You are on page 1of 270

The nutritional status of

disadvantaged preschool children


attending daycares in Salvador,
Northeast Brazil:
a cross-sectional study

Rebecca Leigh Lander

A thesis submitted for the degree of

Doctor of Philosophy

at the University of Otago, Dunedin,

New Zealand

August, 2012
ii
Abstract

In the poor Northeast region of Brazil, grave concerns exist regarding the
prevalence of anaemia and vitamin A deficiency, and the increasing risk of
overweight among preschool children. Brazil has introduced pro-poor social
policies, such as preschool daycare programmes, yet comprehensive studies of the
growth, health, and micronutrient status of disadvantaged daycare children in NE
Brazil are lacking.
We conducted a cross-sectional study of 376 disadvantaged preschoolers
aged three to six years in seven philanthropic daycares in Salvador, NE Brazil and
assessed their growth and biomarkers of morbidity and micronutrient status;
explored potential predisposing factors to parasitic infections, micronutrient
deficiencies, and poor iron status; and examined multiple factors associated with
their somatic growth by a structural equation model (SEM).
Our results showed nearly half of the households were classified as having
extremely low SES and the remainder low SES. The prevalence of underweight,
wasting, and stunting in the preschoolers was low (< 10%), although 13% had
BMIZ-scores > 1SD. The energy density of the daycare meals was low (0.79
kcal/g), yet the nutrient supply was enhanced by iron and folic acid fortification
and was adequate, except for calcium, vitamin A, and thiamine.
Nearly 30% were infected with >1 parasite, with boys and children from
extremely low SES having more than twice the risk of helminth infections
(p<0.05). Helminths negatively impacted on selenium and vitamin B-12
biomarkers (p<0.05), whereas Giardia intestinalis was positively associated with
serum folate (p<0.05). Deworming treatment and vitamin A supplements were
inversely associated with helminths and Giardia, respectively.
Anaemia and zinc, selenium, folate, vitamin A, and B-12 deficiencies were
low (<10%), although nearly a third of children had evidence of chronic
inflammation (AGP > 1 g/L). More than 30% had a genetic haemoglobin disorder.
A significant inverse association was found between haemoglobin and α3.7

iii
thalassemia (p<0.001). The major positive predictors of haemoglobin were serum
selenium, retinol, and zinc, although the latter association was modest. Elevated
AGP was the only positive and significant predictor of serum ferritin (p=0.001).
However, serum soluble transferrin receptor was positively associated with Hb AS
and BMIZ > 1SD (p<0.05).
Based on the SEM, the significant positive direct effect on HAZ-scores was
maternal height (p<0.001), and the major negative associations were being male,
white, of high birth order, and having a helminth infection. Similarly, the main
positive predictor of BMIZ-scores was maternal weight (p<0.001), whereas
extremely low SES and being male were the significant negative direct effects on
BMIZ-scores (p<0.05).
In conclusion, our findings highlight the susceptibility of disadvantaged
preschoolers to both under- and overnutrition and the need for effective
deworming strategies. Overall adequate nutritional status appears to be associated
with some pro-poor initiatives, such as vitamin A supplementation and preschool
daycare programmes. The latter provided a protective environment, by supplying
fortified foods and a micronutrient-rich diet. Thus, we endorse expansion of
daycare programme coverage, with some improvements to increase the energy
density and reduce deficits in the nutrient supply of meals, provision of nutrition
education for the caregivers, and treatment of parasitic infections.

iv
Preface

This cross-sectional study was a collaborative project between the


Department of Human Nutrition, University of Otago, Dunedin, New Zealand and
the Fima Lifshitz Research Unit of the Hospital Universitário Professor Edgard
Santos, associated with the Federal University of Bahia, Salvador, Brazil. The
initial study was a randomized controlled trial (RCT) entitled: “Fortifying daycare
meals to combat anaemia and micronutrient deficiencies in preschool children
from Salvador, NE Brazil”. The candidate spent more than a year preparing for the
RCT, after gaining permission for the trial from the Human Ethics Committees of
the collaborating universities in Brazil and New Zealand. However, few children
were anaemic at baseline, despite the high prevalence published earlier. Thus, the
planned RCT was abandoned and instead, the candidate carried out a cross-
sectional study. The candidate developed the concept and overall design for the
cross-sectional study, assisted by her primary supervisor, Professor RS Gibson.
The candidate together with Professors RS Gibson and Hugo C. Costa-Ribeiro
were responsible for securing the funding.
The candidate, with some assistance from Professor Gibson, was
responsible for writing the Human Ethics Application for the University of Otago,
developing the standard operating procedures for the anthropometry, collection of
fasting blood samples, and faecal specimens, and for compiling all the
questionnaires. The candidate performed the anthropometric measurements on the
preschoolers, and was responsible for supervising the collection of all the data in
Brazil. The candidate assisted with processing of blood and faecal samples, and
was also responsible for data entry and cleaning, compilation of a menu-specific
food composition table for Brazil, and the statistical analysis (STATA) and
interpretation of the data, the latter in consultation with Mrs. Sheila Williams from
the Department of Preventive and Social Medicine (University of Otago).
Dr. Hugo C. Costa-Ribeiro and Dr. Angela P. Mattos (Fima Lifshitz
Research Unit, Hospital Universitário Professor Edgard Santos) were responsible

v
for recruitment of the participating daycares. Ms. Danile L. Barreto (Fima Lifshitz
Research Unit, Hospital Universitário Professor Edgard Santos) was responsible
for the final translation of the questionnaires into Portuguese and provided
oversight of the team of Brazilian nutritionists. The latter administered the pre-
tested questionnaires, conducted the maternal anthropometry, and assisted with
anthropometric data collection.
Mr. Alastair Lander (Research Assistant, Department of Human Nutrition,
University of Otago) was primarily responsible for the labelling, processing and
storage of blood and faecal samples in Brazil, and assisted with data entry. Mr.
Lander also conducted the microscopy analysis and ELISA assays of the faecal
samples for intestinal parasitic infections.
Dr Karl Bailey (Department of Human Nutrition, University of Otago)
conducted the serum ferritin, transferrin receptor, retinol, zinc, selenium, vitamin
B-12, C-reactive protein, and α-1-glycoprotein assays, and assisted with the
compilation of the food composition table for Brazil. Serum and erythrocyte folate
assays were performed by Mrs. Holiday Wilson (Department of Human Nutrition,
University of Otago).
Mr. Abdulmonem Alsaleh, with advice from Professor Ian Morison
(Department of Pathology, University of Otago), conducted the
haemoglobinopathy and thalassemia assessment of the blood samples.
Dr. Ricardo David, Director of Laboratory Services (Hospital Universitário
Professor Edgard Santos), provided oversight of the laboratory technicians
involved in the blood sampling and the haematology laboratory responsible for
the CBC analysis.

vi
Acknowledgements

Without the support of many people, this project would never have become
a reality. Although it is not possible to acknowledge everyone, I would like to
express my gratitude to some of the people involved in the success of this project.
Professor Rosalind S Gibson, my primary supervisor, for your wise
guidance, invaluable mentoring, and faithful encouragement. The opportunity to
learn from you has truly been one of the greatest gifts in my life, and I would not
be where I am without your help.
Mrs. Sheila Williams and Dr. Lisa Houghton, for the generous support and
beneficial suggestions you have both readily given as supervisors. I have greatly
appreciated the contribution of your expertise. Dr. Anne-Louise Heath, convenor
of my PhD committee, for your encouragement and kindness.
Dr. Hugo C. Costa-Ribeiro and Dr. Angela P. Mattos, our Brazilian co-
collaborators, for your support of the field work and willingness to allow many of
the nutritionists from the Fima Lifshitz Research Unit to be involved in this
project. Ms. Danile L. Barreto, for your assistance in leading the team of
nutritionists and the many ways you supported this project. Mr. José R.S.
Lorenzo, for faithfully providing much needed administrative assistance. The
wonderful team of Brazilian nutritionists – Ms. Lissandra A. Santos, Ms. Maria
Luiza A. de Sena, Ms. Carla R.L. Mendonãça, Ms. Ilana P.M. Soares, Ms. Camila
R. Avelar, Ms. Alline F. Lima, Ms. Caline S. Santiago, Ms. Luciana S. Araujo,
Ms. Nina S. de Morais, and Ms. Fernanda F. Sampaio - for your commitment to a
high standard of field work and the joy you brought to each day.
A special thank you to the daycare coordinators, for facilitation of this
project in the midst of busy daily programmes, and the mothers of the children,
for giving up their precious time in order to provide valuable information for us.
To the children, for being fantastic participants!

vii
Dr. Karl Bailey, for your technical expertise and gift of trouble-shooting
interesting problems, and Dr. Ian Gibson, for helping me become better friends
with STATA and for patiently answering many computing questions.
An amazing group of fellow PhD students, for the uplifting monthly group
meetings and many ways you have shown kindness to me. My good friends, both
near and far – Mandy, Jennifer, Beth, Judith, Nicola, and Lisa, for the cups of tea,
laughs, prayers, and encouragement to keep going, even when the road was hard.
Alastair, my beloved husband of 25 years, for your unwavering support,
wonderful sense of humour, gentle humility, and generous contribution of your
skills and expertise. Your commitment to walk together every step of the way has
made all the difference.

viii
This thesis is dedicated to
our daughters,
Rachael and Jenny,
who continue to be a daily source of inspiration

May your deep compassion and respect


for the marginalized and vulnerable
grow ever stronger

“Speak up for those who cannot speak for themselves,


for the rights of all who are destitute...
defend the rights of the poor and needy.”
Proverbs 31:8-9

ix
x
Table of Contents

Abstract ................................................................................................................. iii

Preface .................................................................................................................... v

Acknowledgements .............................................................................................. vii

Table of Contents ................................................................................................. xi

List of Tables ...................................................................................................... xvi

List of Figures ..................................................................................................... xix

List of Abbreviations .......................................................................................... xx

1 Introduction ................................................................................................... 1
1.1 Introduction ............................................................................................. 1
1.2 Objectives ................................................................................................ 3

2 Literature Review ......................................................................................... 5


2.1 Introduction ............................................................................................. 5
2.2 Role of poverty in the aetiology of malnutrition ..................................... 7
2.3 Role of inadequate dietary intakes in the aetiology of malnutrition ...... 9
2.3.1 Iron .................................................................................................... 11
2.3.2 Zinc ................................................................................................... 13
2.3.3 Vitamin A .......................................................................................... 15
2.3.4 Vitamin B-12 and folate .................................................................... 15
2.3.5 Selenium............................................................................................ 16
2.3.6 Interactions between micronutrients ................................................. 17
2.4 Role of host-related factors in the aetiology of malnutrition ................ 17
2.4.1 Intestinal factors ................................................................................ 17
2.4.2 Systemic factors ................................................................................ 20
2.5 Factors associated with the aetiology of anaemia ................................. 23
2.5.1 Iron .................................................................................................... 23
2.5.2 Genetic Hb disorders ......................................................................... 25
2.5.3 Vitamin A .......................................................................................... 25
2.5.4 Folate and vitamin B-12 .................................................................... 27
2.5.5 Zinc ................................................................................................... 28
2.5.6 Selenium............................................................................................ 29
2.5.7 Interactions between micronutrients ................................................. 30
2.5.8 Role of chronic inflammation ........................................................... 31
2.5.9 Interpretation of ferritin, retinol, zinc, and selenium biomarkers ..... 32
2.6 Factors associated with impaired linear growth .................................... 32

xi
2.6.1 Protein................................................................................................ 33
2.6.2 Calcium and vitamin D ...................................................................... 33
2.6.3 Zinc .................................................................................................... 36
2.6.4 Iodine and selenium ........................................................................... 36
2.6.5 Vitamin A .......................................................................................... 37
2.6.6 Parasitic infections ............................................................................. 38
2.6.7 Genetic Hb disorders ......................................................................... 38
2.7 Factors associated with disturbances in body composition ................... 39
2.7.1 Overweight ........................................................................................ 39
2.7.2 Underweight ...................................................................................... 40
2.8 Micronutrient deficiencies associated with impaired immune .................
function .................................................................................................. 41
2.8.1 Zinc .................................................................................................... 41
2.8.2 Vitamin A .......................................................................................... 43
2.8.3 Iron .................................................................................................... 43
2.8.4 Selenium ............................................................................................ 44
2.9 Pro-poor policies in Brazil for the prevention of malnutrition during
childhood ............................................................................................... 45
2.9.1 Wheat and corn flour fortification with iron and folic acid ............... 46
2.9.2 Childhood immunizations and vitamin A supplementation .............. 46
2.9.3 Sewage and sanitation programmes, specifically in Salvador, NE
Brazil ................................................................................................. 47
2.9.4 Conditional cash transfer programme................................................ 48
2.9.5 Maternal education and preschool daycare centres ........................... 49
2.10 Conclusion ............................................................................................. 50

3 Socio-economic and anthropometric status of preschool children


attending peri-urban and urban daycares in Salvador, NE Brazil ......... 51
3.1 Introduction ........................................................................................... 51
3.2 Participants and Methods ...................................................................... 53
3.2.1 Study sites and participants ............................................................... 53
3.2.2 Assessment of socio-demographic and health status......................... 57
3.2.3 Assessment of breastfeeding, infant and young child caring and
feeding practices ................................................................................ 59
3.2.4 Anthropometric assessment ............................................................... 61
3.2.5 Statistical analysis ............................................................................. 63
3.3 Results ................................................................................................... 63
3.3.1 Socio-demographic and health characteristics .................................. 63
3.3.2 Infant and young child caring and feeding practices and current
dietary patterns .................................................................................. 67
3.3.3 Anthropometry .................................................................................. 67
3.3.4 Inter-relationships among variables and household poverty ............. 69
3.4 Discussion.............................................................................................. 70
3.4.1 Prevalence of undernutrition ............................................................. 70
3.4.2 Association between undernutrition and extreme poverty ................ 71
3.4.3 Association between overnutrition and less extreme poverty ........... 72

xii
3.4.4 Association between anthropometry and retrospective child
feeding practices................................................................................ 73
3.4.5 Strengths and limitations ................................................................... 73
3.4.6 Conclusion ........................................................................................ 74

4 The adequacy of daycare meals provided to preschoolers in


Salvador, NE Brazil .................................................................................... 75
4.1 Introduction ........................................................................................... 75
4.2 Participants and Methods ...................................................................... 76
4.2.1 Study sites and participants ............................................................... 76
4.2.2 Comparison of the 20 daycare menus with food-based dietary
guidelines in Brazil ........................................................................... 76
4.2.3 Assessment of the adequacy of the energy and nutrient supply
from the daycare menus ................................................................... 77
4.2.4 Statistical analysis ............................................................................. 79
4.3 Results ................................................................................................... 81
4.3.1 Assessment of compliance of 20 daycare menus to dietary
guidelines .......................................................................................... 81
4.3.2 Assessment of the adequacy of the energy and nutrient supply
from the daycare menus .................................................................... 81
4.3.3 Assessment of the major food sources of energy and selected
nutrients ............................................................................................. 83
4.3.4 Strategies to overcome deficits in the nutrient supply from the
daycare menus ................................................................................... 83
4.4 Discussion ............................................................................................. 85
4.4.1 Adequacy of the energy and nutrient supply of the menus ............... 85
4.4.2 Strategies to overcome observed deficits in nutrient supply from
daycare menus ................................................................................... 89
4.4.3 Strengths and limitations ................................................................... 89
4.4.4 Conclusion ........................................................................................ 90

5 Risk factors associated with intestinal parasite infestation of


preschool children attending daycares in Salvador, NE Brazil .............. 91
5.1 Introduction ........................................................................................... 91
5.2 Participants and Methods ...................................................................... 92
5.2.1 Study sites and participants ............................................................... 92
5.2.2 Collection and microscopic assessment of parasites in stool
samples .............................................................................................. 93
5.2.3 ELISA assays for qualitative detection of Giardia and
Cryptosporidium antigens in faeces .................................................. 95
5.2.4 Assessment of socio-economic status ............................................... 97
5.2.5 Statistical analysis ............................................................................. 97
5.3 Results ................................................................................................... 99
5.3.1 Comparison of characteristics between participants providing
and not providing a stool sample ...................................................... 99
5.3.2 Prevalence of parasitic infections ..................................................... 99

xiii
5.3.3 Risk factors associated with helminth and Giardia infections ........ 102
5.4 Discussion............................................................................................ 102
5.4.1 Risk factors associated with helminth infections ............................ 102
5.4.2 Risk factors associated with Giardia intestinalis ............................ 104
5.4.3 Strengths and limitations ................................................................. 105
5.4.4 Conclusion ....................................................................................... 106

6 Micronutrient status of disadvantaged preschoolers attending


daycares in Salvador, NE Brazil .............................................................. 107
6.1 Introduction ......................................................................................... 107
6.2 Participants and Methods .................................................................... 109
6.2.1 Study sites and participants ............................................................. 109
6.2.2 Assessment of socio-demographic, health, and morbidity status .... 109
6.2.3 Assessment of parasite status .......................................................... 111
6.2.4 Biochemical assessment .................................................................. 111
6.2.5 Statistical analysis ........................................................................... 115
6.3 Results ................................................................................................. 115
6.3.1 Socio-demographic / health characteristics and prevalence of
parasitic infections ........................................................................... 115
6.3.2 Assessment of biomarkers for micronutrients and inflammation .... 117
6.3.3 Predictors of micronutrient biomarkers ........................................... 117
6.4 Discussion............................................................................................ 119
6.4.1 Vitamin A, zinc, and selenium status .............................................. 119
6.4.2 Vitamin B-12 and folate status ........................................................ 126
6.4.3 Strengths and limitations ................................................................. 127
6.4.4 Conclusion ....................................................................................... 128

7 Predictors of haemoglobin and iron status in disadvantaged


preschoolers attending daycares in Salvador, NE Brazil....................... 129
7.1 Introduction ......................................................................................... 129
7.2 Participants and Methods .................................................................... 131
7.2.1 Study sites and participants ............................................................. 131
7.2.2 Assessment of socio-demographic and health status....................... 133
7.2.3 Anthropometric assessment ............................................................. 133
7.2.4 Assessment of parasite status .......................................................... 133
7.2.5 Biochemical assessment .................................................................. 135
7.2.6 Statistical analysis ........................................................................... 139
7.3 Results ................................................................................................. 139
7.3.1 Socio-demographic and health characteristics ................................ 139
7.3.2 Assessment of genetic Hb disorders, haematological and ....................
biochemical iron indices .................................................................. 141
7.3.3 Predictors of Hb and iron biomarker concentrations ....................... 143
7.4 Discussion............................................................................................ 143
7.4.1 Prevalence of anaemia, iron deficiency anaemia and iron ...................
deficiency......................................................................................... 145
7.4.2 Genetic Hb disorders as predictors of Hb ........................................ 145

xiv
7.4.3 Micronutrient predictors of Hb ....................................................... 148
7.4.4 Predictors of iron status ................................................................... 149
7.4.5 Strengths and limitations ................................................................. 151
7.4.6 Conclusion ...................................................................................... 152

8 Factors associated with growth in disadvantaged preschoolers


attending daycare in Salvador, NE Brazil using structural equation
modelling .................................................................................................... 153
8.1 Introduction ......................................................................................... 153
8.2 Participants and Methods .................................................................... 155
8.2.1 Study sites and participants ............................................................. 155
8.2.2 Assessment of socio-demographic and health status ...................... 155
8.2.3 Assessment of child and maternal anthropometry .......................... 157
8.2.4 Assessment of parasite status, Hb and micronutrient biomarkers... 157
8.2.5 Statistical analysis ........................................................................... 157
8.3 Results ................................................................................................. 159
8.3.1 Socio-demographic characteristics of the children ......................... 159
8.3.2 Anthropometric assessment of the children and mothers ............... 161
8.3.3 Comparison between SEM and overall study data ......................... 161
8.3.4 Correlation matrix of variables included in the SEM ..................... 161
8.3.5 Direct and indirect effects on HAZ and BMIZ ............................... 163
8.4 Discussion ........................................................................................... 165
8.4.1 Role of maternal and biological factors on HAZ scores ................. 165
8.4.2 Role of environmental factors and micronutrient status on HAZ
scores ............................................................................................... 168
8.4.3 Role of maternal factors and SES on BMIZ scores ........................ 169
8.4.4 Role of Hb on BMIZ scores ............................................................ 171
8.4.5 Strengths and limitations ................................................................. 171
8.4.6 Conclusion ...................................................................................... 172

9 Summary, Conclusions and Recommendations ..................................... 175


9.1 Summary and Conclusions.................................................................. 175
9.2 Recommendations ............................................................................... 178

References .......................................................................................................... 181

Appendices ......................................................................................................... 233


Appendix A Socio-demographic questionnaire ....................................... 235
Appendix B Socio-economic scoring for household characteristics or
possessions .......................................................................... 244
Appendix C Differences by daycare group for specific feeding and
health characteristics of the participants ............................ 246
Appendix D Instruction for stool collection ............................................ 247

xv
List of Tables

Table 2.1 Prevalence of stunting, wasting, and overweight in children one to 8


four years of age between 1975 and 1996 for Northeast and
Southeast Brazil

Table 2.2 Dietary modifiers influencing iron and zinc absorption 10

Table 2.3 Prevalence of parasitic infections in children living in urban areas 18


in NE Brazil

Table 2.4 Prevalence of anaemia and iron deficiency in cross-sectional 22


studies of children in NE Brazil

Table 2.5 Prevalence of vitamin A deficiency and marginal vitamin A status 26


in cross-sectional studies of children in NE Brazil

Table 2.6 Prevalence of underweight, stunting, wasting and overweight in 34


cross-sectional studies of children in NE Brazil

Table 3.1 Response rate of daycares 56

Table 3.2 Socio-demographic characteristics of the children and households 58

Table 3.3 Health characteristics of the participants 60

Table 3.4 Prevalence of past feeding practices of children and current 62


caregiver feeding practices

Table 3.5 Mean (SD) z-scores and anthropometric measures by sex for the 64
children

Table 3.6 Differences for age, sex and socioeconomic status for the anthro- 66
pometric measures of the children shown by ß-coefficient (95% CI)

Table 3.7 Differences between socioeconomic status for the anthropometric 68


measures of the mothers shown by ß-coefficient (95% CI)

Table 4.1 Comparison of the frequency of compliance (as %) by daycare 78


menus to the dietary guidelines “Ten steps to healthy eating for
children over two years of age” for Brazil

Table 4.2 Macronutrients (as percentage of energy) supplied by Brazilian 80


daycare menus compared to the population target goals

xvi
Table 4.3 Supply of energy from Brazilian daycare menus compared to 82
estimated energy requirements

Table 4.4 Average daily supply of nutrients for a child from daycare menus 84
(n=20) and factors influencing bioavailability of iron and zinc
compared to the estimated average requirements and
recommended nutrient intakes

Table 4.5 Mean percentages contribution of eight food groups to energy and 86
nutrients supply from daycare menus

Table 5.1 Mean (SD) differences for specified variables between the 94
participants who provided a stool sample (n=325) and those who
did not provide a stool sample (n=51)

Table 5.2 Prevalence (% and 95% CI) of intestinal parasite infection 96

Table 5.3 Frequency (%) of intestinal parasite infection, deworming 98


treatment and vitamin A supplementation by daycare group

Table 5.4 Logistic regression models of risk factors (by odds ratio and 95% 100
CI) for infection with helminths and Giardia intestinalis

