet.com