Table 6.1 Socio-economic and health characteristics of the participants and 110
prevalence of intestinal parasitic infections

Table 6.2 Quality control data for the assessment of accuracy and precision 112
of the biomarkers

Table 6.3 Mean (SD) values of micronutrient biomarkers, including adjusted 114
mean values for those impacted by infection

Table 6.4 Prevalence (% and 95% CI) of low micronutrient biomarkers and 116
elevated biomarkers of inflammation

Table 6.5 Predictors of serum retinol, zinc, and selenium based on multiple 118
regression models using age, sex, other micronutrients, chronic
inflammation and parasite status, as shown by ß-coefficient (95%
CI)

Table 6.6 Predictors of serum vitamin B-12, and serum and erythrocyte 120
folate based on multiple regression models using age, sex, other
micronutrients, chronic inflammation and parasite status, as shown
by ß-coefficient (95% CI)

xvii
Table 7.1 Socio-economic and health characteristics of the participants and 132
prevalence of intestinal parasitic infections

Table 7.2 Quality control data for the assessment of accuracy and precision 134
of the iron biomarkers

Table 7.3 Prevalence (% and 95% CI) of haemoglobinopathies and α3.7 136
thalassemia in the preschoolers

Table 7.4 Prevalence of haemoglobinopathies and α3.7 thalassemia in the 136


preschoolers by ethnicity

Table 7.5 Mean (SD) values for haematology and iron biomarkers for 138
normal Hb type (Hb AA) and the four major abnormal Hb variants
for the preschoolers

Table 7.6 Prevalence (% and 95% CI) of selected haematological variables, 140
iron biomarkers, and elevated biomarkers of inflammation

Table 7.7 Predictors of Hb based on multiple regression models using 142


abnormal Hb variants and micronutrient status, as shown by ß-
coefficient (95% CI)

Table 7.8 Predictors of serum ferritin and soluble TfR based on multiple 144
regression models using abnormal Hb variants, chronic
inflammation, parasite status and descriptive variables, as shown
by ß-coefficient (95% CI)

Table 8.1 Socio-demographic characteristics and prevalence of anaemia, low 156


zinc and selenium status of the preschool children

Table 8.2 Anthropometric variables for the preschoolers and their mothers, 158
and mother-child diads

Table 8.3 Comparison between variables for the complete data set included 160
in the structural equation model (SEM) (n=264) and the excluded
incomplete set

Table 8.4 Correlation matrix of biological, social and environmental factors 162
affecting the children’s BMIZ and HAZ

Table 8.5 Direct and indirect effects of SEM variables on HAZ and BMIZ 166

xviii
List of Figures

Figure 2.1 Model of the causes of childhood malnutrition and its short- 6
and long-term consequences

Figure 3.1 Study site location 52

Figure 3.2 Cross-sectional study profile 54

Figure 6.1 Comparison of serum retinol concentrations for male children 122
between Brazilian daycare study (n=188) and US NHANES
III (1988-1994) participants ages 4-8 years

Figure 6.2 Comparison of serum retinol concentrations for female 122


children between Brazilian daycare study (n=170) and US
NHANES III (1988-1994) participants ages 4-8 years

Figure 6.3 Comparison of serum zinc concentrations between Brazilian 124


daycare study (n=358) and US NHANES II (1976-1980)
participants ages 3-8 years

Figure 7.1 Comparison of serum ferritin concentrations between the 146


Brazilian daycare study (n=358) and US NHANES III (1988-
1994) participants ages 3-5 years

Figure 8.1 Structural equation model for growth with outcomes HAZ and 164
BMIZ

xix
List of Abbreviations

AAS Atomic absorption spectrophotometry


ACI Anaemia of chronic inflammation
AFA Mid-upper arm fat area
AGP α-1-glycoprotein
AMA Mid-upper arm muscle area
BMI Body-mass-index
BMIZ Body-mass-index z-score
CBC Complete blood count
CCT Conditional cash transfer
CDC Center for Disease Control and Prevention (US)
CMI Centro de Midia Independente
CRP C-reactive protein
DFE Dietary folate equivalents
EAR Estimated average requirements
EER Estimated energy requirement
FAO Food and Agriculture Organization of the United Nation
FCT Food composition table
Fpn Ferroportin
HAZ Height-for-age z-score
Hb Haemoglobin
IAEA International Atomic Energy Association
IBGE Instituto Brasileiro de Geografia e Estatística
IGF Insulin-like growth factor
INCAP Institute of Nutrition of Central America and Panama
IQR Interquartile range
IVACG International Vitamin A Consultative Group
IYCF Infant and young child feeding
IZiNC International Zinc Consultative Group

xx
MCH Mean cell haemoglobin
MCV Mean cell volume
MUAC Mid-upper arm circumference
NCSDH National Commission on the Social Determinants of Health
(Brazil)
NE Northeast
NEPA Núcleo de Estudos e Pesquisas em Alimentação
NHANES National Health and Nutrition Examination Survey (US)
PAHO Pan American Health Organization
PUFA Polyunsaturated fatty acids
RCT Randomized controlled trial
RBP Retinol binding protein
RDA Recommended daily allowance
RDW Red cell distribution width
RMSR Root mean squared residual
RNI Recommended nutrient intakes
SAF Sodium acetate-acetic acid formalin
SD Standard deviation
SEM Structural equation model
SES Socio-economic status
TE Trace element
TfR Transferrin receptor
UNICAMP Universidade Estadual de Campinas
UNICEF United Nations Children’s Fund
UNU United Nations University
USDA United States Department of Agriculture
WAZ Weight-for-age z-score
WHO World Health Organization
WHZ Weight-for-height z-score

xxi
xxii
1 Introduction

1.1 Introduction
Today, many young children living in urban poverty have a heightened risk
of malnutrition from the lack of access to the basic requirements of life, including
adequate food, clean water, and proper housing (UNICEF 2012). Brazilian
children are no exception; nearly 85% of the population live in urban areas, with
notable health and social inequities between regions (Paim et al. 2011).
Northeastern (NE) Brazil is the poorest region in the country, where young
children living in low-income urban slums are likely to experience increased risk
of morbidity, micronutrient deficiencies, and the double burden of malnutrition
(i.e., risk of under- and overnutrition). Such deleterious outcomes can have long
lasting effects on children, including failure to fulfill both their cognitive and
developmental potential (Grantham-McGregor et al. 2007) as well as the early
onset of non-communicable diseases, such as diabetes and hypertension (Stein et
al. 2005).
The deeply entrenched poverty in the NE region has negatively affected the
nutritional status of children under-five years of age for many decades. Earlier
data suggested this impoverished province had the highest prevalence of
childhood wasting and stunting in Brazil (Monteiro et al. 1992), yet more
recently, the prevalence of overweight in preschoolers from low-income families
living here has rapidly risen (Monteiro et al. 2002). Further, the high prevalence
of anaemia (i.e., ~40%) has also been of great concern in young children in NE
Brazil in recent years (Jordão et al. 2009), a condition assumed to be caused by
iron deficiency. However, little else is known regarding the prevalence of
micronutrient deficiencies of preschoolers living in this region, with the exception
of vitamin A deficiency, which was declared a public health problem in children
under-five years of age more than thirty years ago (WHO 1982).

1
Numerous factors may contribute to malnutrition in preschoolers living in
poor environments in NE Brazil. The amount of available food may be
insufficient and / or lacking in quality for the optimal growth and development of
young children. Traditionally, the staple diets of food-insecure households in
Brazil consist of cereals and legumes, and are devoid of more expensive
micronutrient-dense animal-source foods. Moreover, processed foods that are high
in saturated fats and refined sugars are now widely available and very
inexpensive. Consequently, disadvantaged children are particularly at risk of
energy-rich micronutrient-poor diets (Antunes et al. 2010). Intakes of animal
protein, and key micronutrients, such as iron, zinc, calcium, and vitamin A are
likely to be limited, and their bioavailability compromised by dietary factors such
as phytate, a potent inhibitor of iron and zinc, and to a lesser extent, calcium
absorption
Host-related factors such as intestinal parasitic infections may also limit the
bioavailability of micronutrients among urban slum children in this under-
resourced NE region as traditionally, basic sanitation and hygiene services have
been severely limited (NCSDH 2008; de Lima et al. 2010). Thus, these
disadvantaged children are likely to be at risk of infections with soil-transmitted
helminths and pathogenic protozoal infections (e.g. Giardia intestinalis) (Barreto
et al. 2010). Such infections can cause diarrhoea, anorexia, and damage to the
mucosal wall of the small intestine, leading to reduced absorption of macro- and
micronutrients (Müller & von Allmen 2005; Hall et al. 2008), which may result in
micronutrient deficiencies, as well as impairments in growth and immune
function.
The ethnicity of children living in certain cities in NE Brazil, such as
Salvador, also has the potential to adversely affect their micronutrient status and
growth. For example, children of African-descent have a higher risk of genetic
haemoglobin (Hb) disorders, such as sickle cell disease and thalassemias (Bain
2006), as reported in earlier studies in Salvador (Azevêdo et al 1980; Adorno et al.
2005). Indeed, NE Brazil has the highest percentage of black and brown

2
Brazilians in the country, according to the recent national census (IBGE 2010a),
and Salvador is reportedly the largest African-populated city outside of Africa.
Since the 1990s, Brazil has implemented several key pro-poor policies, such
as mandatory fortification of wheat and corn flour with iron and folic acid, and
supplementation with vitamin A and iron in an effort to improve the nutritional
status of children under-five years of age living in low-income environments.
Sewage and sanitation programmes have also been implemented to decrease the
transmission of parasites in crowded urban areas, such as in Salvador (Barreto et
al. 2007). Provision of universal primary schooling has been an additional focus,
with the enactment of an early childhood education law nearly twenty years ago,
promoting the care and education of Brazilian children less than six years of age
by either public or philanthropic daycare facilities (Brazil Ministry of Health
1996a). High quality early child development programmes are considered to be a
critical step towards reducing the effects of poverty, and enabling disadvantaged
children to achieve both their physical and cognitive developmental potential,
particularly as the programmes provide not only educational stimulation but also
daily meals (Brotanek et al. 2007; Engle et al. 2007). Nonetheless, there have been
few comprehensive studies of the growth, health, and micronutrient status of
preschool children living in low-income environments attending daycare in NE
Brazil.

1.2 Objectives
The overall goal of this research was to examine the nutritional status of a
group of disadvantaged preschoolers aged three to six years attending seven pre-
selected philanthropic urban and peri-urban daycares in Salvador, NE Brazil. This
was achieved by conducting a cross-sectional survey designed to accomplish the
following specific objectives:

3
1. Assessment of the socio-economic status, health characteristics, early
infant and young child feeding practices, and growth and body
composition of the preschoolers and the associations between these
variables (chapter 3).

2. Investigation of the nutrient adequacy of the daycare menus and their


compliance to national food-based dietary guidelines for children over
two years of age (chapter 4).

3. Determination of the prevalence and risk factors associated with


intestinal parasites in the preschoolers (chapter 5).

4. Assessment of the prevalence of deficiencies of vitamin A, zinc,


selenium, vitamin B-12 and folate, and the predictors and underlying
conditions associated with the concentrations of these micronutrient
biomarkers (chapter 6).

5. Assessment of the prevalence of selected genetic Hb disorders, anaemia


and iron deficiency, and the major predictors of Hb and iron biomarkers
(chapter 7).

6. Application of structural equation modelling to examine the direct and


indirect effects of maternal, socio-demographic, nutritional, and
environmental factors on linear and ponderal growth of these
disadvantaged preschoolers (chapter 8).

4
2 Literature Review

2.1 Introduction
Recent analyses from low-income and middle-income countries have
estimated that undernutrition is responsible for 35% of the deaths in children
under-five years of age because of growth failure and micronutrient deficiencies,
particularly of iron, zinc, and vitamin A (Black et al. 2008). However, changing
dietary patterns in some regions, such as Latin America, have resulted in a shift
from undernutrition to overnutrition (i.e., nutrition transition) in preschool-aged
children (de Onis & Blössner 2000; Popkin 2002), with long-term increases in
morbidity outcomes associated with childhood overweight (Baird et al. 2005). The
highest rates of overweight often occur amongst populations with the highest
poverty rates and least education (Drewnowski & Specter 2004). Thus,
malnutrition, including both under- and overnutrition – often termed the double
burden of disease –, may cause deleterious health consequences for young
children living in resource-poor settings (Khor 2008).
Children living in Brazil, the wealthiest country in Latin America with the
eighth largest economy world-wide, do not appear to be at risk of malnutrition
based on national data. These reports over the past three decades indicate a
significant reduction in undernutrition without any concomitant rise in
overnutrition in children less than five years of age (Brazil Ministry of Health
1992; Brazil Ministry of Health 1996b; Brazil Ministry of Health 2007a).
Unfortunately, these national estimates fail to reflect the vastly different rates of
malnutrition which exist regionally between the wealthy Southeast and the very
poor Northeast (Monteiro et al. 1992).
Hence, this literature review will focus on the possible roles of poverty,
inadequate dietary intakes, and host-related factors in the aetiology of malnutrition
among disadvantaged preschoolers living in NE Brazil. Details of the potential
adverse health consequences of malnutrition will also be included, as well as some

5
Figure 2.1 Model of the causes of childhood malnutrition and its short- and
long-term consequences1

Long-term consequences:
Short-term consequences: adult size, cognitive ability, metabolic
morbidity, disability, mortality and cardiovascular disease

Child Malnutrition:
undernutrition / overnutrition,
micronutrient deficiencies,
impaired growth

Inadequate dietary intake: Disease:


energy and micronutrients e.g. parasitic
infections

Household food Inadequate Unhealthy household


insecurity care environment, lack of
health services

Lack of household resources, education,


maternal malnutrition, employment

Lack of community resources: financial,


human, physical, social and natural

Social, economic and political context

1
Modified and reprinted from Black et al. 2008, with permission from Elsevier
Limited (license number: 2953450728852)

6
of the pro-poor policies implemented by the Brazilian government in recent years
to prevent malnutrition in young children.

2.2 Role of poverty in the aetiology of malnutrition


Child malnutrition has multiple causes within society (Smith & Haddad
2000). The underlying fundamental contributor to malnutrition is the unequal
allocation of resources to individuals, communities, and regions as a result of
political and economic structures, national and international policies, and cultural
and religious systems, highlighted by the United Nations Children’s Fund
(UNICEF) conceptual framework, as modified by Black et al. (2008) (Figure 2.1).
Brazil is no exception. Despite extraordinary economic growth over the past four
decades and associated reductions in national poverty rates, deeply entrenched
social and income inequities are evident between rich and poor individuals and
regions (NCSDH 2008). For example, between 1970 and 1988, the prevalence of
those living below the poverty line in the wealthy Southeast region decreased
from 30% to less than 20% (Monteiro et al. 1992). In contrast, in the Northeast,
two-thirds of the population lived in poverty in 1970, and yet by 1988, still more
than half of the population lived below the poverty line in this under-resourced
area of Brazil.
Notable differences in housing, basic sanitation and sewage facilities, access
to health services, maternal education levels, and dietary patterns exist between
the richer Southeast and poorer Northeast regions (Victora et al. 2000; NCSDH
2008). Such disparities have negative health consequences for disadvantaged
young children living in the NE region. For example, infant mortality rates in this
impoverished province have consistently been more than twice as high as the
Southeast region for the past thirty years (Barros et al. 2010). Furthermore, even
in 1996, the prevalence of childhood wasting and stunting was markedly higher in
the Northeast (i.e., more than three-times and ten-times, respectively), compared
to the Southeast, despite the reported improvements between 1975 and 1996 in

7
Table 2.1 Prevalence of stunting, wasting, and overweight in children one to
four years of age between 1975 and 1996 for Northeast and Southeast Brazil1,2

Stunting Wasting Overweight


Region and Year (HAZ < -2SD) (WHZ < -2SD) (WHZ > 2SD)
(%) (%) (%)
Northeast (poorest)
1975 60.8 5.0 2.4
1989 42.4 2.6 1.2
1996 28.8 3.4 2.3
Southeast (richest)
1975 6.0 2.5 6.8
1989 3.4 0.9 10.3
1996 3.0 1.2 8.6

1
Monteiro et al. 1992
2
Monteiro et al. 2002
Abbreviations used: HAZ, Height-for-age z-score; WHZ, Weight-for-height z-
score

8
the nutritional status of children under-five years of age in both regions (Table
2.1).
Of concern is the doubling of the prevalence of overweight in preschoolers
from low-income families living in the NE between 1989 and 1996 (Monteiro et
al. 2002). This trend is not unexpected, as obesity rates nearly tripled (i.e., from
5% to 13%) among the poorest women in Brazil between 1975 and 1997,
although they remained unchanged (i.e., 10–11%) among the wealthiest women
(Monteiro et al. 2004). Increasing obesity in Brazilian women has been linked to
household food insecurity because of the easy access to inexpensive, energy dense
foods (Velasquez-Melendez et al. 2011). Thus, young children living in poor
households in NE Brazil have the potential to be under-nourished because of lack
of available food, or over-nourished as a result of access to poor quality foods
(Assis et al. 2007). These same trends have been reported among children living
in similar environments elsewhere (Adish et al. 1999; Thane et al. 2000;
Villalpando et al. 2003a and 2003b).

2.3 Role of inadequate dietary intakes in the aetiology of


malnutrition
For optimal growth and development in preschool children, the foods
consumed must be sufficient in amount and quality to meet the high energy and
micronutrient requirements of the growing child (PAHO/INCAP 1999). However,
the diets of young children living in low-income settings such as the favelas
(slums) of NE Brazil are traditionally based on cereals and legumes. Consumption
of expensive animal-source foods, particularly meat, poultry, fish or eggs,
together with dark green and yellow fruits and vegetables is low, making it a
complex challenge to meet the energy and nutrient requirements of these young
children (Antunes et al. 2010).
The quality of the overall diet depends on both the energy (i.e., kcal/g) and
micronutrient density (i.e., micronutrient per MJ), as well as the bioavailability of
nutrients. The term ‘bioavailability’ is defined as the proportion of the total

9
Table 2.2 Dietary modifiers influencing iron and zinc absorption1

Non-haem iron Zinc

Enhancers Enhancers
Meat, poultry, fish and other seafood Meat, poultry, fish and other seafood,
eggs, whey protein
Ascorbic acid —
Other organic acids Organic acids
(citric, lactic, malic, tartaric) (citric, lactic, malic, tartaric)
Inhibitors Inhibitors
Phytate Phytate
Polyphenols (e.g., tannin) —
Calcium2 —
Certain processed soy products Certain processed soy products

1
Modified from Gibson & Ferguson (2008) and reproduced with permission from
HarvestPlus www.harvestplus.org. This technical monograph can be found at this
web address: http://www.harvestplus.org/content/interactive-24-hour-recall-
assessing-adequacy-iron-and-zinc-intakes-developing-countries.
2
Suggested from short-term supplementation studies only (Minihane &
Fairweather-Tait 1998)

10
micronutrient in a food or diet that is absorbed and utilized for normal
physiological functions (Gibson et al. 2000a), and is affected by both diet and
host-related factors (e.g., age, ethnicity, nutritional status) (Hunt 2005; Gibson
2007). For many micronutrients, the dietary factors influencing bioavailability are
not well established, with the notable exception of iron and zinc. For these
micronutrients, mathematical models have been developed to predict their
bioavailability, based on the enhancing and inhibiting factors in the whole diet
that are known to modify bioavailability (Gibson 2005).

2.3.1 Iron

Two forms of iron are found in food, haem iron and non-haem iron. Haem
iron is derived mainly from Hb and myoglobin in meat, poultry or fish, and is
much more readily absorbed than non-haem iron. Further, absorption of haem iron
is not affected by dietary components, with the exception of calcium (Hunt 2005).
Non-haem iron is present in a wide variety of both animal and plant-based foods,
and is a mandatory fortificant in all wheat and maize-flour based products in
Brazil (Brazil Ministry of Health 2002). Furthermore, unlike haem iron,
absorption of non-haem iron is affected by the iron content of the meal as well as
the simultaneous ingestion of several dietary components known to modify non-
haem iron absorption (Table 2.2). For example, the inclusion of cellular animal-
source foods such as meat, poultry and fish, vitamin C, and other organic acids
(citric, lactic, malic, and tartaric acid) produced during fermentation enhance non-
haem iron absorption, particularly when strong inhibitors of non-haem absorption,
such as phytic acid and polyphenols, are also present in the diet (Lynch 1997;
Gibson & Ferguson 2008).
Of the non-haem iron enhancers, vitamin C is the most potent because of its
ability to reduce ferric to ferrous iron via the action of ferrireductase duodenal
cytochrome b in the intestine, resulting in increased iron solubility (Teucher et al.
2004). Its enhancing effect is most apparent when consumption of meat is low and
iron status is suboptimal (Gibson & Ferguson 2008). Cellular animal-source foods
are also important enhancers of non-haem iron. Their precise mechanism is

11
uncertain. Perhaps partially digested peptides from muscle proteins bind iron via
their cysteine and histidine residues to form soluble complexes that are available
for absorption. Other non-protein components in muscle tissue may also enhance
iron absorption (Hurrell et al. 2006). Unfortunately, consumption of expensive
cellular animal products is unlikely in poor households in NE Brazil, although
vitamin-C rich fruits (e.g., oranges and acerola) are less expensive and thus more
likely to be consumed (Salles-Costa et al. 2010).
Diets in impoverished households in Brazil are commonly based on
unrefined cereals (e.g., corn, maize, and wheat) and beans (Salles-Costa et al.
2010). Hence, they contain large amounts of phytic acid (myo-inositol
hexaphosphate), often present as magnesium, calcium, and potassium phytate
(Lönnerdal 2000). Phytic acid is a potent inhibitor of non-haem iron absorption. It
chelates iron (and zinc and calcium), forming insoluble complexes in the
gastrointestinal tract that cannot be digested or absorbed in humans because of the
absence of intestinal phytase enzymes (Iqbal et al. 1994).
Absorption of non-haem iron is also inhibited by polyphenols found in tea
and coffee as well as in certain cereals (red sorghum) (Zijp et al. 2000; Ruel
2001), attributed to the formation of insoluble iron-phenolic compounds (Gibson
& Ferguson 2008). Consumption of coffee is common among preschoolers in
Brazil, particularly among the poor (Antunes et al. 2010), and sub-optimal iron
status has been identified in infants and toddlers in some other Latin American
countries (e.g., Guatemala) where coffee is also introduced early in life (Dewey et
al. 1997).
Several bioavailability algorithms have been constructed to estimate the
bioavailability of iron in meals. Of the six bioavailability algorithms developed
for iron absorption (Monsen & Balintfy 1982; Tseng et al. 1997; Du et al. 2000;
Hallberg & Hulthén 2000; Reddy et al. 2000; Bhargava et al. 2001), use of the
oldest and most simple model by Monsen and Balintfy (1982) has recently been
recommended (Beard et al. 2007). This simple model estimates iron
bioavailability in diets from quantitative data on the intake of meat, fish and
poultry (g/1000 kcal) and ascorbic acid (mg/1000 kcal), preferably in individual

12
meals. However, it does not take into account the inhibitory effects of tea, coffee,
and phytates on non-haem iron absorption (Monsen & Balintfy 1982). In a study
of Filipino religious sisters who consumed iron-biofortified and unfortified rice
for nine months, bioavailable iron estimated using Monsen’s model was most
closely correlated with total body iron calculated from serum ferritin and
transferrin receptor (TfR) concentrations using the algorithm of Cook et al. (2003)
compared with the other algorithms used to calculate bioavailable iron (Beard et
al. 2007).

2.3.2 Zinc

Like iron, the absorption of zinc is affected by several enhancing and


inhibiting factors within the diet (Table 2.2). The main dietary sources of readily
available zinc are meat, liver, fish and shellfish (Gibson 2005). Zinc absorption is
primarily enhanced by dietary protein, particularly protein from cellular animal-
source foods (Gibson & Ferguson 2008). During digestion of cellular animal
protein, some of the liberated amino acids such as histidine, are strong chelators of
zinc, and thus have the potential to increase zinc solubility (Lönnerdal 2000).
Organic acids produced during fermentation also have the potential to enhance
zinc absorption through the formation of soluble ligands. Indeed, the addition of
yogurt to a plant-based meal has been shown to increase zinc absorption in
women in Mexico (Rosado et al. 2005), but unfortunately, fermented milk
products are not commonly consumed in NE Brazil.
Zinc absorption is primarily a saturable, carrier-mediated process (Miller et
al. 2007), so that solubilised zinc is more readily absorbed, even in the presence of
inhibitors such as phytic acid, calcium, and soybean protein (Lönnerdal 2000;
Gibson & Ferguson 2008). Of these inhibitors, phytic acid has the greatest
inhibitory effect. Various processing methods, such as soaking and fermentation,
have been utilized to reduce the levels of phytate in legumes and traditional
cereal-based porridges, thereby enhancing absorption of zinc, and to a lesser
extent, non-haem iron and calcium (Gibson et al. 2003; Hurrell et al. 2003). Early
animal studies suggested that the inhibitory effect of phytate on zinc absorption

13
was exacerbated by large amounts of calcium in the diet, attributed to the
formation of a highly insoluble calcium-zinc-phytate complex (Forbes et al.
1983). However, more recent studies indicate that calcium does not inhibit zinc
absorption (Lönnerdal et al. 1984) even when in the presence of a high phytate
diet (Hunt & Beiseigel 2009), despite the earlier findings (Likusi & Forbes 1965).
Finally, absorption and endogenous faecal excretion of zinc is influenced by
the total zinc content of the diet (Hambidge et al. 2010). With habitually low
intakes of dietary zinc, homeostatic mechanisms upregulate zinc absorption and
retention and the re-absorption of endogenous zinc, and vice versa with higher
zinc intakes (Hambidge 2003; Hambidge et al. 2010).
Four algorithms have been developed to estimate zinc bioavailability in
diets (Murphy et al. 1992; WHO/FAO/IAEA 1996; Hotz & Brown 2004; Miller et
al. 2007). The major factors considered to influence the bioavailability of zinc in
three of these models (Murphy et al. 1992; Hotz & Brown 2004; Miller et al.
2007) are the intakes of dietary zinc (mg/d) and phytate (mg/d). The most recent
mathematical model of Miller et al. (2007) has been developed from stable isotope
studies in adults that have measured both zinc absorption and intestinal excretion
of endogenous zinc from whole day diets containing varying levels of phytate.
The data generated suggest that the predicted effect of dietary phytate on zinc
absorption is larger than previously estimated, and is linear with no threshold.
Furthermore, these investigators emphasized the limited ability of individuals to
adapt to high phytate diets by increasing zinc absorption when they are consuming
adequate amounts of zinc (Hambidge et al. 2008). Indeed, Hunt and co-workers
suggest that such an adaption is only possible after consuming low-zinc, low-
phytate diets for several weeks (Hunt et al. 2008). However, these studies were
performed on adults, and more experimental data that includes dietary zinc,
absorbed zinc, and endogenous faecal zinc excretion measured with a range of
dietary phytate intakes in young children, are needed (Hambidge et al. 2008).

14
2.3.3 Vitamin A

Unlike the minerals, the bioefficacy of vitamin A (i.e., the efficiency in


which nutrients are absorbed and converted to the active form of the nutrient)
(West et al. 2002) is not affected by phytate but instead by the food sources of
vitamin A, as well as the food matrix and fat content of the diet (de Pee & Bloem
2007). Animal-source foods provide highly bioavailable vitamin A in the form of
retinol or retinol esters (Miller et al. 2002), whereas the pro-vitamin A carotenoids
in plant sources have a much lower bioavailability. Of the latter, those in orange
and yellow fruits are more readily bioavailable than those in leafy green
vegetables (de Pee & Bloem 2007). Carotenoids in orange and yellow fruits are
dissolved in oil droplets in chromoplasts and more readily released during
digestion, whereas beta-carotene in leafy vegetables is more tightly bound to
proteins in the chloroplasts and must be freed from their food matrix in order to be
absorbed (Castenmiller & West 1998). The plant-based diets of young children in
low-income environments such as NE Brazil are very low in fat. Their diets often
require small amounts (3–5 grams/meal) of added dietary fat, preferably from
animal products (West et al. 2002), to promote absorption of the pro-vitamin A
carotenoids, and thus achieve a normal vitamin A status (Miller et al. 2002;
Sommer & Davidson 2002).

2.3.4 Vitamin B-12 and folate

Dietary vitamin B-12 is only found naturally in animal-source foods such as


muscle meats, eggs, and dairy products (de Benoist 2008; Stabler & Allen 2004),
and these foods are not affordable for most low-income households, such as those
in the favelas of Salvador, NE Brazil. Although fermented sauces can provide an
additional source of vitamin B-12 (Stabler & Allen 2004), such condiments are
not normally consumed in Brazil, except among the Asian immigrant population.
To facilitate complete digestion, vitamin B-12 must first be released from carrier
food proteins by the action of acid and pepsin in the stomach and then the vitamin
B-12 is available to bind to intrinsic factor for absorption (Russell 2001).

15
In contrast, folate is widely distributed in foods, primarily as
pteroylpolyglutamates. Rich food sources of folate include liver, yeast, leafy
vegetables, fruits, legumes, and nuts. As described earlier for pro-vitamin A
carotenoids in leafy vegetables, absorption of native folates found in certain plants
may also be hindered by the food matrix, becoming entrapped in their insoluble
matrix or cellular structure, thus inhibiting folate release (McNulty & Pentieva
2004; Caudill 2010). Before absorption, the polyglutamate forms of folate must
first be deconjugated to the monoglutamate form in the gastrointestinal tract by
the enzyme folate conjugase (Molloy 2002). The bioavailability of naturally
occurring polyglutamate forms of folate in foods is lower (i.e., about 50%)
compared to that of synthetic folic acid (pteroylmonoglutamic acid) (i.e., about
85%). The latter is used in mandatory fortification of all wheat and corn flour
products in Brazil (Brazil Ministry of Health 2002).

2.3.5 Selenium

Soil concentrations of iodine, zinc, and selenium influence concentrations in


plant-based staples and thus intakes of these micronutrients, provided locally
grown plant-based staples are consumed. Iodine is likely to be lacking in the soil
in some regions of Brazil (Horton & Miloff 2010). However, iodine deficiency is
unlikely in Brazil because mandatory fortification of salt with iodine was
implemented in 1995, and recent reports indicate that about 96% of Brazilian
households are consuming iodized salt (Medeiros-Neto 2009). Widespread zinc
deficiency in soils has also been observed in Brazil, with zinc deficiency
identified in certain crops such as maize and beans (Alloway 2004), two staple
foods consumed by disadvantaged Brazilian households. In contrast, low levels of
selenium in soils and plants have only been reported in southern Brazil (Sillanpää
& Jansson 1992), and there is a paucity of data on the soil selenium content in NE
Brazil. The primary food sources of selenium are sea foods, organ meats (e.g.,
liver and kidney) and muscle meats (Thomson et al. 2007), in addition to Brazil
nuts which have a very high selenium content. Unfortunately, none of these foods

16
are affordable for poor households in NE Brazil. Instead, grain products probably
provide their major food source of selenium.

2.3.6 Interactions between micronutrients

Competitive interactions between some micronutrients can also negatively


affect their absorption and utilization. Examples identified include the interaction
between supplemental iron and zinc (Lönnerdal 2000). High concentrations of
iron in relation to zinc can negatively affect zinc absorption when given
simultaneously in an aqueous solution because of competition in the intestine for a
common nonspecific pathway for iron and zinc absorption (Whittaker 1998). On
the contrary, the iron in fortified foods does not appear to have a negative effect
on zinc absorption (Davidsson et al. 1995; Fair-weather Tait et al. 1995). This has
been attributed to the presence of dietary ligands in food that can chelate iron and
zinc, thereby enhancing their absorption (Sandström et al. 1985).

2.4 Role of host-related factors in the aetiology of


malnutrition
Besides inadequate dietary intakes, several host-related factors, specifically
intestinal and systemic factors, can also have a negative impact on micronutrient
bioavailability, and thus micronutrient status.

2.4.1 Intestinal factors

Intestinal factors include both luminal and mucosal components. Atrophic


gastritis is a key example of a significant luminal factor which can influence
nutrient bioavailability (Gibson 2007), predominantly via its association with
hypochlorhydria (i.e., a decreased secretion of hydrochloric acid from the parietal
cells of the stomach). This low acid gastric condition is often linked to infection
with the bacterium Helicobacter pylori, prevalent amongst young children in low-
income settings in NE Brazil (Parente et al. 2006; Braga et al. 2007). Such
alterations in the pH of the stomach have been associated with malabsorption of

17
Table 2.3 Prevalence of parasitic infections in children living in urban areas in NE Brazil

Author City/Area Study Sample size Age Ascaris Trichuris Hookworm Giardia
(year) (sample location) Design (n) (yrs) (%) (%) (%) (%)

Tsuyuoka et Aracaju, Sergipe


Cross-sectional 360 7 – 16 28.7 15.6 6.4 10.6
al. (1999) (public schools)
108
<8 25.9 16.7 4.6 18.5
(subsample)

Prado et al. Salvador, Bahia Longitudinal


0.5 – 3.8 —
18

597 23.1 17.2 14.5


(2005) (city-wide) (baseline)

Genser et Salvador, Bahia Longitudinal


902 0–3 16.0 14.3 — 9.5
al. (2006) (city-wide) (baseline)

Lima et al. Fortaleza, Ceará RCT1


63 1.5 – 6.0 16.0 13.0 — 3.0
(2010) (community) (baseline)

1
Randomized controlled trial
certain nutrients, most notably vitamin B-12, folate, and ß-carotene. For example,
low acidity in the stomach may inhibit the release of protein-bound vitamin B-12
from food (Carmel 1997), resulting in little vitamin B-12 available for subsequent
binding to intrinsic factor, an essential step for the absorption of vitamin B-12, as
noted earlier (Russell 2000). Likewise, the bioavailability of some minerals may
also be decreased as a result of low gastric acidity, which compromises the
solubilization of iron, calcium, and possibly zinc in the gastro-intestinal tract
(Russell 2001).
Gastric atrophy has also been associated with non-specific bacterial
overgrowth in the small intestine which can cause diarrhoea and poor absorption
of macronutrients and vitamins, specifically vitamin B-12 (Russell 2000).
Furthermore, the bacteria in the small intestine may also utilize available
micronutrients for growth, especially vitamin B-12 freed from food sources
(Russell 2000). Bacterial overgrowth is often present in association with
environmental enteropathy (Menzies et al. 1999), a condition that has been
described among children living in urban slums in Brazil (dos Reis et al. 2007),
and in many other low income countries (Menzies et al. 1999).
The precise aetiology of environmental enteropathy is uncertain. The
condition is characterized by atrophy of the villi in the small bowel with
infiltration of inflammatory T-cells. Such damage reduces the capacity of the
intestine to absorb both macro- and micronutrients, and hence has the potential to
exacerbate malnutrition among urban slum preschoolers in NE Brazil. There is
some evidence that subclinical enteropathy may also be an important aetiological
factor in the high prevalence of zinc deficiency that has been reported among
young disadvantaged children living in low income countries. Zinc homeostasis
appears to be perturbed in these children, as evidenced by their failure to reduce
endogenous faecal zinc excretion, despite the presence of zinc deficiency (Manary
et al. 2010).
Similarly, intestinal parasitic infections such as soil-transmitted helminths
(e.g., Ascaris and Trichuris) and protozoans (e.g., Giardia and Cryptosporidia)
are also common among children living in the crowded urban and semi-urban

19
environments of NE Brazil where hygiene and sanitation conditions are poor
(Barreto et al. 2010) (Table 2.3). Infection with these intestinal parasites can also
reduce absorption of macro- and micronutrients by causing damage to the mucosal
wall of the small intestine with consequent reductions in its absorptive surface and
heightened intestinal permeability (Crompton & Nesheim 2002; Müller & von
Allmen 2005). Such infections can also cause increased peristalsis resulting in a
faster transit time of nutrients through the intestine with concomitant decreased
time available for nutrient solubilisation, and subsequent absorption.
Certainly, malabsorption of vitamin A has been reported with infections of
Ascaris (roundworm) (Taren et al. 1987; Koski & Scott 2001) and Trichuris
(whipworm) (Kongsbak et al. 2006a), as well as with Giardia (Quihui-Cota et al.
2008). The latter has also been linked to reduced absorption of vitamin B-12
(Hjelt et al. 1992; Olivares et al. 2002). A reduction in zinc absorption arising
from increased losses from exacerbated intestinal motility has also been
associated with Trichuris infection (Kongsbak et al. 2006b). In trichuriasis, there
can also be direct and indirect loss of blood, and as a consequence, reduction of
available iron (Ramdath et al. 1995; Le et al. 2007). The blood loss is caused by
the ability of Trichuris to attach securely to the caecal epithelium, with heavy
infections causing damage to the mucosa and its blood vessels (Crompton &
Nesheim 2002; Thurnham & Northrop-Clewes 2007).

2.4.2 Systemic factors

Systemic factors, such as age, ethnicity, and nutrient status of the host, can
also influence the absorption and utilization of some micronutrients (Gibson
2007). Age is known to play a role in absorption among the older population, in
whom the gastrointestinal tract becomes less efficient in absorbing certain
nutrients, most notably vitamin B-12, calcium, and vitamin D (Russell 2000).
However, during childhood, the absorption and utilization of certain
micronutrients may be increased due to the high nutrient demands during periods
of rapid growth (Gibson 2007).

20
Likewise, ethnicity may also play an indirect role in altered nutrient
absorption and / or utilization. For example, Africans have a higher risk of certain
genetic Hb disorders, such as thalassemias or sickle cell disease (Sonati et al.
1991; Masmas et al. 2006). In thalassemias, accelerated iron absorption occurs
despite adequate iron stores, as a result of ineffective erythropoeisis
(Zimmermann et al. 2008a), whereas in sickle cell disease, zinc deficiency can be
induced by hyperzincuria (Prasad 1983; Leonard et al. 1998; Zemel et al. 2002)
and by treatment with chelating drugs (Fung 2010). In Salvador, the capital city of
the state of Bahia in NE Brazil, the population is approximately 80% African or
African/Brazilian (IBGE 2010a), and several studies have confirmed the high
prevalence of sickle cell disease or trait, as well as α3.7thalassemia in this
population (Azevédo et al. 1980; Couto et al. 2003; Adorno et al. 2005). Ethnicity
in Brazil is usually based on self-described physical appearance (e.g., skin colour,
facial and hair characteristics), rather than ancestry. The self-descriptive terms
white, black, brown (mixed), yellow (Asian) and Indian are used routinely in the
Brazilian national census (IBGE 2010a), and these same terms will be used
throughout this thesis.
Finally, it is well recognized that for some nutrients, homeostatic
adaptations play a role in meeting the increased nutrient needs during periods of
rapid growth and in response to inadequate intakes. For example, absorption of
iron is increased when iron stores are depleted, as may occur in disadvantaged
preschool children who consume low intakes of readily available haem iron from
cellular animal-source foods (Oliveira et al. 2007), whereas re-absorption of
endogenous zinc plays a major role in zinc homeostasis (Hambidge & Krebs
2001). For example, when habitual zinc intakes are inadequate, faecal excretion of
endogenous zinc may be reduced in an effort to maintain zinc homeostasis.
However, such an adaptive mechanism may be compromised in the urban
preschoolers studied here who are living in crowded and poor living
environments, where there is a high risk for chronic asymptomatic gastrointestinal
parasitic infestations and / or tropical enteropathy. In such circumstances, both
absorption of zinc and subsequent re-absorption of endogenous zinc may be

21
Table 2.4 Prevalence of anaemia1 and iron deficiency2 in cross-sectional studies of children in NE Brazil

Author City/Area Sample size Age Anaemia Iron Deficiency


(year) (sample location) (n) (yrs) (%) (%)

Salvador, Bahia
Assis et al. (2004) 603 0.5 – 6 46.3 —
(city-wide)
212
3–5 28.3 —
(subsample)
Pernambuco
Oliveira et al. (2006) 746 0.5 – 5 40.6 —
22

(state-wide)
129
3–4 23.3 —
(subsample)
110
4–5 25.5 —
(subsample)
Recife, Pernambuco
Vieira et al. (2007) 162 0.5 – 5 55.6 30.8
(public daycare centres)

1
Anaemia defined as Hb < 110 g/L
2
Iron deficiency defined as serum ferritin < 12.0 µg/L, irrespective of infection
limited (Manary et al. 2010). Few data exist on homeostatic adaptations for the
vitamins, with little evidence to date that absorption of vitamin A, for example, is
influenced by the nutrient status of the host (Biesalski 1997).
Other factors that have the potential to play a very significant role in the
absorption and utilization of nutrients, and hence the nutritional status of children,
are acute and chronic infections. The latter can impair nutrient intakes as a result
of illness-induced anorexia, at a time when both metabolic requirements and
catabolic losses are increased, while simultaneously compromising nutrient
absorption (Bhutta 2006). In resource-poor environments such as the crowded
settings of the urban and semi-urban poor of Salvador, NE Brazil, where children
are likely to be malnourished (Assis et al. 2007), the adverse synergy between
poor nutrition and infection is of particular concern (Schaible & Kaufmann 2007;
Prentice et al. 2008).

2.5 Factors associated with the aetiology of anaemia

2.5.1 Iron

The high prevalence of anaemia (i.e., ~ 50%) among Brazilian children


under-five years of age has been of grave concern in recent years (Jordão et al.
2009) (Table 2.4). The underlying cause of anaemia in children has been assumed
to be iron deficiency (Coutinho et al. 2005), although few studies in Brazil
(Bortolini & Vitolo 2010), particularly in the poor NE region, have examined the
presence of tissue iron deficiency or storage iron depletion.
Tissue iron concentrations are regulated by TfR, a cell surface glycoprotein
that transports transferrin iron across cell membranes (Cook et al. 1993).
Expression of TfR by cells is up-regulated in mild iron deficiency when iron
stores are depleted because the availability of iron to tissues is compromised. A
soluble form of TfR circulates in serum and reflects the number of TfR receptors
expressed on the cell surface (Lynch 2007). Hence, with increasing severity of
iron deficiency, soluble TfR concentrations in serum gradually increase, thus

23
providing a reliable indicator of tissue iron concentrations, even in the presence of
inflammation.
Ferritin is the major iron storage protein, and circulating levels in the serum
reflect the amount of storage iron in the liver in the absence of inflammation (Mei
et al. 2005). However, because ferritin is an acute phase protein, serum levels are
elevated in the presence of acute and chronic inflammation. Once liver iron stores
are exhausted, indicated by a marked decrease in serum ferritin concentrations in
the absence of inflammation, levels of circulating iron (reflected by serum iron)
fall and the supply of iron to the bone marrow is reduced. As a consequence, there
is a marked reduction in the concentration of Hb in the erythrocytes resulting in
the development of microcytic, hypochromic anaemia (Hallberg 2001).
Currently, the indicators recommended by the World Health Organization
(WHO) for assessing and interpreting the iron status of a population are Hb and
serum ferritin, together with a biomarker of chronic infection, preferably α-1-
glycoprotein (AGP) (WHO/CDC 2007). However, in settings such as the crowded
urban slums in Salvador where the prevalence of infections among young children
is likely to be high, soluble TfR in serum may also be useful (Asobayire et al.
2001), because unlike serum ferritin, it is not influenced by inflammation and
infection (WHO/CDC 2007).
Even though studies in Brazil on the prevalence of iron deficiency, based on
serum soluble TfR or serum ferritin are limited (Vieira et al. 2007; Bortolini &
Vitolo 2010), daily supplementation with liquid ferrous sulphate drops (1–2
mg/kg) for all children from 6–18 months of age has been recommended since
1995 (Brazil Ministry of Health 2005), and national iron fortification of all corn
and wheat flour products was introduced in 2004 (Brazil Ministry of Health
2002). To our knowledge, whether the implementation of these strategies has been
associated with a reduction in the prevalence of iron deficiency anaemia or iron
deficiency without anaemia among young children in NE Brazil has not been
investigated (Assunção et al. 2007; Shibukawa et al. 2008).

24
2.5.2 Genetic Hb disorders

Caution is needed when attributing the anaemia in children living in


Salvador, NE Brazil to iron deficiency, when such a high proportion of the
population is African or African/Brazilian (IBGE 2010a), often with genetic Hb
disorders (Azevédo et al. 1980; Couto et al. 2003; Adorno et al. 2005). Mild to
severe anaemia may occur as a result of ineffective erythropoiesis (i.e., an
increase in the number of immature red blood cells destroyed within the bone
marrow) in these conditions, the severity depending on whether the genetic Hb
disorder is heterozygous or homozygous. Furthermore, the presence of certain
genetic Hb disorders may compromise the usefulness of serum ferritin and soluble
TfR for assessing the prevalence of iron deficiency (Hillman et al. 2005). In
thalassemias, for example, even when concentrations of Hb and mean cell volume
are low and indicative of microcytic hypochromic anaemia, concentrations of both
serum ferritin and soluble TfR levels can be elevated. This trend is caused by the
ineffective erythropoiesis associated with thalassemias, a condition that stimulates
an increase in dietary iron absorption even when iron stores are adequate (Earley
et al. 1990; Rees et al. 1998). Moreover, because the majority of TfR is derived
from the erythroid bone marrow, conditions such as ineffective erythropoiesis that
increase the cellular demand for iron will up-regulate the formation of TfR, thus
elevating serum soluble TfR concentrations (Earley et al. 1990; Rees et al. 1998;
Gibson 2005). However, when iron deficiency exists in the absence of these
genetic Hb disorders and inflammation, then Hb and serum ferritin may be
positively correlated.

2.5.3 Vitamin A

The presence of vitamin A deficiency is an additional factor that may


contribute to the high prevalence of anaemia among young children in NE Brazil.
The precise role of vitamin A in the aetiology of anaemia is uncertain. Several
mechanisms have been implicated, including a possible role for vitamin A in the
absorption of iron, as well as in the mobilization of iron from existing spleen or
liver stores into the bone marrow to support increased erythropoiesis. These

25
Table 2.5 Prevalence of vitamin A deficiency1 and marginal vitamin A2 status in cross-sectional studies of children in NE Brazil

Vitamin A Marginal vitamin


Author City/Area Sample size Age
deficiency A status
(year) (sample location) (n) (yrs)
(%) (%)
Martins et al. Sergipe
607 0.5 – 5 32.1 —
(2004) (state-wide)
117
2–5 30.5 —
(subsample)
Fernandes et al. Recife, Pernambuco
311 0.5 – 5 6.6 22.2
(2005) (public daycare centres)
26

Teresina, Piauí
Paiva et al. (2006) 631 3–6 15.4 29.0
(public daycare centres)

Vieira et al. Fortaleza, Ceará


102 0.2 – 9 2.9 20.6
(2008) (community)

de Azevedo et al. Recife, Pernambuco


344 2–5 7.7 29.6
(2010) (public daycare centres)

1
Vitamin A deficiency defined as serum retinol < 0.7 µmol/L
2
Marginal vitamin A status defined as serum retinol 0.7 – 1.05 µmol/L
processes may be impaired, even in the presence of marginal vitamin A deficiency
(i.e., serum retinol 0.7–1.05 µmol/L), so that iron is retained in the liver and
spleen and less available for erythropoiesis (Bloem 1995; Semba & Bloem 2002).
Additionally, iron-dependent erythropoiesis may be hindered by a low vitamin A
status because of the role of vitamin A in stimulating production of erythropoietin
(Evans 2005; Zimmermann et al. 2006).
To assess risk of vitamin A deficiency at the population level, serum retinol
is the recommended biomarker (Sommer & Davidson 2002), although retinol
binding protein (RBP) is being increasingly employed in low-income settings
where correlation between serum retinol and RBP has been established. The latter
is more stable and less expensive (de Pee & Dary 2002). A single cutoff for serum
retinol concentrations indicative of vitamin A deficiency of < 0.70 µmol/L has
been recommended by WHO (1996) and the International Vitamin A Consultative
Group (IVACG) (de Pee & Dary 2002), although in population groups where the
prevalence of vitamin A deficiency is low, serum retinol concentrations indicative
of suboptimal vitamin A status, defined as 0.70–1.05 µmol/ L, are sometimes used
(Ballew et al. 2001).
Deficiency of vitamin A was declared a public health problem in young
children in NE Brazil more than three decades ago (WHO 1982). Since that time,
efforts to increase the coverage of vitamin A supplementation programmes have
been instigated in this region, although routine implementation is still lacking
(Osório et al. 2004; de Almeida et al. 2010). Nevertheless, recent studies have
indicated that the high prevalence (i.e., > 30%) of vitamin A deficiency reported
earlier has now decreased to lower rates (i.e., 3–15%) (Table 2.5). The prevalence
of marginal vitamin A deficiency is still high among children (i.e., > 20%) in NE
Brazil, and has been associated with anaemia in some studies (Osório et al. 2004),
as reported elsewhere (van Stuijvenberg et al. 1997; Thurlow et al. 2005).

2.5.4 Folate and vitamin B-12

Anaemia often co-exists with other micronutrient deficiencies besides iron


and vitamin A, such as folate, vitamin B-12, zinc, and / or selenium, although

27
limited data exist on biomarkers for these latter micronutrients in preschool
children in NE Brazil. Severe deficiency of folate or vitamin B-12 can result in
megaloblastic anaemia, characterized by abnormally large erythrocytes attributed
to decreased cell division of bone marrow cells (Scott 2007). Such abnormalities
in cell replication arise because the production of 5,10-methylene-
tetrahydrofolate, essential for the synthesis of DNA, is impaired. Specifically,
when folate status is low, the supply of methylene-tetrahydrofolate
monoglutamate required to transfer the methyl group to generate tetrahydrofolate
is limited. When vitamin B-12 is in short supply, the methylation cycle cannot
function properly because of a reduction in the vitamin B-12 dependent methyl
acceptor, methionine synthase, resulting in a decreased availability of
tetrahydrofolate to support DNA synthesis (Scott 2007).
To assess folate status, both serum folate, an indicator of recent folate
intake, and erythrocyte folate, a reflection of liver folate levels and thus folate
stores, are recommended (Gibson 2005; de Benoist 2008). Cut-offs for serum and
erythrocyte folate are based on data available from the third U.S. National Health
and Nutrition Examination Survey (NHANES) (1988-1994), which takes into
account age, sex and race, and is based on microbiological assay analysis (Pfeiffer
et al. 2007). Measurement of total serum vitamin B-12 is the biochemical test
recommended for routine screening, with the caveat that reference limits are assay
specific (de Benoist 2008). Where possible, because of the low sensitivity of
serum vitamin B-12, it is advisable to measure methylmalonic acid in serum or
urine, as elevated levels are a sensitive and specific marker of functional vitamin
B-12 deficiency. However, the analytical methods available for methymalonic
acid are complex and not routinely accessible (Gibson 2005).

2.5.5 Zinc

There are several plausible mechanisms whereby low zinc status could
potentially have an impact on Hb concentrations, and thus anaemia (Gibson et al.
2008). Several zinc-dependent enzyme systems are important in the synthesis of
Hb. Specifically, aminolevulinic acid dehydrase assists with haem synthesis

28
(Garnica 1981), whereas thymidine kinase and DNA polymerase are involved in
the synthesis of DNA (Spivak et al. 1992). Interference with DNA synthesis
induces abnormal cell replication, especially in rapidly proliferating cells such as
the haematopoietic system. GATA-1, a zinc-finger transcription factor, is also
necessary for normal haematopoiesis (Labbaye et al. 1995). Other mechanisms
whereby zinc may have a role in anaemia include its role in stabilizing red cell
membranes (O’Dell 2000), as well as the association of zinc with increases in
plasma insulin-like growth factor (IGF)-1 which stimulates erythropoiesis
(Nishiyama et al. 1999).
To our knowledge, whether zinc has a role in the anaemia reported in young
children in NE Brazil has not been investigated. One study of preschoolers (ages
1-5 years) in southeast Brazil reported a significant improvement in Hb following
zinc supplementation for four months (Silva et al. 2006). However, whether this
haematological response can be attributed to the role of zinc in Hb synthesis, or an
indirect effect arising from a reduction in the anaemia of chronic inflammation
(ACI) is not clear. No biomarkers of infection were measured in this study.
For zinc, serum or plasma zinc is considered a useful biomarker of zinc
status at the population level but not at the individual level (Hess et al. 2007),
provided the serum or plasma is collected and analysed using recommended
procedures (Hotz & Brown 2004). Suggested lower reference limits for serum
zinc concentrations based on age group, sex, fasting status, and time of day of
collection derived from the NHANES II data (1976-1980) are available (Hotz et
al. 2003).

2.5.6 Selenium

Low serum selenium concentrations have also been associated with anaemia
in children and adults (Van Nhien et al. 2008; Semba et al. 2009) through a
mechanism primarily involving the antioxidant function of selenium. For
example, reductions in the activity of key selenoenzymes, such as glutathione
peroxidase, can negatively affect Hb concentrations via the diminished oxidation
protection of red blood cells (Nagababu et al. 2003). Likewise, decreased activity

29
of circulating thioredoxin reductase, another selenoprotein, may contribute to the
upregulation of hepatic haem oxygenase-1, which can initiate haem catabolism
(Mostert et al. 2003). Furthermore, accelerated oxidative stress associated with
low selenium status may increase levels of circulating interleukin-6, which in turn
upregulate hepcidin, thus inhibiting absorption and mobilization of iron from
hepatic stores (Roy & Andrews 2005; Young et al. 2009).
Serum selenium has been used in several studies of children as a biomarker
of population selenium status. No accepted reference ranges exist for interpreting
serum selenium because of the wide-spread geographical variations in serum
selenium concentrations arising from differences in soil selenium levels.
However, cut-offs for plasma selenium for assessing adequate selenium status
have been devised by Thomson (2004), based on the concentrations required for
optimal activity of certain selenoproteins and protection against some cancers.

2.5.7 Interactions between micronutrients

Several important inter-relationships between micronutrients have been


described among children living in poor resource settings, which have the
potential to exacerbate the risk of anaemia. For example, suboptimal zinc status
among children living in low-income settings has been associated with a low
vitamin A status (Muñoz et al. 2000; Rahman et al. 2002; Thurlow et al. 2006),
which also has a role in anaemia, as noted earlier. Zinc is essential for the
synthesis of RBP in the liver, the protein responsible for the hepatic release,
transport, and tissue utilization of vitamin A (Christian & West 1998). To date,
one study in southeastern Brazil attempted to investigate the association between
zinc and vitamin A status in preschool children (Ferraz et al. 2007). However,
adventitious contamination during the blood sample collection may have
contributed to the lack of association because reported serum zinc concentrations
are above the normal range for children of this age group (Pilch & Senti 1984).
No studies in NE Brazil to date have examined these inter-relationships, in part
because there have been no studies on multiple micronutrient concentrations.

30
In some studies of children, serum selenium concentrations have been
positively associated with serum zinc (Krittaphol et al. 2006; Lander et al. 2008).
The release of redox-inert zinc from metallothionein to the zinc dependent
enzyme, copper-zinc superoxide dismutase, is dependent on the role of
selenoproteins, specifically glutathione peroxidase (Maret 2000; Lyons et al.
2004), which may explain this association.

2.5.8 Role of chronic inflammation

Chronic infection plays an important role in the aetiology of the ACI.


Initially, ACI is normocytic, normochromic anaemia, but when prolonged,
microcytic hypochromic anaemia may develop, as seen in iron deficiency
anaemia. Characteristic features of ACI are hypoferremia (i.e., a decrease in
serum iron and total iron binding capacity) arising from defects in the release of
iron from the reticulo-endothelial cells ─ previously termed the “mucosal block”.
Together, these defects result in a shortage of iron in the bone marrow, despite
adequate iron stores (Thurnham & Northrop-Clewes, 2007).
Recent research indicates that the principal mechanism in ACI is the
hepcidin-ferroportin interaction. During inflammation, pro-inflammatory
cytokines such as interleukin-6 stimulate an increase in circulating hepcidin, a
small peptide hormone produced mainly by the liver but also by adipose tissue.
Hepcidin is responsible for controlling the iron export protein ferroportin (Fpn),
which is involved in the transfer of iron from the enterocyte to the circulating
transferrin pool (Lynch 2007). Hepcidin binds to Fpn, causing Fpn to be
internalized and degraded (Nemeth et al. 2004), thus blocking the export of iron
from iron storage into the circulation. As a consequence, functional iron
deficiency develops, characterized by elevated soluble TfR levels, despite
adequate iron stores. It is noteworthy that in overweight school age children
(Aeberli et al. 2009), obesity-related inflammation has also been associated with
increased hepcidin secretion and lower tissue iron status (as indicated by elevated
serum soluble TfR concentrations) in some studies, although serum hepcidin is
not routinely measured.

31
Additionally, zinc and vitamin A have essential roles in immune function.
Deficiency of zinc and / or vitamin A may result in a decreased resistance to
infection, and thus an increase in the ACI, again through the response by hepcidin
to infection or inflammation.

2.5.9 Interpretation of ferritin, retinol, zinc, and selenium biomarkers

Care must be taken in interpreting serum levels of ferritin, retinol, zinc, and
possibly selenium as acute and chronic infections influence concentrations of
these micronutrients in serum. Limited data are available on the effect of acute
phase reactions on selenium, although one study has shown an inverse
relationship between serum selenium and C-reactive protein (CRP), possibly
attributed to a redistribution of selenium from serum and liver to muscle tissue
during the acute phase response (Maehira et al. 2002). However, even subclinical
infections have been shown to reduce both serum zinc and retinol concentrations
(Thurnham et al. 2003; Hotz & Brown 2004), and elevate serum ferritin levels
(Tomkins 2003). Such effects may result in misclassification of micronutrient
deficiencies (Brown et al. 1993; Stephensen & Gildengorin 2000; Thurnham et al.
2005). To avoid such misclassification, biomarkers of acute and chronic infections
such as serum CRP and AGP, respectively, should be measured so that
concentrations of serum zinc, retinol, and ferritin can be adjusted accordingly,
when infections exist (Thurnham et al. 2005).

2.6 Factors associated with impaired linear growth


Chronic inadequate dietary intakes and / or diminished micronutrient
absorption can often lead to deficiencies of certain growth-limiting nutrients, and
thus contribute to impaired linear growth. Childhood stunting is defined as length
or height-for-age less than -2 standard deviations (SD) below the age-specific
median of the reference population (HAZ < -2 SD) (Kuczmarski et al. 2000) and
has been associated with increased risk of mortality (Black et al. 2008). However,
even mild stunting (HAZ < -1 SD) is associated with a heightened risk of

32
morbidity, impaired cognition, and delayed motor development in children
(Grantham-McGregor et al. 1991; Hambidge 1997; Stephensen 1999). Based on
recent studies, the prevalence of mild stunting in NE Brazil remains high (> 15%)
(Table 2.6), and could be related to the co-existence of multiple growth-limiting
nutrient deficiencies among children living in peri-urban low socio-economic
households (Allen 1994; Rosado 1999). However, few studies have examined the
relationship between childhood stunting and deficits in protein, together with
deficiencies of the growth-limiting micronutrients, in this impoverished region.

2.6.1 Protein

Insufficient intake of protein limits the supply of the appropriate balance of


essential amino acids for lean tissue synthesis, as well as nitrogen for the synthesis
of nonessential amino acids (Gibson & Hotz 2001). Such limitations, particularly
of sulphur-rich amino acids (e.g., methionine and cystine), reduces synthesis of
3’-phosphadenosine 5’-phosphosulphate, a high energy sulphate compound
needed to produce glycosaminoglycans. The latter are essential for the synthesis
of cartilage at the growth plates and thus for skeletal growth (Golden 2009). Low
intakes of protein, especially from animal-source foods, are usually accompanied
by low intakes of other key growth-limiting micronutrients, such as calcium and
zinc, also found in sources of animal protein (Allen 1994). Hence, it is not
surprising that improvements in linear growth have been reported in children in
some low-income countries following increased intakes of animal-source foods
(Marquis et al. 1997; Grillenberger et al. 2006). Unfortunately, expensive animal-
source foods are seldom consumed in poor Brazilian households (Antunes et al.
2010), as noted earlier.

2.6.2 Calcium and vitamin D

As might be expected, low intakes of bioavailable calcium have been


reported in deprived Brazilian households, where traditional diets are based on
cereals and legumes and intakes of dairy products, a readily available source of
calcium, are low (Pettifor 2004; Salles-Costa et al. 2010). Calcium plays an

33
Table 2.6 Prevalence of underweight, stunting, wasting and overweight in cross-sectional studies of children in NE Brazil

Sample Mild Mild Mild Over-


Author City/Area Age 1
size Underweight Underweight Stunting Stunting4
2 3
Wasting 5
Wasting6 weight7
(year) (Location) (yrs)
(n) (%) (%) (%) (%) (%) (%) (%)

Assis et Salvador, Bahia


1740 0.5 – 6 3.2 17.2 3.0 15.1 — — —
al. (2007) (city-wide)

Viera et Fortaleza, Ceará


102 0.2 – 9 8.8 41.2 2.9 15.7 0.0 19.6 —
al. (2008) (community)
34

Ferreira &
Alagoas
Luciano 1386 0–5 2.9 — 10.4 — 1.2 — 9.7
(state-wide)
(2010)
242
3–4 2.1 — 7.9 — 1.2 — 9.5
(subs)
161
4–5 3.7 — 8.1 — 1.9 — 5.6
(subs)
1
Underweight: WAZ < -2SD
2
Mild underweight: WAZ < -1 to ≥ -2SD
3
Stunting: HAZ < -2SD
4
Mild stunting: HAZ < -1 to ≥ -2SD
5
Wasting: WHZ < -2SD
6
Mild wasting: WHZ < -1 to ≥ -2SD
7
Overweight: WHZ > 2SD
important structural role in the formation of bone, bone mineral density, and bone
remodelling, and children with marked and prolonged deficits in calcium intake
may develop thin, demineralised bones (Golden 2009). A positive correlation has
been observed between habitual calcium intakes from birth and bone mineral
content at five years of age (Lee et al. 1993).
Deficits in calcium intakes have been associated with nutritional rickets in
vitamin D-replete children of African ethnicity (Pettifor 1991; de Lucia et al.
2003). Nutritional rickets is characterized by a failure to properly mineralize bone
tissue and is more commonly linked to vitamin D deficiency. The mechanism
whereby low calcium intakes induce rickets is not clear (Abrams 2002). Low
calcium intakes may increase requirements for vitamin D (Clements et al. 1987),
which in turn is known to play an important role in calcium homeostasis and
parathyroid function. 1,25-dihydroxyvitamin D3 potentiates the synthesis of IGF-1
and stimulates chondrocyte proliferation and cell differentiation (Soliman et al.
2008). Thus, in vitamin D-deficient children, reduced concentrations of IGF-1
may contribute to defective matrix calcification and irregular maturation of
chondrocytes, resulting in growth-plate abnormalities and impaired linear growth
(David 1991; Soliman et al. 2008).
The Brazilian diet is typically low in vitamin D as well as calcium. Oily fish
and organ meats, the natural food sources of vitamin D (Gibson 2005), are neither
affordable nor culturally acceptable (Bandeira et al. 2006). Nevertheless,
childhood rickets is not common in equatorial areas, such as NE Brazil, because
of the high number of direct sunshine hours year-round. However, children of
African descent living in urban areas such as Salvador may be at an increased risk
of vitamin D deficiency because darker-skinned individuals require more time in
the sunshine to produce vitamin D, and urban children often tend to play inside
(de Lucia et al. 2003). Furthermore, ozone pollution in the air of large
industrialized cities has been shown to significantly reduce the amount of ultra-
violet radiation available for vitamin D metabolism (Pettifor 2004), and hence
vitamin D deficiency could be of concern if Salvador becomes more industrialized
in the future.

35
2.6.3 Zinc

Numerous studies have reported a positive association between sub-optimal


zinc status and stunting in children in low-income settings (Gibson et al. 1991;
Cavan et al. 1993; Ruz et al. 1997; Gibson et al. 2007; Gibson et al. 2011).
However, there is a paucity of data on the zinc status of Brazilian children. In one
study, Beinner et al. (2010) reported no association between serum zinc
concentrations and stunting, perhaps because of the low levels of stunting found
in this Brazilian study. In studies elsewhere (Gibson et al. 1991; Gibson et al.
2007; Gibson et al. 2011), boys have had a higher risk of low zinc status and
lower growth percentiles than girls, attributed to their faster growth rate in early
childhood and greater lean body mass, and thus higher requirements for zinc.
Zinc potentiates the effect of vitamin D on bone metabolism through the
stimulation of DNA synthesis in bone cells (Salgueiro et al. 2002). Zinc may also
enhance the effect of IGF-I on bone formation and mineralization by increasing
proliferation of osteoblastic cells (i.e., activating aminoacyl-tRNA synthetase in
osteoblasts) and inhibiting osteoclastic bone resorption (Yamaguchi & Fukagawa
2005; Cole & Lifshitz 2008). A reduction in bioavailable zinc induced by high
phytate cereal and legume-based diets may affect membrane signalling systems
and other intracellular messengers that coordinate cell proliferation in response to
IGF-1, thus impairing the necessary cell division required for growth (MacDonald
2000).
Furthermore, zinc interacts with several hormones involved in bone growth,
including thyroid hormones, testosterone, insulin and somatomedin-c (Salgueiro
et al. 2002). Finally, even mild zinc deficiency has the potential to indirectly
affect linear growth in children through increasing the susceptibility to infection
or through impaired taste acuity, resulting in decreased appetite and lower food
intakes (Hambidge 1997; Gibson & Hotz 2001; Rivera et al. 2003).

2.6.4 Iodine and selenium

Delayed linear growth can also be the result of severe iodine deficiency
(Allen 1994; Kapil & Bhavna 2002) because iodine is necessary for the

36
production of thyroid hormones (i.e., thyroxine and tri-iodothyronine ), which in
turn affect the secretion of growth hormone and the cellular response to IGF-1
(Gibson & Hotz 2001; Viteri & Gonzalez 2002). However, the risk of iodine
deficiency in Brazil is likely to be low because of the mandatory iodization of salt,
as noted earlier, and the capacity of the body to store iodine. However, selenium is
involved in the metabolism of iodine, specifically via selenium-dependent
iodothyronine deiodinase enzymes which are required for the conversion of
thyroxine to tri-iodothyronine, the more active form of thyroid hormones (Lyons
et al. 2004). Hence, it is possible that impairments in linear growth could arise
through an interaction between selenium and iodine, if children in NE Brazil are
at high risk to low selenium status.

2.6.5 Vitamin A

The role of vitamin A on linear growth is less well understood. Several


studies have examined the effect of vitamin A supplementation on linear growth,
but the evidence is inconsistent (Rahmathullah et al. 1991; Hadi et al. 2000), even
in NE Brazil (Chen et al. 2003; Lima et al. 2010). It is possible that a positive
impact on growth is only observed when vitamin A is the first limiting nutrient
(West et al. 1997). Nevertheless, despite these equivocal results, vitamin A is
necessary for the activity of cells in the epiphyseal cartilage, and hence does have
a direct metabolic effect on bone growth (Fawzi et al. 1997; Gibson & Hotz
2001). Furthermore, retinoic acid, an active metabolite of vitamin A, is involved
in the activation of gene transcription via a nuclear receptor and may also affect
growth through its link with thyroid hormone receptors (Umesono et al. 1988).
Vitamin A may also have an indirect influence on linear growth through its
role in the maintenance of immune function, as vitamin A affects both mucosal
integrity and systemic immunity. Hence, in vitamin A deficiency there is an
increased susceptibility to infections, which may thus impact on growth (Gibson
& Hotz 2001; Hadi et al. 2004).

37
2.6.6 Parasitic infections

Several other factors, including parasitic infections, can cause delayed linear
growth in young children via several well-established mechanisms. For example,
stunting has been associated with infections of Ascaris lumbricoides and Trichuris
trichiura in Brazilian children (Saldiva et al. 1999; Jardim-Botelho et al. 2008a),
and in NE Brazil specifically, with Giardia intestinalis in preschoolers (Prado et
al. 2005; Matos et al. 2008) and Schistosoma mansoni in school-aged children,
especially boys (Parraga et al. 1996; Assis et al. 2004a). Parasitic infections are
often associated with loss of appetite and malabsorption, which can result in
nutrient deficits, thus contributing to impaired growth, as noted earlier
(Stephenson et al. 2000a; Hall et al. 2008).

2.6.7 Genetic Hb disorders

An additional factor that could potentially contribute to impaired linear


growth in some NE Brazilian children of African descent is genetic Hb disorders
(Cipolotti et al. 2000). Children with sickle cell disease and thalassemias have
increased energy and nutrient requirements, attributed to an elevated resting
energy expenditure and protein turnover. These disturbances are probably
associated with the hyperactive bone marrow and increased cardiac demands that
arise from chronic anaemia (Barden et al. 2002; Fung 2010). Furthermore,
increased oxidative stress associated with these conditions may contribute to
higher nutrient requirements (Walter et al. 2006; Claster et al. 2009), which may
not be met because of reduced dietary intakes as a result of illness-induced
anorexia (Kawchak et al. 2007). Indeed, deficits in intakes of some of the growth-
limiting nutrients (i.e., zinc, calcium, and vitamin D) have been suggested in
children with these conditions (Reed et al. 1987; Kawchak et al. 2007; Fung
2010).
Both sickle cell disease and thalassemias have been associated with zinc
deficiency, which can lead to impaired linear growth as noted earlier, as well as to
decreased lean tissue synthesis. The aetiology of zinc deficiency in these disorders
is uncertain. Possible mechanisms that have been proposed in sickle cell disease

38
include heightened requirements secondary to erythrocyte hemolysis, increased
urinary zinc losses and protein turnover, chronic inflammation, decreased dietary
zinc intake and inadequate net intestinal absorption (Leonard et al. 1998; Zemel et
al. 2002). In thalassemias, damage to the proximal tubules and increases in urinary
zinc may be involved (Fung 2010). Furthermore, chelation therapy used to treat
iron overload in both thalassemia and sickle cell anaemia patients may cause a
secondary zinc deficiency (Fung 2010).

2.7 Factors associated with disturbances in body


composition

2.7.1 Overweight

The rising prevalence of overweight in preschoolers in Brazil is of


increasing concern (Simon et al. 2008; de Novaes Oliveira et al. 2010; Shoeps et
al. 2011). Childhood overweight is associated with many long-term adverse health
effects, including hypertension, atherosclerosis, hyperinsulinaemia, impaired
glucose tolerance, adverse serum lipid profiles, and increased risk of cancer later
in life; these are discussed in detail in the WHO/FAO expert consultation (WHO
2003).
Furthermore, functional iron deficiency has also been associated with
overweight in children (Pinhas-Hamiel et al. 2003; Zimmermann et al. 2008b;
Aeberli et al. 2009; Amato et al. 2010; Cepeda-Lopez et al. 2011), although to
date, this relationship has not been investigated in NE Brazil. Conditions of
chronic inflammation, such as overweight or obesity, up-regulate
proinflammatory cytokines such as interleukin-6 and leptin, which in turn up-
regulate hepcidin expression from the liver and adipose tissue. As a result, there is
a reduction in iron absorption as well as a decrease in recycling and mobilization
of iron from hepatic stores (Roy & Andrews 2005; Thurnham & Northrop-Clewes
2007; Young et al. 2009).
Among children in Brazil from disadvantaged households, both mild
stunting and overweight often co-exist (Sawaya et al. 1998; Hoffman et al. 2000a;

39
Hoffman et al. 2000b). Such a pattern during early childhood has been associated
with a higher risk of adolescent and adult obesity and long-term non-
communicable diseases in several countries, including Brazil (Sawaya & Roberts
2003; Stein et al. 2005). Several mechanisms have been proposed to explain this
phenomenon. It may be associated, at least in part, with zinc deficiency induced
by the consumption of cheap, high-energy low-zinc density foods in poor
households (Popkin et al. 1996; Fernald & Neufeld 2007; Antunes et al. 2010).
Children with low zinc status are not only shorter, but are often fatter (Prentice
1993; Perrone et al. 1998; Gibson et al. 2000b), attributed to the role of zinc in
accrual of fat-free mass (Arsenault et al. 2008). However, as noted earlier, this
relationship has not been investigated because data on the zinc status of
preschoolers in NE Brazil are limited.
Alternative mechanisms proposed include the postulation that low energy
intakes during growth may result in a decrease in the synthesis of IGF-1, and thus
reduced lipolysis leading to increased fat deposition over time (Hoffman et al.
2000a; Veiga et al. 2010). At the same time, the higher ratio of cortisol to insulin
that is characteristic of this pattern results in reductions in accrual of fat-free mass
and linear growth, as well as cortisol-induced truncal obesity (Sawaya et al. 1998).
Another suggested mechanism involves the decreased ability of stunted
children to self-regulate their energy intake, which could in turn lead to
opportunistic overeating, as a result of underlying metabolic changes or
behavioural factors associated with long-term food insecurity (Hoffman et al.
2000b). The impact of an inadequate nutrient supply during intrauterine life and
subsequent long-term alterations to insulin sensitivity and energy metabolism may
be another factor that may account for the increased susceptibility to gain weight
(Hales & Barker 2001).

2.7.2 Underweight

In low-income settings, underweight in childhood is caused by numerous


factors, including inadequate intakes of energy and nutrients through lack of
available food, or as a result of illness-induced anorexia, malabsorption, and / or

40
increased metabolic requirements associated with parasitic, viral or bacterial
infections, as noted earlier (sections 2.3 and 2.4). Diarrhoeal illness, when severe
and chronic, can be a major contributor to weight loss in young children living in
impoverished settings (Bhan et al. 2001). There are several reports of diarrhoea
among young children in NE Brazil (Newman et al. 2001; Strina et al. 2003;
Genser et al. 2006), attributed in some studies to the presence of intestinal
parasites, poor sanitation conditions, and low socioeconomic status (Genser et al.
2006). Hence, it is not surprising that Issler et al. (1996) reported that Brazilian
children living in extreme poverty had a much higher risk of underweight (i.e.,
nearly four times higher) than those living in relatively less poverty.

2.8 Micronutrient deficiencies associated with impaired


immune function
Optimal nutritional status and the integrity of the immune system are closely
linked, with adequate nutriture of protein, zinc, vitamin A, iron, and selenium of
paramount importance for a properly functioning immune system (Semba 1994;
Fraker et al. 2000; Beard 2001; Arthur et al. 2003). Thus deficiencies of these
nutrients, especially when multiple deficiencies co-exist, can decrease resistance
to pathogens, and increase the susceptibility of young children to infections,
affecting both their duration and severity (Bhaskaram 2001). This review will
focus on the role of micronutrients in immune function, particularly cell-mediated
immunity involving thymus dependent lymphocytes (T-cells) that affect the
destruction of bacteria and virus-infected cells.

2.8.1 Zinc

Zinc is known to play a central role in the immune system, influencing


multiple aspects that range from maintenance of the physical skin barrier to both
cell-mediated and humoral immunity (Shankar & Prasad 1998; Black 2001). Cells
with a rapid turnover, such as the epithelial linings of the respiratory system and
gastrointestinal tract, are especially sensitive to zinc deficiency. In the latter, the

41
villi atrophy, the activity of brush border disaccharidase may be reduced, and
intestinal transport impaired (Bhandari et al. 1996). Even in mild zinc deficiency,
an increased susceptibility to recurrent episodes of acute lower respiratory
infection (e.g., pneumonia) and diarrhoea (Hambidge 1992; Black & Sazawal
2001) may occur. Thus, it is not surprising that a recent meta-analysis reported a
beneficial effect of zinc supplements on both the incidence of acute lower
respiratory infection and diarrhoea, particularly for the latter among children over
12 months of age (Brown et al. 2009).
Zinc is also essential for the cell mediated arm of the immune system,
particularly T-cell mediated function. In zinc deficiency, there is a reduction in the
size of the thymus, the central organ for the development of T-lymphocytes. As a
result, there is a decrease in lymphocyte proliferation and loss of delayed
hypersensitivity responses (an in vivo measure of cell-mediated immune response)
(Fraker et al. 2000; Sherman 2000; Cunningham-Rundles et al. 2005).
Furthermore, zinc deficiency is associated with apoptosis in several immune
system cell types, including a reduction in the activity and function of neutrophils,
natural killer cells, and macrophages (Wirth et al. 1989; Rink & Gabriel 2000;
Semba 2004). Moreover, because zinc is necessary for the generation of reactive
oxygen species in stimulated neutrophils, zinc deficiency diminishes protection
from the damaging effects of oxygen radicals generated during immune activation
(Shankar & Prasad 1998; Hasegawa et al. 2000).
The humoral component of the immune system is also affected by zinc
deficiency, although to a lesser extent than the cell-mediated arm. Zinc deficiency
affects the development of lymphocytes in the bone marrow (B-cells), reducing
both their number and function. As a result, specific antibody responses are
reduced in zinc deficiency, with abnormalities in mitogenic and cytokine
responses (Fraker et al 2000; Sherman 2000). Thus, the depletion of both T- and
B-cells in zinc deficiency leads to a substantial impairment of the lymphocytic
branch of the immune system.

42
2.8.2 Vitamin A

Vitamin A is necessary for the maintenance of the integrity of mucosal


epithelia, and also like zinc, has a role in both cell-mediated and humoral
immunity (Gerster 1997; Villamor & Fawzi 2005). Consequently, even mild
vitamin A deficiency has been associated with impairment in the integrity of the
gut and changes in the respiratory tract, including loss of microvilli in the
gastrointestinal tract, reduction in the production of mucous-secreting ciliated
epithelium of the respiratory tract, and loss of mucin and goblet cells in both tracts
(Semba 2004). Such disturbances have been associated with increased
susceptibility to gastro-intestinal and respiratory diseases in children (Bloem et al.
1990; Thurnham et al. 2000).
Vitamin A deficiency may also influence T-cell lymphocyte
immunocompetence through several mechanisms. For example, circulating T-cell
subsets (i.e., CD4+, CD8+) may be reduced, and the production and activation of
natural killer cells inhibited in vitamin A deficiency (Semba 2004). In contrast,
the total number of neutrophils and macrophages are increased in vitamin A
deficiency, although their protective function is impaired as indicated by
decreased phagocytosis and killing of pathogens (Stephensen 2001). Finally,
vitamin A deficiency reduces the Th-2 mediated antibody response leading to a
compromised immunity to infections that are dependent upon antibody responses.
These include tetanus toxoid and diphtheria antigens, with a consequent
diminished response to vaccines (Semba et al. 1992; Rahman et al. 1999).

2.8.3 Iron

Iron plays an integral role in the immune response as it is essential for


proper cell differentiation and cell growth (Wintergerst et al. 2007). Iron is also a
critical component of peroxide- and nitrous oxide-generating enzymes which are
vital for the normal enzymatic functioning of immune cells (Beard 2001).
Activation of lymphocytes, particularly T-cell proliferation, is iron-dependent. T-
cells stimulate TfR production on the cell surface by an interleukin-2-dependent
pathway, and thus facilitate tissue iron uptake (Thibault et al. 1993).

43
Consequently, iron deficiency decreases T-cell subsets. In contrast, B-cell-
mediated immunity is unaffected by iron deficiency as it is less sensitive to
changes in iron homeostasis (Wintergerst et al. 2007).
In iron deficiency, neutrophils have a decreased activity of myeloperoxidase
and a reduction in migration to inflammation sites, with a consequent impaired
intracellular killing of bacteria (Field et al. 2002). Likewise, natural killer cells
may have a diminished killing activity in iron deficiency because they require
sufficient iron for differentiation and proliferation. In some iron supplementation
RCTs, lower morbidity scores have been reported in Indonesian anaemic school
children who received iron supplements compared to their anaemic counterparts
receiving the placebo (Chwang et al. 1988). Similarly, a lower incidence of upper
respiratory tract infections in iron-deficient schoolchildren in Sri Lanka receiving
iron supplements has also been noted (de Silva et al. 2003).
However, any alteration in cellular iron homeostasis may have unfavourable
functional consequences on the immune system. Even iron overload can result in
diminished proliferation of T-cells and impaired phagocytic capacity of
neutrophils and macrophages (Weiss 2002). In the presence of iron, less nitric
oxide is formed, which is an essential effector molecule of macrophages to fight
infectious pathogens (Theurl et al. 2005). Furthermore, there is some evidence
that iron supplementation of iron-replete children living in areas with high rates of
malaria and other infections may be harmful (Oppenheimer 2001; Iannotti et al.
2006), not only because iron is necessary for the growth of iron-dependent
pathogenic bacteria (Ratledge 2007; Prentice et al. 2008), but also because iron
modulates the effect of the Th-1 mediated immune response, which in turn affects
iron homeostasis (Weiss 2002).

2.8.4 Selenium

Selenium also influences the cellular-mediated component of the immune


system. Deficiency of selenium can result in reduced T-cell counts and impaired
neutrophil, macrophage, and natural killer cell activity (Arthur et al. 2003; Lyons
et al. 2004). Furthermore, selenium plays a key role in redox regulation and

44
antioxidant function, mediated through certain selenoenzymes. Glutathione
peroxidases remove the potentially damaging lipid hydroperoxides, hydrogen
peroxide, and peroxinitrite produced during oxidative stress (Field et al. 2002;
Lyons et al. 2004). Likewise, the selenoenzyme thioredoxin reductase affects the
redox regulation of several key enzymes, which bind to DNA and activate
expression of genes encoding proteins (e.g., cytokines) involved in the immune
response (Ferenčík & Ebringer 2003).
Reductions in antioxidant function associated with chronic marginal
selenium intake have been associated with increased susceptibility to viral
infections (Lyons et al. 2004; Thomson 2004). Furthermore, in cases of selenium
deficiency, it appears that harmless viruses may undergo mutations to more
virulent forms with higher pathogenicity (Wintergerst et al. 2007). For example,
even in the presence of marginal selenium deficiency, influenza viruses can cause
more serious lung pathology (Beck 1999) and HIV infection can progress more
quickly to AIDS (Baum & Shor-Posner 1998).

2.9 Pro-poor policies in Brazil for the prevention of


malnutrition during childhood
Childhood malnutrition has often been associated with poverty in Brazil
(Issler et al. 1996; Victora et al. 2000), a country with a long history of
widespread regional and social health inequalities (Paim et al. 2011). However,
numerous pro-poor measures have been implemented over the past three decades
in an attempt to improve the nutritional status of children under-five years of age
in Brazil. These include mandatory fortification of flour with iron and folic acid,
free childhood vaccinations, vitamin A supplementation for children aged under-
five years, improved urban sanitation and sewage programmes, a conditional cash
transfer programme for poor families, universal primary education, and the
establishment of daycare centres for preschool children.

45
2.9.1 Wheat and corn flour fortification with iron and folic acid

Mandatory iron and folic acid fortification of all wheat and corn flours was
introduced at the national level in 2004. This strategy aimed to provide the poor
with widely available and affordable fortified products in an effort to improve the
population’s iron and folate status. Little is known about the impact of the flour
fortification programme on iron status in preschoolers, although Assunção et al.
(2007) found no change in the prevalence of anaemia at one- and two-years post-
fortification in children aged less than five years in southern Brazil.
Similarly, data are lacking on the effect of mandatory folic acid fortification
on serum and erythrocyte folate levels in preschoolers. Instead, research has
focused on the effectiveness of folic acid fortification to protect against neural
tube defects at birth (Pacheco et al. 2009; Orioli et al. 2011). However, concerns
have been raised recently about whether national fortification programmes with
folic acid may lead to inappropriately high folate concentrations in young
children, with potentially negative effects of unmetabolized folic acid (Smith
2010). Therefore, an assessment of the folate status of Brazilian preschoolers
consuming folate-fortified wheat and corn-flour products is warranted.

2.9.2 Childhood immunizations and vitamin A supplementation

Free routine childhood immunization programmes have been well-


established in Brazil since the 1980’s (Barreto et al. 2011). Brazil now produces
most of its own vaccines and provides immunizations through 30,000 health units
and additional temporary vaccination points bi-annually during national
vaccination days, with near universal coverage (Victora et al. 2011). Routine
universal vaccinations include BCG; poliomyelitis; measles, mumps and rubella
(MMR); diphtheria, pertussis and tetanus (DPT); Haemophilus influenza type b
(Hib); hepatitis B; yellow fever; rotavirus; 10-valent pneumococcal; and
meningococcal C conjugate vaccines (Barreto et al. 2011), with no cases of
poliomyelitis reported in Brazil since 1989, nor measles since 2000 (Paim et al.
2011).

46
Similarly, vitamin A supplements are also widely available in NE Brazil for
children aged less than five years (de Almeida et al. 2010). Nevertheless,
supplement distribution is irregular and coverage is low, particularly in the under-
resourced state of Bahia (Martins et al. 2007). Little is known about the effect of
vitamin A supplementation on the nutritional status of preschoolers in this region,
with conflicting evidence on growth outcomes, as discussed in section 2.6.5.
However, other studies in NE Brazil have reported vitamin A supplementation
may be protective against diarrhoea (Walser et al. 1996) and certain parasite
infections (Lima et al. 2010). It is possible vitamin A supplementation may induce
an enhanced cellular immune response, although such responses seem to vary
depending on the type of parasite (Long et al. 2006). Certainly, studies in other
countries have suggested children may benefit less from vitamin A
supplementation when administered in the presence of co-existing parasites,
particularly helminths (Long et al. 2006; Rosado et al. 2009). These findings
highlight the need for deworming treatment in conjunction with vitamin A
supplementation to achieve the maximum health benefits in at-risk preschoolers
(Tanumihardjo et al. 2004).

2.9.3 Sewage and sanitation programmes, specifically in Salvador, NE Brazil

To date, Brazil does not have a population-wide programme to control


parasitic infections, except for schistosomiasis (Katz 1998; do Amaral et al.
2006). Instead, Brazil has focused on the implementation of sanitation and sewage
programmes as a measure to decrease the presence and transmission of soil-
transmitted helminths and Giardia intestinalis infections, commonly found in
crowded urban cities in NE Brazil, such as Salvador. Certainly, studies conducted
in Salvador have highlighted the significantly higher prevalence of helminth
infections in school-aged children living in communities without public sanitation
(Moraes et al. 2004). Further, the major risk factors for giardiasis identified in
preschool children in Salvador have included lack of a household toilet and public
rubbish collection, and the presence of visible sewage nearby (Prado et al. 2003;
Matos et al. 2008).

47
A city-wide sanitation programme was implemented in Salvador between
1997 and 2003, during which time a decrease in the prevalence of infections of
Ascaris (i.e., 24% to 12%), Trichuris (i.e., 18% to 5%) and Giardia (14% to 5%)
in children aged less than four years was reported (Barreto et al. 2007; Barreto et
al. 2010). Unfortunately, fewer sewage connections were implemented in areas
where the prevalence of helminth infection was the highest because of the
financial constraints of those living in greater poverty (Barreto et al. 2010).

2.9.4 Conditional cash transfer programme

Low-income families in the NE region of Brazil are more than twice as


dependent on the conditional cash transfer (CCT) programme, Bolsa Familia, than
the poor living in wealthier regions of the country (de Bem Lignani et al. 2011).
Bolsa Familia seeks to alleviate poverty by providing greater purchasing power
for these families and promoting increased household food security, with the cash
transfer paid directly to the primary female in the family group (Paes-Sousa et al.
2011). Some Latin American CCT programmes, such as the Mexico’s
Oportunidades, have shown modest improvements in childhood growth and
anaemia rates (Rivera et al. 2004; Leroy et al. 2008). However, the impact of
Bolsa Familia on the anthropometric status of preschoolers in Brazil is less clear,
and their effect on anaemia has not been evaluated (Leroy et al. 2009). For
example, Paes-Sousa et al. (2011) found that children aged less than five years
living in CCT-participating households had an increased chance (i.e., 26% higher
chance) of having adequate height-for-age and weight-for-age than children not
exposed to the programme. Yet Morris et al. (2004) reported lower weight gains
in preschoolers of CCT beneficiaries. The latter finding may be attributed to the
misconception that eligibility in the programme depended on a child remaining
malnourished.
It is of interest that in NE Brazil, the decrease in child undernutrition over
the past decade has been associated with the increased purchasing power of
extremely poor households (de Lima et al. 2010). Yet the changes in dietary
patterns reported by Bolsa Familia participants in this region, the most food

48
insecure households in Brazil, are alarming. Increases in the consumption of fat,
sugar, soft drinks, and coffee have been reported, particularly of processed food
and high-density energy-rich foods, probably because they are affordable,
appealing in taste, and easy to prepare (de Bem Lignani et al. 2011). Some
investigators have suggested Bolsa Familia recipients should receive nutrition
education and capacity strengthening by learning new skills in order to adopt
healthy eating behaviours (Lindsay et al. 2009). However, to date, only cash
transfers are given.

2.9.5 Maternal education and preschool daycare centres

Improvement in maternal education through the provision of universal


primary schooling over the past two decades has also been identified as
contributing to the decline of child undernutrition in Brazil (Monteiro et al. 2009).
Indeed, functional illiteracy (i.e., less than four years of formal education) has
decreased from more than 60% to less than 20% between 1986 and 2006,
respectively, in the NE region (de Lima et al. 2010; Paim et al. 2011).
In 1996, an early childhood education law was established which legislated
care and education for Brazilian children from birth to six years of age (Brazil
Ministry of Health, 1996a). This law was introduced because of the evidence that
early childhood education programmes can greatly assist disadvantaged children
in reaching their developmental potential (Caughy et al. 1994; Barnett 1998;
Votruba-Drzal et al. 2004; Engle et al. 2007). Nevertheless, these programmes
have not been fully implemented. As a result, only half of Brazilian children aged
four to six years of age attend daycare (Freitas et al. 2008), with less than 30% of
the poorest children enrolled (Marins & Rezende 2011). Furthermore, evidence
for the nutritional benefits of daycares between regions is conflicting. For
example, da Silva Pereira et al. (2010) reported a decrease in undernutrition and
anaemia in preschool daycare attendees in SE Brazil. However, Gurgel et al.
(2005) found an increased risk of parasite infection in those attending public
daycares in NE Brazil as a result of classroom overcrowding and lack of hygienic
conditions in the daycares. Additionally, few studies have evaluated the nutrient

49
adequacy of the daycare meals (de Azevedo et al. 2010; Gomes et al. 2010), and
whether they meet the food-based dietary guidelines for feeding children over two
years of age, as recommended by the Ministry of Health (Brazil Ministry of
Health 2007b).

2.10 Conclusion
In conclusion, preschool children living in low-income households in the
under-resourced NE region of Brazil are very likely to be susceptible to the
adverse health risks associated with poverty. These may include an inadequate
supply of selected key micronutrients in their diets, and a heightened exposure to
parasitic infections. As a result, the micronutrient status of the preschoolers may
be compromised, with the potential to impair their immune function, thus
perpetuating a negative cycle, with long-term adverse consequences on growth
and health outcomes. Moreover, in cities such as Salvador in NE Brazil, where a
high percentage of the population are of African or African/Brazilian descent, the
presence of genetic Hb disorders may be an additional factor with a negative
impact on the nutritional status of these children. Brazil has implemented
numerous pro-poor policies in recent years, yet little is known about their
influence on the growth, micronutrient status, and morbidity of disadvantaged
preschoolers. Hence, a comprehensive investigation of the nutritional status of
disadvantaged preschoolers attending philanthropic daycares in Salvador would
provide some much needed information on the possible benefits of these pro-poor
policies.

50
3 Socio-economic and anthropometric status
of preschool children attending peri-urban
and urban daycares in Salvador, NE Brazil

3.1 Introduction
The prevalence of severe undernutrition has decreased in children less than
five years of age in many low- and middle-income countries over the past decade
because of the rapid nutrition transition (i.e., the change in dietary patterns
resulting in a shift from undernutrition to overnutrition), which is occurring
worldwide (Popkin 2002; de Onis et al. 2004a). These trends have been
particularly marked in Brazil (Stein et al. 2005; Monteiro et al. 2009), where
nationwide rates of stunting, underweight, and wasting have nearly halved
between 1997 and 2008 (Brazil Ministry of Health 1996b; WHO 2008). Instead,
the risk of overweight is an emerging problem among Brazilian preschool children
(Victora et al. 2010).
Despite the significant progress made by Brazil (Barros et al. 2010),
regional social inequities persist particularly between the rich south and the poor
north, with significant impacts on health outcomes for children living in these two
regions (Monteiro et al. 2009). The northeast of Brazil is one of the poorest and
most populous regions in the country (da Veiga et al. 2004; César et al. 2006),
with a large proportion of the population living without adequate housing and
sanitation (Assis et al. 2007). This region also has the highest rates of low
birthweight (Lira et al. 1998), and infant (Barros et al. 2010) and child mortality
(IBGE 2010a) in Brazil.
Yet, obesity rates in pre-adolescent children in the northeast are increasing
more rapidly than anywhere else in Brazil (da Veiga et al. 2004). This trend, like
undernutrition, is associated with several adverse health consequences. Young
children who become overweight are at high risk of adolescent and adult obesity

51
Figure 3.1 Study site location1

Study site

1
Reprinted from the World Factbook: Brazil (ISSN 1553-8133), under the public
domain license

52
and its associated risk of non-communicable diseases such as hypertension, type 2
diabetes mellitus, other cardiovascular disease risk factors, and increased
mortality (WHO Expert Consultation 2004; Baird et al. 2005), whereas children
with undernutrition are at high risk for impaired physical and mental
development, as well as increased morbidity and mortality (Sigman et al. 1991;
Schroeder & Brown 1994).
Salvador is the capital city of Bahia, the northeast state of Brazil, with the
largest African population of any city outside of Africa (IBGE 2010a). Of the
inhabitants, 80% are of African descent, many with little formal education and a
total family income of less than the minimum monthly salary (i.e., $415 Brazilian
Reis) (IBGE 2010a). Non-government daycare centres established in the urban
and peri-urban areas of the city provide care for preschool children living in these
poor socio-economic environments, with the aim of enhancing their growth and
development. However, few studies have investigated the nutritional status of
preschool children attending these non-government daycare centres. Several other
factors such as low birthweight, frequent infections, large family size, inadequate
breast-feeding, and poor complementary feeding practices during the vulnerable
period of transition from breastfeeding to the family diet, are also known to play a
major role in the subsequent growth and development of preschoolers (Issler et al.
1996; Grantham-McGregor & Ani 2002). Therefore, the present study aims to: (1)
assess the socio-economic status, health characteristics, early infant and young
child feeding (IYCF) practices, and growth and body composition of a pre-
selected group of preschool children aged three to six years attending urban and
peri-urban daycares in Salvador, NE Brazil; and (2) explore associations among
these variables.

3.2 Participants and Methods

3.2.1 Study sites and participants

This cross-sectional study was conducted in Salvador, the capital city of the
state of Bahia, located in NE Brazil (Figure 3.1) between August – November,

53
Figure 3.2 Cross-sectional study profile

7 selected philanthropic daycares agreed to participate


Available children
n=438

Consent Excluded
n=376 n=62

SES & health Q


n=376 Child moved
n=16

Anthropometry
n=361-364 Chronic illness
n=3

Faecal samples
n=325
Consent
refused
n=43
CBC n=359
Biochemistry n=355-358
Haemoglobinopathies
n=358 Non-participation
α3.7 thalassemia n=306 n=62

Completed the study n=376


Complete data sets obtained n=264

Abbreviations used: Q, Questionnaire; CBC, Complete blood count

54
2010. Salvador is a city with a population of three million which has a high degree
of social inequality, with high-rise apartments and crowded urban slum dwellings
closely interwoven throughout the city (Assis et al. 2007). Seven philanthropically
funded preschool daycares, one located within the city centre and six in peri-urban
areas, were pre-selected for participation in this study. The daycares in the peri-
urban areas were located approximately 20 kilometres north and north-east of the
city centre daycare. One peri-urban daycare was located within a large gated
complex with separate buildings for preschool daycare, primary and secondary
school. The other five peri-urban daycares were located within a radius of two
kilometers in one large favela community with a population of 65,000 (CMI
Brasil 2007), where residents lived primarily in permanently constructed houses
of concrete blocks or wood with cement, tile or earthen flooring. All daycares
were accessible by road, although the heavy rain during the wet season disrupted
city-wide transportation.
The children enrolled in the daycares (maximum class size of 25 children)
were from low-income families and attended full-time daycare (i.e., 8am–5pm)
five days per week, except holidays, until school age. All of the daycares provided
breakfast, a mid-morning snack, a main lunch meal, an afternoon snack, and a
milk drink or soup before the children went home. Filtered drinking water was
also available throughout the day. Further, flushable toilets, hand basins, and
shower facilities were provided in each daycare for the children. Age of enrolment
varied by daycare, with the peri-urban gated daycare enrolling children from six
months of age, the city centre daycare from two years of age, and the five peri-
urban favela daycares from three years of age.
Inclusion criteria for the study were apparently healthy children enrolled in
the three- and four-year old daycare classes for the 2010 school year (February to
December). Of the 438 eligible children, the caregivers of 376 children (i.e., 86%)
gave permission for participation in the study, with 99% (95/96), 86% (79/92) and
81% (202/250) recruited from the peri-urban gated daycare, city centre daycare,
and peri-urban favela daycares, respectively. From the latter, approximately 40

55
Table 3.1 Response rate of daycares

Daycare Daycare No. Roll Refusals Response Rate


(%)

"Santa Casa de Misericordia"


City centre 1 92 13 79/92 (86)
Peri-urban favela community 2 50 12 38/50 (76)
3 50 21 29/50 (58)
4 50 4 46/50 (92)
5 50 9 41/50 (82)
6 50 2 48/50 (96)
"Mansão do Caminho"
Peri-urban gated complex 7 96 1 95/96 (99)

Overall total 7 438 62 376/ 438 (86)

56
children per daycare were recruited for the study. Reasons for non-participation
included children on the roll who had already moved or were moving during the
year (n=16), children not attending daycare because of chronic illness (n=3), or
whose caregivers refused consent (n=43), as shown in Figure 3.2. It was not
possible to collect socio-demographic and anthropometric data from the families
who refused to allow their child to participate.
The study protocol was approved by the Human Ethics Committees of the
Federal University of Bahia, Salvador, Bahia, Brazil and the University of Otago,
Dunedin, New Zealand. Written approval to carry out the study was given from
the overarching philanthropic organizations: “Santa Casa de Misericordia” was
responsible for the city centre and five peri-urban favela daycares and “Mansão do
Caminho” was responsible for the peri-urban gated daycare. Informed written
permission to participate in the study was given by the parents or primary
guardians of the children.

3.2.2 Assessment of socio-demographic and health status

A culturally appropriate structured questionnaire (Appendix A) was


developed to collect information on the socio-demographic, health, and morbidity
characteristics of the participants. Trained Brazilian nutritionists administered the
pre-tested questionnaire to the mother or guardian of each participant at the
daycare. Data on parental education and occupation, sanitation, ownership of
possessions, and other household characteristics were obtained. From these data,
an overall socio-economic status (SES) score was developed for each participant
based on the model of Issler et al. (1996) designed to assess the poverty level of
Brazilian families in low socio-economic urban populations. Points were assigned
for family and housing size and structure, parental education and occupation,
marital status, ownership of house and household possessions, toilet and sewage
facilities, type of drinking water, availability of electricity, and susceptibility of
the house to flooding during heavy rain (Appendix B).
The ethnicity of the children was determined by skin colour, and hair and

57
Table 3.2 Socio-demographic characteristics of the children and households

n %
Sex (male) 196/376 52.1
Child ethnicity
Black 154/365 42.2
Brown 189/365 51.8
White 22/365 6.0
Father Education
No schooling 15/376 4.0
Primary school 178/376 47.4
High school 111/376 29.5
> High school 11/376 2.9
Unknown 61/376 16.2
Mother Education
No schooling 13/376 3.5
Primary school 212/376 56.4
High school 144/376 38.3
> High school 7/376 1.8
Father Occupation
Regular wage earner 170/376 45.2
Casual wage earner 125/376 33.2
Business or trade owner 11/376 2.9
Unemployed 24/376 6.4
Unknown 46/376 12.2
Mother Occupation
Regular wage earner 81/376 21.5
Casual wage earner 129/376 34.4
Business or trade owner 5/376 1.3
Domestic worker 35/376 9.3
Unemployed 126/376 33.5
Total income ( ≤ 1 min monthly salary1) 227/376 60.5
Mother Marital Status
Married / Partner 228/376 60.6
Divorced 13/376 3.5
Single 135/376 35.9

n Mean (SD)
Maternal Age (years) 368 29.7 (7.0)
Maternal Height (cm) 312 159.1 (6.7)
Maternal Weight (kg) 324 66.4 (15.4)
Maternal BMI 312 26.2 (5.7)
1
R$415 (equivalent to USD$205) = minimum monthly salary (IBGE 2010a)

58
facial characteristics, as performed by the 2010 national census (IBGE 2010a)
(see section 2.4.2). Information on health status variables and morbidity for each
child included maternal reports of deworming treatment within the past six
months, and the number and reasons for previous hospitalizations, including acute
respiratory infections and diarrhoea. Reports on the use of vitamin and / or
mineral supplements, including iron syrup within the previous six months was
also obtained. Review of the child’s health card, provided either by the caregiver
or the daycare, was used to verify and / or record information on vitamin A
supplementation, immunizations, birthweight and birth length. Information about
self-reported regular smoking in the home, either by the mother or another adult,
and history of asthma in the mother or siblings was also recorded.

3.2.3 Assessment of breastfeeding, infant and young child caring and feeding
practices

Information based on self-reports by the mothers or primary caregivers on


caregiver practices, child feeding, and related hygiene practices was collected
using the pre-tested questionnaire (Appendix A). Caregiver practice variables
included food preparation and feeding of the child, food-handling hygiene, hand
washing of both caregiver and child prior to mealtime, and specific practices
related to child illness (e.g., giving or withholding certain foods during or after
illness). Self-reported family food assistance received within the previous six
months was also recorded.
Questions on child feeding practices included a retrospective recall of early
IYCF practices. From these questions, selected indicators assessing optimal IYCF
practices were compiled, as specified by the WHO (2010) and Mukuria et al.
(2006). The following indicators were included: (i) early initiation of
breastfeeding; (ii) prelacteal feeding within the first three days of life; (iii) ever
breastfed; (iv) continued breastfeeding at one year; (v) continued breastfeeding at
two years; (vi) continued breastfeeding at three years; (vii) duration of
breastfeeding, and (viii) introduction to solid, semi-solid or soft foods at six
months of age.

59
Table 3.3 Health characteristics of the participants (n=376)

n %

Dietary supplements 204 54.3


Vitamin A 200 53.2
Iron syrup within 6 months 70 18.6

Vaccinations 315 83.8


BCG 281 74.7
Polio 291 77.4
DPT (Diptheria, Pertussis, Tetanus) 288 76.8
Measles 272 72.3
Rotavirus 197 52.4
Pneumococcus 7 1.9
Influenza A (H1N1) 133 35.4

Deworming treatment within 6 months 192 51.1

Hospitalizations 187 49.7


1-2 times 139/187 74.3
3 or more times 48/187 25.7

Reasons for hospitalizations


Diarrhoea 16/187 8.6
Upper respiratory infection 93/187 49.7
Other reasons 78/187 41.7

Smoking
Mother smoking in the house 47 12.5
Adult smoking in the house 89 23.7
History of asthma in mother or sibling 68 18.1

60
3.2.4 Anthropometric assessment

Anthropometric measurements, including weight, height, mid-upper arm


circumference (MUAC), and skin-fold measurements at two sites (triceps and
subscapular) were taken with children wearing light clothes and no shoes using
standardized techniques and calibrated equipment (WHO 2004a). Measurements
were taken in duplicate and a third measurement was taken if the difference
between the first two measurements was outside the maximum allowable
difference for that measure as recommended by the WHO (de Onis et al. 2004b)
(i.e., 100 g for weight, 7.0 mm for height, 5.0 mm for arm circumference, and 2.0
mm for both triceps and subscapular skinfolds). Weight was measured to the
nearest 0.1 kg using a weighing scale (Camry Mechanical Bathroom Scale,
China), levelled and calibrated daily. Standing height was measured to the nearest
0.1 cm using a portable stadiometer (O’Leary model STAT 7X, Ellard
Instrumentation Ltd, Monroe, WA USA). Mid-upper-arm circumference was
measured at the midpoint of the upper right arm with a fiberglass insertion tape
(Ross Lab, Columbus, OH USA). Skin-fold measurements were also taken on the
right side using precision calipers (Lange, Cambridge Scientific Industries Inc,
Cambridge, MA USA); the precise location of the measurement sites was taken
from Frisancho (1990). Maternal weight was measured to the nearest 0.1 kg, and
maternal height was measured to the nearest 0.1 cm, using a modified tape-
measure (Microtoise, CMS Weight Equipment Ltd, London, UK), fixed to a
portable wooden board. For all children, z-scores were calculated for height-for-
age (HAZ), weight-for-age (WAZ), weight-for-height (WHZ), and body-mass-
index (BMIZ) using the updated US Center for Disease Control and Prevention
(CDC) 2000 Growth Reference in the United States (Kuczmarski et al. 2000),
based on the method described by Vidmar et al. (2004). These growth reference
data were selected because of the high number of black and brown participants in
the study. Children were classified as stunted, underweight or wasted based on
anthropometric z-scores for height-for-age (HAZ), weight-for-age (WAZ) or

61
Table 3.4 Prevalence of past feeding practices of children and current
caregiver feeding practices (n=376)

n %
Past feeding practices:
Early initiation of breastfeeding1 145 40.4
Prelacteal feed within first 3 days of life2 68 18.1
Ever breastfed 357 95.0
Breastfed at 12 months 198 52.7
Breastfed at 24 months 139 37.0
Breastfed at 36 months 64 17.0
Duration of breastfeeding (months) 370 16.3 (13.7)4
Introduction of solid, semi-solid or soft foods3 162 43.1

Current caregiver feeding practices:


Preparation of child's food
Mother 323 85.9
Care-giver 51 13.6
Sibling 2 0.5
Feeding of child
Mother 292 77.7
Care-giver 40 10.6
Sibling 7 1.9
Child self feeds 37 9.8
Caregiver covers food until it is eaten 354 94.2
Caregiver washes raw food before giving to child 372 98.9
Caregiver washes own hands before feeding child 341 90.7
Caregiver washes child's hands or face before eating 321 85.4
Caregiver gives special foods when child is ill 112 29.8
Caregiver withholds certain foods when child is ill 92 24.5
Caregiver giving special foods after child's illness 40 10.6

All IYCF practices are based on WHO indicators for assessing IYCF practices (WHO 2010),
unless otherwise stated.
Proportion of children:
1
Put to the breast within one hour of birth
2
Given liquids or foods other than breast milk prior to the establishment of regular breastfeeding
(Mukuria et al. 2006)
3
Introduced solid, semi-solid or soft foods at 6 months of age
4
Mean (SD)

62
weight-for-height (WHZ) of < -2 SD. Mildly stunted, underweight or wasted was
classified as < -1 to ≥ -2 SD. Overweight or obesity in the children was based on
> 1 to ≤ 2 SD, or > 2 SD body-mass-index z-score (BMIZ) or WHZ, respectively.
Mid-upper arm fat area (AFA) and mid-upper arm muscle area (AMA) for the
children and BMI for the mothers / caregivers were calculated using standard
equations (Frisancho 1990).

3.2.5 Statistical analysis

Selected characteristics of the children, households, and their parents are


presented as percentages (%) for categorical variables and means with SD for
continuous variables. The daycares were divided into three groups (i.e., city
centre, peri-urban favela daycares, and peri-urban gated daycare), and compared
for the following variables using Fisher’s exact test: age group, sex, ever
breastfed, introduction of solids, dietary supplements including iron syrup within
the past six months and vitamin A supplementation, deworming treatment within
the past six months, and childhood vaccinations (details shown in Appendix C).
Overall SES scores developed according to Issler et al. (1996) were divided into
two categories – extremely low and low SES, based on scores ≤ 34 and ≥ 35,
respectively. Mean (SD) anthropometric variables for the children were
summarized by sex. Differences for age, sex, and socio-economic status for the
anthropometric variables of the children and mothers were analyzed using
multiple regression, adjusted by daycare clustering. The sandwich estimator was
used to obtain robust standard errors, to account for the sampling procedure.
Statistical analyses were carried out using STATA version 11 (Stata Corporation,
College Station, TX USA).

3.3 Results

3.3.1 Socio-demographic and health characteristics

The participation response rates of the daycares are presented in Table 3.1.

63
Table 3.5 Mean (SD) z-scores and anthropometric measures by sex for the children1 (n=361-364)

Difference (95% CI)


Variable n Females n Males p
(males – females)

Birth weight (g) 104 3055.5 (573.4) 123 3215.4 (633.0) 159.9 (-38.8, 358.6) 0.099

Birth length (cm) 88 47.9 (3.6) 98 49.4 (2.5) 1.5 (-0.2, 3.1) 0.071

Height-for-age z-score (HAZ) 171 0.60 (1.03) 193 0.37 (1.08) -0.23 (-0.42, -0.05) 0.020

Weight-for age z-score (WAZ) 171 0.21 (0.99) 193 0.07 (1.05) -0.15 (-0.28, -0.02) 0.034
64

Weight-for-height z-score (WHZ) 170 -0.17 (1.01) 193 -0.29 (1.06) -0.12 (-0.23, -0.001) 0.047

Body-mass-index z-score (BMIZ) 171 -0.11 (1.12) 193 -0.33 (1.17) -0.23 (-0.37, -.08) 0.007

Mid-upper arm fat area (AFA) (cm2) 169 8.01 (2.89) 192 6.75 (2.03) -1.26 (-1.76, -0.77) 0.001

Mid-upper arm muscle area (AMA) (cm2) 169 15.88 (2.37) 192 16.14 (2.08) 0.26 (-0.09, 0.60) 0.120

1
Adjusted by age and daycare cluster
Of the 376 participants, 196 were males and 180 were females (Table 3.2) who
ranged in age from 3.00 to 5.65 years. The mean (SD) age of the children was 4.2
years (0.61). One-third of the children (127/376) were of high birth order (i.e.,
third- or subsequent-born). The ethnicity of the participants was 42.2% (154/365)
black, 51.8% (189/365) brown, and 6.0% (22/365) white (Table 3.2). Of the
children with a known birth weight, 14% (32/227) had a birth weight of less than
2500g.
Socio-demographic characteristics of the households are shown in Table
3.2. Primary school was the highest level of education for 56.4% of the mothers
and 47.4% of the fathers; less than 3% of the parents were educated beyond high
school. One-third of the mothers were unemployed, with another third working as
casual wage earners. Nearly 40% of the mothers / caregivers were single or
divorced, 16% did not know the paternal education level, and 12% did not know
the father’s occupation. Overall, 60.5% of the households received the minimum
monthly salary of $415 Brazilian Reis (i.e., USD$205) or less. The mean (SD) age
of the mothers was 29.7 (7.0) years, with a mean (SD) height, weight, and BMI of
159.1 (6.7) cm, 66.4 (15.4) kg, and 26.2 (5.7) kg/m2, respectively. Of the mothers
/ caregivers, 14% consumed two or fewer meals per day, based on self-reports.
Health characteristics of the participants are shown in Table 3.3, with
differences by daycare presented in Appendix C. Over half of the children
(54.3%) had ever been given dietary supplements, with 19% having received iron
syrup within the past six months and 53% vitamin A capsules at some time. More
children attending the city centre daycare had received dietary supplements
(p=0.008) compared with the other daycares, although vitamin A supplementation
per se was more common among children attending the five peri-urban favela
daycares (p<0.001). Of the children, 51% received deworming treatment within
the previous six months, with treatment being more frequent among the children
at the peri-urban gated daycare (p=0.004). Most of the children received routine
childhood vaccinations (83.8%), although coverage was higher (p<0.001) among

65
Table 3.6 Differences for age, sex and socioeconomic status for the anthropometric measures of the children shown by
ß-coefficient (95% CI)

Change in age for a


Difference between sex1 Difference between SES1
Variable n one year increment1 p p p
(males-females) (extremely low- low SES)
(older-younger)

Child's birth weight (g) 227 n/a 160.38 0.096 -145.29 0.068
(-36.89, 357.65) (-304.53, 13.95)
Child's birth length (cm) 186 n/a 1.47 0.074 0.00 0.994
(-0.18, 3.13) (-1.17, 1.17)
HAZ 364 -0.06 0.548 -0.22 0.021 -0.17 0.089
66

(-0.28, 0.16) (-0.40, -0.04) (-0.38, 0.03)


WAZ 364 -0.19 0.066 -0.12 0.042 -0.35 0.001
(-0.40, 0.02) (-0.24, -0.01) (-0.50, -0.20)
WHZ 363 -0.19 0.054 -0.09 0.107 -0.37 0.004
(-0.39, 0.00) (-0.21, 0.03) (-0.57, -0.16)
BMIZ 364 -0.12 0.270 -0.20 0.013 -0.41 0.002
(-0.35, 0.11) (-0.34, -0.06) (-0.61, -0.21)
AFA (cm2) 361 0.28 0.252 -1.24 < 0.001 -0.47 0.011
(-0.25, 0.82) (-1.71, -0.76) (-0.80, -0.14)
AMA (cm2) 361 0.62 0.021 0.30 0.063 -0.81 0.004
(0.13, 1.12) (-0.02, 0.63) (-1.26, -0.36)
1
Parameters adjusted for other terms in model
Abbreviations used: SES, Socioeconomic status; n/a, Not applicable
the children attending the five peri-urban favela daycares. Of the children, 50%
had been hospitalized at least once; 25.7% were hospitalized three or more times.
Upper respiratory infections accounted for almost half (49.7%) of the
hospitalizations, whereas hospitalization for diarrhoeal episodes was much less
frequent (8.6%). Nearly a quarter of the households (23.7%) reported an adult
regularly smoking in the house, with 12.5% of the mothers smoking indoors.

3.3.2 Infant and young child caring and feeding practices and current dietary
patterns

Past feeding practices, based on the WHO indicators for assessing IYCF
practices, are shown in Table 3.4, with specific differences by daycare presented
in Appendix C. Almost all the children (95%) had been breastfed at some time. Of
these, 53% had been breastfed for one year of age, 37% for two years, and 17%
for three years. A quarter of the mothers reported stopping breastfeeding when
their child entered daycare. Solid, semi-solid or soft foods were introduced to
43.1% of the children from six months of age, although more children at the peri-
urban gated daycare had received solids at less than six months of age compared
with the other daycares (p=0.001).
Current caregiver feeding and hygiene practices are also shown in Table 3.4.
Over 90% of the mothers reported covering food, and washing both raw food and
their own hands before feeding their child. These practices were independent of
mother’s education level or the child’s birth order.

3.3.3 Anthropometry

Table 3.5 presents the mean (SD) anthropometric z-scores and indices of the
children and differences by sex. Males had significantly lower z-scores for height-
for-age, weight-for-age, weight-for-height, and BMI and AFA. Differences for
age, sex, and socio-economic status for the anthropometric variables of the
children and mothers are shown in Table 3.6. No significant differences were
observed by age group, except AMA (p=0.021), with the older children having a

67
Table 3.7 Differences between socioeconomic status for the anthropometric measures of the mothers shown
by ß-coefficient (95% CI)

Difference between SES1


Variable n p
(extremely low SES – low SES)

Maternal age (years) 368 -0.69 (-2.15, 0.77) 0.301

Maternal height (cm) 312 -0.48 (-2.0, 1.04) 0.480

Maternal weight (kg) 324 -3.93 (-7.66, -0.20) 0.041


68

Maternal BMI 312 -1.51 (-3.0, -0.05) 0.044

1
Adjusted for daycare clustering
greater mid-upper-arm muscle area.
Very few children (1.9%; 7/364) were stunted (HAZ < -2 SD), although
5.5% (20/364) had a HAZ < -1 to ≥ -2 SD. Height-for-age z-scores were
positively associated with breastfeeding at 12 months of age (p=0.007) but not at
24 months or 36 months of age. This relationship was independent of mother’s
education level.
Less than 2% (7/364) of the children were underweight (i.e., WAZ < -2
SD), but 12.1% (44/364) were mildly underweight (i.e., WAZ < -1 to ≥ -2 SD).
Likewise, few children (5.2%; 19/363) were wasted, but 15.7% (57/363) were
mildly wasted (i.e, WHZ < -1 to ≥ -2 SD). Overweight in the children was 10.7%
(39/364), based on a BMIZ >1 to ≤ 2 SD; 2.5% (9/364) were obese (i.e., BMIZ >
2 SD). Of the mothers or primary caregivers, 32.1% (100/312) were overweight
(BMI ≥ 25 to < 30) and 21.5% (67/312) were obese (BMI ≥ 30).

3.3.4 Inter-relationships among variables and household poverty

Based on the overall SES score, 48.4% (182/376) of the households were
classified as having an extremely low SES and the remainder a low SES. No
significant difference existed between the two SES groups by daycare location or
ethnicity. However, mothers in the extremely low SES group had more children
(p=0.003) (results not shown), who tended to have a lower birth weight (p=0.068)
than those in the low SES group (Table 3.6). Furthermore, mothers or the primary
caregivers in the extremely low SES group reported they had fewer meals (two or
less) per day (p=0.024) (results not shown), although child caring and feeding
practices did not differ between the two SES groups. Likewise, there were no
significant differences between the two SES groups for dietary supplement use
and vaccinations given to the children, or for the number or reasons for
hospitalizations. However, smoking was more frequent in the extremely low SES
group, with more mothers or caregivers smoking (p=0.001), and more adults
routinely smoking in the house (p=0.039) compared to the low SES group.
Children in the extremely low SES group had significantly lower (p<0.05)
WAZ, WHZ, BMIZ, AFA, and AMA than the children in the low SES group

69
(Table 3.6). Similarly, mothers in the extremely low SES group had significantly
lower weight (p=0.041) and BMI (p=0.044) than those in the low SES group
(Table 3.7).

3.4 Discussion
The preschool children studied here all lived in low-income environments in
urban and peri-urban areas of Salvador, NE Brazil and consumed most of their
food during the week in the daycare. Nevertheless, risk of mild-to-moderate
undernutrition still persisted, especially among the male children and children
living in households classified as extremely poor. Our findings also highlight that,
even in these settings, there is a tendency for some children to become
overweight, especially for those living in slightly improved circumstances.

3.4.1 Prevalence of undernutrition

None of the children in our study were severely undernourished. Prevalence


estimates for mild undernutrition, based on mild underweight and wasting (but not
mild stunting), were comparable to toddlers attending daycares in São Paulo,
southern Brazil, where household poverty levels are not as low (de Novaes
Oliveira et al. 2010). Much higher rates of both stunting and underweight than
those reported here have been reported in earlier studies of preschool Brazilian
children from poor families (Issler et al. 1996; Post & Victoria 2001; César et al.
2006), emphasizing that improvements have been made even among the lower
SES households. These improvements have been attributed to expansions in
healthcare and pro-poor social policies, and increases in both purchasing power
and maternal schooling among poor families (Monteiro et al. 2009; Barros et al.
2010). Indeed, it is likely the NE region will meet the first United Nations
Millennium Development Goal of reducing the number of underweight children
under-five years of age by half between 1990 and 2015 (de Lima et al. 2010), as
has been reported for Brazil as a whole (Monteiro et al. 2009).

70
The boys in our study, irrespective of level of poverty, were shorter, lighter,
with a lower content of body fat (based on AFA) than the girls. Such sex-related
differences in growth have been reported by others elsewhere (Espo et al. 2002;
Marriott et al. 2010), and linked in part to the faster growth rates of males during
early childhood and thus higher requirements for energy (Butte et al. 2000) and
nutrients, most notably zinc (Cavan et al. 1993; Ruz et al. 1997; Gibson et al.
2007). The latter plays a critical role in the growth and development of young
children because of its role in cell proliferation (MacDonald 2000).

3.4.2 Association between undernutrition and extreme poverty

Levels of poverty were a major factor influencing the anthropometric status


of these children. Children living in households classified as extremely poor were
more underweight and wasted, with significantly lower BMI z-scores, and
anthropometric indicators of muscle mass and total body fat, than their
counterparts in the less impoverished households. This finding is not unexpected.
Studies in Salvador and elsewhere in Latin America have shown that children
over two years of age living in crowded households, with a low monthly per
capita income, have significantly increased risks of mild-to-moderate
undernutrition, based on weight-for-age deficits (Pelto et al. 1991; Assis et al.
2007). These differences were again accentuated in the boys, and existed even
though all the children, irrespective of daycare, were provided with the same
foods during the week by the daycare: weekly menus were the same across all the
daycares studied here.
These findings suggest that the children from the extremely impoverished
households were probably experiencing greater deficits in the quantity and / or the
quality of the food they consumed in the weekends and vacations compared with
those in the less impoverished households. Certainly, mothers of the children in
the extremely impoverished households reportedly ate fewer daily meals, had a
lower mean body weight and BMI (p<0.05) (Table 3.7), and more children to
feed, than their counterparts in the less impoverished households. Further,
consumption of expensive cellular animal protein by the children in these low

71
income homes was probably beyond their purchasing power, given that 60% of
the households had a total income below the estimated minimum monthly salary
(IBGE 2010a). Cellular animal foods are rich sources of zinc, haem iron, and
vitamin B-12, micronutrients which have been associated with improvements in
growth (Allen et al. 1992) and muscle mass (Grillenberger et al. 2006; Neumann
et al. 2007) in studies of children in other countries. Low birthweight may have
been an additional factor contributing to the moderate undernutrition of the
preschool children living in extreme poverty, although data on birthweight was
limited. Low birthweight infants have higher nutrient requirements because of
their smaller hepatic reserves for certain nutrients at birth and higher growth rates.
The higher rates of maternal smoking in the extremely poor households, together
with poor nutritional status of the mothers during pregnancy and lactation,
probably contributed to the risk for low birthweight (Horta et al. 1997; Dijkhuizen
et al. 2001; Belkacemi et al. 2010).

3.4.3 Association between overnutrition and less extreme poverty

It is of interest that even among the children studied here who all lived in
poor environments, those from the households classified as less impoverished,
were more at risk of overweight and obesity than their counterparts in the
extremely impoverished group (17.1% vs. 9.0%; p=0.029), especially the girls.
This same trend has been reported among adolescent girls in a shanty town in the
city of São Paulo, Brazil (Sawaya et al. 1995) and elsewhere (Wang et al. 1998).
Moreover, this trend was also evident among the mothers: less impoverished
mothers had a significantly higher BMI than those who were extremely
impoverished (Table 3.7), with 32% classified as overweight and 22% as obese.
These findings are of concern because a high maternal BMI has been associated
with the increased risk of overweight and obesity in children (Fernald & Neufeld,
2007; Balaban et al. 2010). Further, they support the trend that the burden of
disease has shifted among low-income Brazilian women from under- to
overnutrition during the last three decades (Monteiro et al. 2004; Stein et al. 2005;
Goulet et al. 2006). This is unfortunate because these low-income women in

72
particular lack not only the resources to purchase better quality food, but also the
information and understanding to improve feeding practices for their young
children. Clearly, preventive measures such as nutrition education strategies
designed specifically for this population, including appropriate incentives, must
be implemented to avoid a generational cycle of overweight and obesity in these
households (Vereecken et al. 2004; Lindsay et al. 2008).

3.4.4 Association between anthropometry and retrospective child feeding


practices

The positive association found here between HAZ-scores and breastfeeding


at one year of age, but not at two or three years, is not surprising, as children
beyond 12 months of age receive most of their nutrient needs from food, rather
than breastmilk. Further, the lack of other relationships between anthropometry of
the preschool children and past IYCF practices in this study was consistent with
those of Marriott et al. (2010) for Cambodian children based on self-reports from
the Demographic and Health Survey. Like our findings, relative wealth was also
associated with a reduced risk of underweight and stunting in this Cambodian
study. Whether the foods consumed by these preschool children from Salvador
during the earlier complementary feeding period were adequate to meet their
energy and nutrient needs and support physiological growth and development was
not investigated here. Breastfeeding patterns for these poor households were
similar to reports elsewhere in Brazil (Romulus-Nieuwelink et al. 2011). We did
not find any relationship between breastfeeding practices and maternal education
or employment, unlike an earlier study in Timor-Leste (Senarath et al. 2007),
possibly because of the limited range of schooling among the mothers compared
to the Timor-Leste study.

3.4.5 Strengths and limitations

We recognize that our study has some strengths and limitations. Our
findings are based on a cross-sectional study so that causal inferences cannot be
made from the associations reported here. The restriction of the daycare selection

73
to philanthropically funded preschool daycares limits extrapolation of the results
to all preschool children attending daycares in the urban and peri-urban slums of
Salvador. Nevertheless, we had a large sample size and high response rate. We
also checked the health data with clinic cards and collected anthropometry data on
the mothers and the children. Further, we expressed the past IYCF practices using
WHO indicators in order to compare data across countries. Finally, the mean BMI
of the children examined here was comparable to that for preschool children 3.00-
5.99 years of age living in the state of Bahia, NE Brazil (i.e., 15.3) reported in the
most recent Brazilian National Nutrition survey (IBGE 2010b).

3.4.6 Conclusion

In conclusion, our results suggest that preschool children living in


disadvantaged environments in Salvador may be nutritionally vulnerable to the
adverse health consequences arising from both the immediate risks of
undernutrition and those associated with the chronic risks of overnutrition,
depending on the social strata of their households. Analysis of the dietary intakes
from the foods and beverages provided by the daycares would provide a greater
understanding of the adequacy of the energy and nutrient supply and their
potential contribution to the nutritional status of these children.

This chapter has been accepted for publication in a modified format as:
Factors associated with growth and intestinal parasitic infections in
preschoolers attending philanthropic daycares in Salvador, Northeast Brazil
Rebecca L. Lander, Alastair G. Lander, Hugo C. Costa-Ribeiro,
Danile L. Barreto, Angela P. Mattos, Sheila M. Williams,
Lisa Houghton, Rosalind S. Gibson
Cadernos de Saúde Pública (in press).

74
4 The adequacy of daycare meals provided
to preschoolers in Salvador, NE Brazil

4.1 Introduction
Brazilian charitable organizations have played an important role historically
in providing assistance to low-income families, particularly in the
socioeconomically depressed NE region of Brazil. Santa Casa de Misericordia
(‘Holy House of Mercy’), a philanthropic arm of the Catholic Church, began its
social service work in the city centre of Salvador, Bahia, in the 16th century
(c.1525) and continues to this day. In 2002, the city-based childcare programme
expanded to include five daycares located in Barros da Paz, a peri-urban favela
(slum) on the outskirts of Salvador. Similarly, Mansão do Caminho (‘Mansion of
the Way’) established a charitable work fifty years ago in Pau da Lima, another
peri-urban slum area of Salvador, providing daycare and schooling for local
children 0-18 years. Both of these non-profit organizations offer not only
education, school uniforms, and a stimulating learning environment, but they also
supply daily meals and snacks, five days a week, for the children of poor families.
Since 2004, the national pro-poor social policies of Brazil have included
mandatory fortification of all wheat and corn flour products with iron and folic
acid in an effort to enhance iron and folate status of the population (Darnton-Hill
et al. 1999; Brazil Ministry of Health 2002). This strategy is recommended for
Latin America by the Pan American Health Organization (PAHO) (2004) because
these fortified products are widely available and affordable staples in the diets,
and thus, accessible to the poor.
Brazil has also developed important key nutritional messages to promote
healthy eating for the general population (Sichieri et al. 2010) and to encourage
healthy eating in schools (Brazil Ministry of Health 2006). The Ministry of Health
has designed specific food-based dietary guidelines for feeding children over two
years of age (Brazil Ministry of Health 2007b), as recommended by PAHO

75
(PAHO/INCAP 1999) and the WHO (1998). Nevertheless, despite this new policy
development, few studies have examined whether the menus in the philanthropic
daycares comply with these guidelines, particularly for preschool children, and
quantified the energy and nutrients supplied by the daycare meals. Therefore, the
present study aims to: (1) assess the compliance of the daycare menus to national
food-based dietary guidelines for children over two years of age; (2) examine the
nutrient adequacy of the daycare menus, and (3) recommend strategies, where
necessary, to overcome any deficits in the energy and nutrients supplied by the
daycare menus for a pre-selected group of preschool children age three to six
years attending urban and peri-urban daycares in Salvador, NE Brazil.

4.2 Participants and Methods

4.2.1 Study sites and participants

This cross-sectional study was conducted in Salvador, the capital city of the
state of Bahia, located in NE Brazil between August – November, 2010. Seven
philanthropically funded preschool daycares located within the city centre and in
peri-urban areas were pre-selected for participation in this study (as described in
Chapter 3). The children enrolled in the daycares were from low-income families
and attended daycare all day five days per week, except holidays, until school age.
They received most of their food during the week while attending the daycare and
very little at home. The study protocol was approved by the responsible Ethics
committees and overarching daycare organizations, as described in section 3.2.1.

4.2.2 Comparison of the 20 daycare menus with food-based dietary


guidelines in Brazil

All of the daycares provided breakfast, a mid-morning snack, a main lunch


meal, an afternoon snack, and a soup or milk drink before the children went home.
Menus of all foods and beverages, including serving sizes, served to the children
in the daycare over 20 days were obtained from Santa Casa de Misericordia. The
frequency of compliance of the 20 daily menus to the Brazilian dietary guidelines

76
for children, “The ten steps to healthy eating for children over two years of age”
was determined, and expressed as a percentage (Brazil Ministry of Health 2007).

4.2.3 Assessment of the adequacy of the energy and nutrient supply from the
daycare menus

Energy and nutrients supplied by each of the 20 daily menus were calculated
by developing a menu-specific Brazilian food composition database. Nutrient
values were derived primarily from two Brazilian food composition tables
(FCTs), Tabela Brasileira de Composição de Alimentos (TACO), version 2
(NEPA/UNICAMP 2006) and Tabela de Composicao de Alimentos: suporte para
decisão nutricional, 2nd edition (Philippi 2002), augmented, where necessary, with
nutrient values for raw foods from the US Department of Agriculture (USDA)
National Nutrient Database for Standard Reference (2005) and the WorldFood
Dietary Assessment System (Bunch & Murphy 1997). For composite dishes, the
raw ingredients from the daycare recipes were used to calculate their energy and
nutrient content, adjusted for changes in nutrient retention and yield after cooking
by using appropriate retention (USDA 2007) and yield factors (Matthews &
Garrison 1975), where necessary. To estimate the potential bioavailability of zinc
and iron from the daycare menus, median (1st, 3rd quartile (IQR)) molar ratios of
phytate:zinc and phytate:iron were calculated, and available iron estimated using
the algorithm of Monsen and Balintfy (1982), as recommended by Beard et al.
(2007).
The nutrient values for iron for food products containing wheat and corn
flour in the Brazilian FCTs did not reflect the mandatory fortification levels for
iron (4.2 mg per 100 g wheat and corn flour). Consequently, the iron content of
these products was adjusted to reflect the additional iron contributed by the
fortified flours. In addition, many folate values were missing from the Brazilian
FCTs, so all folate values for the database were compiled from the USDA
database, and adjusted where necessary to take into account the additional

77
Table 4.1 Comparison of the frequency of compliance (as %) by daycare
menus to the dietary guidelines “Ten steps to healthy eating for children over two
years of age” for Brazil

Steps Healthy Eating Steps Menu compliance


Frequency %

1 Offer variety of vegetables 20/20 100


Offer vegetables in five daily serving sizes 0/20 0

2 Offer beans once a day four times a week 4/4 100


Offer source of vitamin C (e.g., orange juice) 13/16 81
with beans

3 Fatty foods served maximum once a week 4/4 100


Limited amounts of preferred oils (e.g., 19/20 95
soybean oil)

4 Moderate use of salt n/a

5 Three meals and two snacks offered per day 20/20 100
Fruit offered for snacks and dessert 20/20 100

6 Sweets, cakes and biscuits offered maximum 4/4 100


twice a week

7 Avoid daily consumption of soft drinks 20/20 100

8 Encourage child to eat slowly and chew food well n/a

9 Keep child’s weight within healthy limits for age n/a

10 Encourage child to be active n/a

n/a = Not applicable

78
contribution from the mandatory fortificant – synthetic folic acid – used in wheat
and corn flour products in Brazil (i.e., 150 µg per 100 g flour). Consequently, all
folate values were expressed as dietary folate equivalents (DFE), and calculated
using the following formula: food folate (µg/100 g) + (1.7 x µg synthetic folic
acid). This formula is based on two assumptions: (1) the bioavailability of food
folate is ~50%; and (2) the bioavailability of synthetic folic acid in fortified food
is ~85%. Thus, synthetic folic acid is 85/50 or 1.7 times more available than food
folate (Molloy 2002). The contribution of the fortified flour products to the total
supply of iron (mg/day) and folate (µg DFE/day) from the menus was also
estimated.
The adequacy of the usual provision of energy (kcal/day; kJ/day) and
selected nutrients supplied by the 20 menus for a child ranging in age from three
to six years was evaluated qualitatively by comparing the median (IQR) supply
with the estimated average requirements (EARs) for energy (FAO/WHO/UNU
2001) and nutrients (WHO/FAO 2006) and the recommended nutrient intakes
(RNIs) (WHO/FAO 2004), as recommended by Barr et al. (2002). The
contribution of eight food groups to the supply of energy and selected nutrients
from the daily menus was also determined.
In addition, the mean (SD) usual energy density (i.e., kcal/g) of the daily
menus, meals, and snacks were calculated, as well as the contribution of food and
beverages separately to the overall energy density. The percentage of energy
supplied from macronutrients (protein, carbohydrate, and fat) was calculated and
then compared to the population target range set for these nutrients by WHO
(2003) and FAO (2010).

4.2.4 Statistical analysis

All calculations for the analysis of the nutrient supply from the menus were
performed using the STATA statistical software package version 11.0 (Stata
Corporation, College Station, TX USA).

79
Table 4.2 Macronutrients (as percentage of energy) supplied by Brazilian
daycare menus compared to the population target goals

% of energy provided % of energy


from menu target goal

Protein 17 10-151

Fat 21 25-352

Carbohydrate 62 55-751

1
Specified by WHO (2003)
2
Updated recommendations for children 2-18 yrs from FAO (2010)

80
4.3 Results

4.3.1 Assessment of compliance of 20 daycare menus to dietary guidelines

The frequency of compliance (as %) by the 20 daycare menus, with the


Brazilian dietary guidelines for children over two years of age is shown in Table
4.1. Compliance with the guidelines was high, with the notable exceptions of
failing to offer vegetables in five daily serving sizes, and not providing a
consistent source of vitamin C with the servings of beans.

4.3.2 Assessment of the adequacy of the energy and nutrient supply from the
daycare menus

The carbohydrate and protein supplied by the menus, expressed as a


percentage of energy, met or exceeded the WHO (2003) target range for these
macronutrients, whereas the fat was below the desirable range specified by the
Food and Agriculture Organization (FAO) (2010) (Table 4.2). Table 4.3 presents
the median (IQR) supply of energy from the 20 daycare menus compared to the
FAO/WHO/UNU (2001) estimated energy requirements (EER), and highlights
that the median energy supply was below the EAR for a child ranging in age from
three to six years. In terms of mean (SD) energy density (i.e., kcal/g), the value
based on both foods and beverages supplied by the menus was 0.79 (0.14) kcal/g,
whereas the values based exclusively on either foods or beverages were 1.06
(0.16) and 0.57 (0.19) kcal/g, respectively. The energy densities for breakfast,
lunch, dinner, and all the snacks combined were 0.83 (0.31), 1.19 (0.16), 0.59
(0.45), and 0.50 (0.08) kcal/g, respectively.
Moderate bioavailability was assumed for the supply of iron and zinc
because the 20 menus had low levels of phytate, with median phytate to iron and
phytate to zinc molar ratios of 2.9 and 7.9, respectively, and a median (IQR)
supply of meat, poultry and organ meat of 80 (25, 120) g/day, vitamin C of 100
(70, 174) mg/day, and available iron of 0.75 (0.64, 0.94) mg/day (Table 4.4).
The median (IQR) supply of selected nutrients from the 20 daycare menus

81
Table 4.3 Supply of energy from Brazilian daycare menus compared to
estimated energy requirements (EER)1

Median (IQR) EER Percentage


energy (kcal/d) (kcal/d) EER

Boy age 3-4 y 814 (735, 875) 1250 64.4

Girl age 3-4 y 814 (735, 875) 1150 70.0

Boy age 4-5 y 814 (735, 875) 1350 59.6

Girl age 4-5 y 814 (735, 875) 1250 64.4

Boy age 5-6 y 814 (735, 875) 1475 54.5

Girl age 5-6 y 814 (735, 875) 1325 60.7

1
Based on age, sex, and moderate physical activity level from FAO/WHO/UNU
(2001)

82
compared to the WHO/FAO (2006) nutrient requirement estimates are shown in
Table 4.4. The median usual supply of iron, zinc, niacin, vitamin B-12, folate, and
vitamin C for a preschool child three to six years of age was above the RNI (Table
4.4), indicating the supply was likely to be adequate, based on the qualitative
approach of Barr et al. (2002). In contrast, because the median usual supply of
calcium for a child aged three to six years, and thiamine and vitamin A for a child
aged four to six years of age, was below the EAR (WHO/FAO 2006), there was a
risk that the supply was inadequate, with a probability of adequacy of 50 percent
or less. For the micronutrients with a median supply at a level between the EAR
and RNI (i.e., vitamin B-6, thiamine, vitamin A and riboflavin, depending on the
child’s age), the probability of adequacy for the supply of these micronutrients
was more than 50 percent but less than 97.5 percent for at least one of the age
groups specified (Barr et al. 2002). Hence, the supply of these micronutrients
probably needs to be improved, given the difficulty of establishing their exact
requirements.

4.3.3 Assessment of the major food sources of energy and selected nutrients

Iron and folate were provided primarily by legumes (61% and 35%,
respectively) and cereals (17% and 25%, respectively), both unfortified and
fortified (Table 4.5). The iron and folic acid fortificants in the wheat and corn
flour-based products contributed ~12% of the median daily iron supply (1.6
mg/day) and ~21% of the median daily folate supply (44 µg /day). Major food
sources of vitamin C were vitamin C-rich fruit juices (63%), followed by fruits
and vegetables (24%) (Table 4.5). Meat, poultry and organ meats supplied 11% of
the iron, 30% of zinc, 70% of the vitamin A, and 85% of the vitamin B-12. As
expected, dairy products provided the main source of calcium (i.e., 65%).

4.3.4 Strategies to overcome deficits in the nutrient supply from the daycare
menus

Menus were modified to overcome the observed deficits in the supply of fat,

83
Table 4.4 Average daily supply of nutrients for a child from daycare menus
(n=20) and factors influencing bioavailability of iron and zinc compared to the
estimated average requirements and recommended nutrient intakes1

Nutrient Median (IQR) EAR EAR RNI


supply from
Brazil menu 1-3 yrs 4-6 yrs 4-6 yrs

Calcium (mg) 388 417 500 600


(274, 398)
Iron (mg) 13.5 5.8 6.32 6.3
(12.5, 14.8)
Zinc (mg) 5.7 2.03 3.03 4.8
(5.0, 6.2)
Thiamine (mg) 0.41 0.4 0.5 0.6
(0.37, 0.55)
Riboflavin (mg) 0.53 0.4 0.5 0.6
(0.48 0.69)
Niacin (mg) 8.1 5 6 8
(7.1, 9.7)
Vitamin B-6 (mg) 0.48 0.4 0.5 0.6
(0.42, 0.65)
Vitamin B-12 (µg) 1.3 0.7 1.0 1.2
(1.07, 2.33)
Folate (µg DFE) 209 120 160 200
(184, 279)
Vitamin C (mg) 100 25 25 30
(70, 174)
Vitamin A (µg RE) 298 286 321 450
(221, 372)
Phytate (mg) 457
(375, 503)
Phytate:Zn molar ratio 7.9
Phytate:Fe molar ratio 2.9
Available Fe (mg) 0.75
(0.64, 0.94)
Meat, poultry, organ 80
meat (g/day) (25, 120)

1
From WHO/FAO (2006)
2
Assuming moderate bioavailability; the RNI value is used because of the skewed
distribution of requirements for iron by young children (WHO/FAO 2006)
3
IZINCG EAR and RDA recommendations for mixed diets (Brown et al. 2004)

84
calcium, vitamin A, and thiamine on the assumption that the food supplied by the
daycare provided most, but not all, of the daily supply of energy and nutrients in
the children’s diets. The observed calcium deficit could be overcome by
substitution of the daily fruit juice snack with 300 g whole milk (without added
sugar), irrespective of the age of the child. Further, the usual median supply of
thiamine and vitamin A could be adjusted so that it fell at least between the
corresponding EAR and RNI by increasing (a) the size of the main composite dish
served at lunch to not less than 125 g for a child aged 3-4 years, or not less than
200 g for a child aged 4-6 years, and (b) the portion sizes of the accompanying
dishes by 50%. These proposed changes would result in 25% of the energy being
provided by fat in the revised menus, as recommended by FAO (2010).

4.4 Discussion
To our knowledge, this study is the first to assess the compliance of menus
compiled for daycares in Brazil with their food-based dietary guidelines for
children over two years of age. Inclusion of the recommended number of meals
and snacks, servings of beans and / or animal source foods at the main meal each
day, fruit for snacks and desserts while excluding high intakes of fatty foods,
biscuits and soft drinks emphasize the positive efforts made to ensure that the
daycare menus comply with the food-based dietary guidelines for children over
two years of age (Brazil Ministry of Health 2007b), as well as the general
recommendations for kindergarten food supply (Brazil Ministry of Health 2006).

4.4.1 Adequacy of the energy and nutrient supply of the menus

The daycare menus failed to meet the estimated energy requirements


(FAO/WHO/UNU 2001) for a preschool child aged three to six years. Such
shortfalls in the usual energy supply by the daycare menus are consistent with the
findings of Gomes et al. (2010) who reported a deficit in energy intakes in
Brazilian preschoolers (n=647) attending philanthropic daycares. However, the fat
supply noted here was markedly lower (i.e. 50%) compared to the fat intake

85
Table 4.5 Mean percentages contribution of eight food groups to energy and nutrients supply from daycare menus

Nutrient Cereal/grains Meat/poultry/ Dairy Yellow/orange Roots Legumes Other fruit Juice
organ meat fruit & vegetables & tubers & vegetables

Energy 22.8 16.6 26.0 3.3 3.1 7.6 14.8 5.8


Calcium 6.3 2.2 65.4 2.5 1.6 15.2 5.6 1.2
Iron 16.5 11.2 3.4 2.3 1.8 60.7 3.4 0.6
Zinc 17.6 29.7 13.8 6.9 3.0 23.1 4.9 1.0
Thiamine 41.3 10.7 17.8 3.5 4.4 4.7 15.6 2.0
Riboflavin 12.3 18.7 48.2 4.3 2.3 2.9 8.8 2.5
86

Niacin 42.8 29.6 3.7 3.3 3.1 5.0 8.4 4.1


Vitamin B6 13.3 26.3 18.5 3.6 5.2 5.5 25.9 1.7
Vitamin B12 0.6 84.9 13.5 0.8 0.2 0.0 0.0 0.0
Folate 25.4 13.7 6.7 4.8 2.2 34.7 8.8 3.6
Vitamin C 0.7 0.8 1.1 7.3 1.7 1.6 24.1 62.7
Vitamin A 2.8 69.6 11.0 3.9 4.5 1.8 4.6 1.8
Phytate 24.9 4.1 1.7 1.3 4.0 61.3 1.5 1.2
reported by Gomes et al. (2010), and contributed a lower proportion of the energy
than the recommended level (FAO 2010). Based on self-reports from daycare
staff, children from extremely low SES households received nearly all their food
from daycare. Thus, for these extremely poor children, energy intakes were
probably inadequate, as reflected by their suboptimal growth, being more
underweight and wasted, with lower BMI z-scores, muscle mass and total body fat
(Chapter 3).
An adequate supply of fat in the daycare menus is necessary to promote the
absorption of fat-soluble vitamins (Jalal et al. 1998) (e.g., vitamin A) and provide
a sufficient supply of polyunsaturated fatty acids (PUFAs) for optimal child
growth (Barbarich et al. 2006) and immune function (Prentice & van der Merwe
2011), especially for the children receiving little additional food at home. Soybean
oil was the most commonly used source of fat in the daycare recipes, and is
recommended in moderate amounts in the dietary guidelines for Brazil, as it
provides a high level of PUFAs, especially α-linolenic acid (ALA, 18:3n-3). Only
one daycare recipe included palm oil, which contains very low levels of PUFAs
(Michaelsen et al. 2011).
The shortfall in energy supply noted here was attributed in part to the low
energy density of the daycare menus, estimated to be only 60–80% of that
reported in other studies of preschoolers (Kral et al. 2007; Leahy et al. 2008), with
the energy density supplied exclusively from foods being particularly low. This
finding is not surprising, not only because of their limited supply of fat, but also
because of the semi-liquid consistency of the dinners served and large amount of
beverages consumed. For example, beverages such as milk drinks and juice
contributed 32% of the total energy supply of the menus, despite their lower
energy density compared to that for solid foods (Kral et al. 2007).
The low energy density of the daycare menus may have been in response to
the rising concern of overweight and obesity among preschoolers in Brazil (Simon
et al. 2008; Balaban et al. 2010; Shoeps et al. 2011) and elsewhere (Fernald &
Neufeld 2007; Corvalán et al. 2008; Leahy et al. 2008). Certainly, some daycare
institutions in Chile and the US (Corvalán et al. 2008; Leahy et al. 2008) have

87
reduced the energy density of their meals by decreasing the sugar and the
nonessential fatty acid content and increasing the amount of fruit and vegetables.
The dietary diversity of the daycare menus was high, supplying micronutrients
from eight food groups, including animal-source foods and vitamin C and pro-
vitamin A-rich fruits and vegetables (Table 4.5). Hence, the supply of readily
absorbable iron and zinc, vitamin B-12, vitamin C, and niacin in the daycare
menus was adequate (Table 4.4), all micronutrients that play an important role in
promoting growth, cognition, and immune function in young children (Fraker et
al. 2000; Neumann et al. 2003; Arsenault et al. 2009).
Several characteristics of the daycare menus could potentially facilitate the
absorption of non-haem iron and / or zinc from the daycare meals. For example,
inclusion of animal-source foods in the main meal together with vitamin C-rich
fruits and vegetables was likely to improve absorption of both the iron fortificant
in the cereal flours and the non-haem iron and zinc in the accompanying legumes
(Teucher et al. 2004; Hunt 2005). Iron fortification resulted in low phytate to iron
molar ratios (i.e., < 6:1), and likely moderate bioavailability of iron from the
composite meals containing either meat or vegetables high in ascorbic acid
(Hurrell & Egli 2010). Finally, the practice by the daycares of soaking dried beans
overnight, and discarding the water before cooking has the potential to enhance
iron and zinc absorption through the reduction of water-soluble phytates
(Hallberg et al. 1989; Gibson et al. 1998), potent inhibitors of non-haem iron and
zinc absorption (Lönnerdal 2000; Hunt 2005).
Calcium and vitamin A were the nutrients most likely to be inadequate in
the daycare diet (Table 4.4). Calcium is a major structural component of bones
and teeth, but also acts as a regulatory agent in several metabolic processes (Lee et
al. 1993; Lanou et al. 2005), whereas vitamin A is important for young children
through its role in hematopoiesis, the immune system, and the integrity of
epithelial cells (Villamor & Fawzi 2005; West et al. 2007). Hence, some
modifications to the menus must be made to overcome these shortfalls, even
though dairy products, pro-vitamin A carotenoid-containing fruits and vegetables
(papaya, mango, carrots, and pumpkin), and cooked liver (once a week) were

88
regular components of the daycare menus. In contrast, there was an adequate
supply of iron and folate in the daycare menus, in part as a result of Brazil’s
mandatory wheat and maize flour fortification policy (Brazil Ministry of Health
2002). This supply was unlikely to exceed the upper tolerable levels for iron (i.e.,
40 mg) or folate (i.e., 400 µg DFE) for preschool children (WHO/FAO 2006).

4.4.2 Strategies to overcome observed deficits in nutrient supply from


daycare menus

To overcome the observed deficits in the supply of calcium, vitamin A, and


thiamine we recommend substituting a daily fruit juice snack with cow’s milk and
providing larger serving sizes of the dishes at lunch time, particularly the main
composite dish. The daycares should also consider increasing the supply of solid
food instead of beverages in the menus, a practice that will increase both the
energy density (Leahy et al. 2008) and the satiety (Kral et al. 2007; de Graaf
2011; Popkin 2011) of the meals, and discourage consumption of cheap high
energy dense foods in the home (Antunes et al. 2010; de Bem Lignani et al. 2011).

4.4.3 Strengths and limitations

This is the first study to evaluate the adequacy of the energy and nutrient
supply from the menus for preschool children used in philanthropically funded
daycares in Salvador, NE Brazil and to provide some suggestions for their
improvement. Our study has also quantified the contribution of fortified wheat
and maize flour products to the iron and folate supply from these daycare meals,
assuming the flour fortificant levels were the minimum levels specified.
Nevertheless, our study has some limitations. The results are not based on actual
measurements of food intake by the children, but instead on the energy and
nutrients supplied by standard serving sizes (irrespective of age) of the 20 daycare
menus. As a result, we have no information on the distribution of usual intakes
among the preschoolers. However, very little food was leftover on a daily basis,
except by children who were not feeling well. Further, children were reported to
receive very small amounts of food, if any, from home during the week, so the

89
usual weekday supply probably reflects the usual weekday intake of energy and
nutrients. Six of the seven daycares in this study used these same menus, and the
menus used by the other daycare established by Mansão do Caminho were said to
be comparable. Foods and beverages consumed by the preschoolers during the
weekends were not available. Finally, information regarding the type of iron
fortificant, and the exact levels of the iron and folic acid fortificants used in the
daycare menus was not available, so our estimates for the usual weekday supply
of iron and folate were based on the minimum recommended fortification levels.

4.4.4 Conclusion

In conclusion, the daycares are commended for their efforts to meet the
nationally recommended food-based dietary guidelines for children over two
years of age and for providing menus with such a high dietary quality for this
group of low-income Brazilian preschoolers. Some modifications are needed,
however, to ensure that each child is supplied with his / her requirements for
calcium, thiamine, and vitamin A, and the desirable range of energy derived from
fat. Inadequate intakes of vitamin A may increase the susceptibility of the
preschoolers to certain parasitic infections, especially in a setting of urban
poverty. Thus, an investigation of the parasite status of these preschoolers is
warranted.

This chapter has been submitted for publication in a modified format as:
Adequacy of daycare meals for disadvantaged preschoolers in Salvador, Brazil
Rebecca L. Lander, Hugo C. Costa-Ribeiro, Jeane dos Santos Ferreira, Sued
Macedo Ramos Burgues, Danile L. Barreto, Rosalind S. Gibson
(in process).

90
5 Risk factors associated with intestinal
parasite infestation of preschool children
attending daycares in Salvador, NE Brazil

5.1 Introduction
Infections from gastro-intestinal parasites, including soil-transmitted
helminths and pathogenic protozoa, are widespread in crowded urban
environments with poor sanitation (Crompton & Nesheim 2002; Sclar et al. 2005).
Hence, it is not surprising that in the urban and peri-urban areas of the NE region
of Brazil, one of the most under-resourced areas in the country, risk of intestinal
parasitic infection is high, especially in view of the warm and humid tropical
environment (Utzinger & Keiser 2006; Hotez 2008) and the lack of access to a
public water supply and a sanitation system. Indeed, less than 30% of preschool
children in the region live in households with access to both water and sewage
services (de Lima et al. 2010). Thus, preschool and school-aged children living in
such environments are more at risk to helminth infections (Hotez 2008) than older
children or adults because of their increased activity in potentially infective
environments and lack of conscientious sanitary behaviours (Harhay et al. 2010).
Helminth infections, such as hookworm, Ascaris lumbricoides
(roundworm), and Trichuris trichiura (whipworm), have been linked with
micronutrient deficiencies, notably zinc, vitamin A, and iron (Tanumihardjo et al.
2004), as well as impaired growth in children (Stephenson et al. 2000a), induced
by anorexia, nausea and vomiting, reductions in digestion and absorption, and
enhanced nutrient losses (Koski & Scott 2001). Indeed, in some preschool and
school-aged children, depletion in body iron from blood loss arising from
hookworm and T. trichiura has been associated with an increased risk of anemia
and impaired cognition (Sakti et al. 1999; Stephenson et al. 2000b; Stoltzfus et al.
2000).

91
The pathogenic protozoan infections, particularly Giardia intestinalis (syn
Giardia duodenalis, Giardia lamblia), Cryptosporidium (var species) and
Entamoeba histolytica are also commonly found in poor urban settings (Harhay et
al. 2010). Children attending daycare centres are especially at risk of Giardia
intestinalis and Cryptosporidium infections because of the ease of faecal-oral
transmission of these organisms between a host child and other children and
daycare staff (Thompson 2000; Correia do Nascimento et al. 2009). Like
helminths, protozoan infections can also pose a significant health threat in young
children, causing chronic diarrhoea, anorexia, and abdominal pain, all of which
can have a negative impact on micronutrient absorption, micronutrient status,
immune function, growth, and sometimes cognition.
Therefore, the present study aims to: (1) assess the prevalence of intestinal
parasites in a pre-selected group of preschool children aged three to six years
attending urban and peri-urban daycares in Salvador, NE Brazil; and (2) explore
the risk factors associated with parasite infestation.

5.2 Participants and Methods

5.2.1 Study sites and participants

This cross-sectional study was conducted between August – November,


2010 in Salvador, the capital city of the state of Bahia. The latter is one of the
poorest areas in Brazil, and is located in the NE region where a large proportion of
the population is living without adequate housing and sanitation (de Lima et al.
2010). Even in the city of Salvador, appropriate sewerage coverage is still lacking
in high risk areas of this densely populated city (Prado et al. 2003), despite
implementation of a city-wide sanitation intervention within the past decade
(Moraes et al. 2004; Barreto et al. 2007).
The participants in this study were from low-income families and attended
philanthropically-supported daycares located in urban and peri-urban areas of
Salvador; details of the study design are reported earlier (see Chapter 3). Of the
376 children, 86% (n=325) provided a stool sample for parasite examination, the

92
basis for the findings reported here. The study protocol was approved by the
responsible Ethics committees and overarching daycare organizations, as
described in section 3.2.1, and informed written consent to participate in the study
was given by the parents or primary guardians of the children.

5.2.2 Collection and microscopic assessment of parasites in stool samples

Stool collection containers and instructions were distributed to the


caregivers of the 376 participants. The plastic containers, labeled with the child’s
full name and birth date, had a scoop attached to the screw-top lid, and contained
no preservatives. The container and collection instructions (Appendix D) were
placed together in a sealable plastic bag and distributed by the daycare
coordinators to the caregivers when they came to pick up their child from daycare.
The caregivers were instructed to bring the stool sample back to the daycare the
following morning. A replacement container and instructions were provided if the
caregiver failed to return the child’s stool sample, up to a maximum of three
replacements. A total of 325 samples were returned during the collection phase.
Only one stool sample was collected from each participant.
The stool sample containers were collected from the daycares each morning
and transported directly to the laboratory in a chilled box. In the laboratory, a
small amount of stool (2–4 grams) was thoroughly homogenized into ~10 ml
sodium acetate-acetic acid formalin (SAF) (Ft. Richard Laboratories, Auckland)
by manual stirring with a plastic spatula. Next, a faecal concentrate of the SAF
mixture was prepared by using the standard formalin-ethyl acetate sedimentation
concentration procedure (Garcia 2009). A standard amount of sediment (100 µl)
from each faecal concentrate was placed onto a slide and then examined for the
presence of helminthic and protozoal intestinal parasites by microscopy.
Trichrome stain microscopy examination was used to confirm the presence of
Entamoeba histolytica / dispar cysts, although whether the cysts were pathogenic
Entamoeba histolytica or non-pathogenic Entamoeba dispar could not be
ascertained. The presence of other non-pathogenic protozoal cysts, such as

93
Table 5.1 Mean (SD)1 differences for specified variables between the
participants who provided a stool sample (n=325) and those who did not provide
a stool sample (n=51)

Variable n Faecal samples n No faecal samples p


Mean (SD) Mean (SD)

Age (years) 325 4.26 (0.61) 51 4.08 (0.63) 0.052

HAZ 322 0.47 (1.06) 42 0.54 (1.14) 0.703

WAZ 322 0.13 (1.03) 42 0.16 (0.98) 0.854

WHZ 321 -0.23 (1.05) 42 -0.23 (0.93) 0.980

BMIZ 322 -0.22 (1.17) 42 -0.25 (0.98) 0.906

n (%) n (%) p

Sex (male) 325 173 (53.2) 51 23 (45.1) 0.295

Mother education 325 197 (60.6) 51 28 (54.9) 0.446


(low)

SES group 325 157 (48.3) 51 25 (49.0) 1.000


(extremely low)

Deworming 325 169 (52.0) 51 23 (45.1) 0.371


treatment

Vitamin A 325 181 (55.7) 51 19 (37.3) 0.016


supplementation

1
Unless shown as a frequency (%)

94
Entamoeba coli, Endolimax nana, Iodamoeba butschlii, and Chilomastix mesnili,
was also noted. A semi-quantitative egg-burden estimate of positive helminth
samples was also determined based on the number of eggs counted during the
microscopy examination. Light, moderate, and heavy loads were defined as ≤1
egg per 10 low power fields (10x), 2–9 eggs per 10 low power fields (10x), and
≥1 egg per low power field (10x), respectively. Microscopy examination of the
faecal concentrates was repeated on a random sub-sample (10% in total) by a
laboratory scientist not involved with this study to determine reproducibility of
the methods used.

5.2.3 ELISA assays for qualitative detection of Giardia and Cryptosporidium


antigens in faeces

A separate aliquot of the SAF mixture (~2 ml) was used to detect the
presence of Giardia and Cryptosporidium antigens using an enzyme-linked
immunosorbent assay (ELISA) (TechLab Inc, Virginia). The
GIARDIA/CRYPTOSPORIDIUM CHEK® is an in vitro immuno-assay which uses
monoclonal and polyclonal antibodies to cell-surface antigens of Giardia and an
oocyst antigen of Cryptosporidium sp. Briefly, the immobilized monoclonal
antibodies bind the Giardia and Cryptosporidium antigens if the antigens are
present. A conjugate is then added which binds to the antigen / antibody complex,
with any unbound materials removed during the washing steps. Following the
addition of a substrate of tetramethylbenzidine and peroxide, a specific shade of
colour is detected due to the enzyme-antibody-antigen complexes that form in the
presence of Giardia and / or Cryptosporidium antigens and the conjugate. The
intensity of colour was measured by a calibrated spectrophotometric ELISA plate
reader and positive results for the presence of Giardia and / or Cryptosporidium
antigens gave an absorbance of ≥ 0.090, showing a pale yellow to strong yellow
visual colour (Wampole® 2006). Each positive sample of this combined antigen
assay was then re-tested, once with a Giardia-specific ELISA test and once with a
Cryptosporidium-specific ELISA test, according to the manufacturer’s

95
Table 5.2 Prevalence (% and 95% CI) of intestinal parasite infection (n=325)

Variable n % (95% CI)

Presence of parasites 95 29.2 (24.3, 34.5)

Helminths 58 17.8 (13.8, 22.5)


Trichuris trichiura 39 12.0 (8.7, 16.0)
Ascaris lumbricoides 34 10.5 (7.4, 14.3)
Hookworm 3 0.9 (0.2, 2.7)
Giardia intestinalis 42 12.9 (9.5, 17.1)
Entamoeba histolytica / dispar 12 3.7 (1.9, 6.4)
Cryptosporidium sp. 1 0.3 (0.01, 1.7)
Hymenolepsis nana 1 0.3 (0.01, 1.7)
Strongyloides stercoralis 1 0.3 (0.01, 1.7)

Presence of non-pathogenic 80 24.6 (20.0, 29.7)


protozoal cysts

Presence of ≥ 2 parasites 30 9.2 (6.3, 12.9)


Presence of ≥ 3 parasites 8 2.5 (1.1, 4.9)

96
instructions (TechLab Inc., Virginia), in order to determine whether the sample
was positive for Giardia, Cryptosporidium or both protozoans (TechLab Inc.,
Virginia).

5.2.4 Assessment of socio-economic status

A culturally appropriate structured questionnaire was developed to collect


information on the socio-demographic and health characteristics of the
participants, as described previously (see Chapter 3). An overall SES score was
developed for each participant from socio-demographic and sanitation data
collected from the caregivers (Appendix B), as described in section 3.2.1.
Information on deworming treatment within the past six months and vitamin A
supplementation was provided by maternal report and review of the child’s health
card, provided by the caregiver or daycare.

5.2.5 Statistical analysis

Differences in age, anthropometric indices, sex, mother’s education, SES,


use of deworming treatment and vitamin A supplementation between the
participants who provided stool samples and those who did not were compared
using a Student’s two sample t-test for the continuous variables and Fisher’s exact
test for the categorical variables. The prevalence (as % and 95% CI, with poisson
distribution used for frequencies of < 10) of helminthic and protozoal intestinal
parasites was calculated based on the results of the microscopic examination and
the ELISA method for antigen detection of Giardia intestinalis and
Cryptosporidium sp. The daycares were divided into three groups (i.e., city centre,
peri-urban favela daycares, and peri-urban gated daycare) for comparison of the
prevalence of infection with helminths, selected protozoal intestinal parasites and
non-pathogenic protozoal cysts, deworming treatment, and vitamin A
supplementation using Fisher’s exact test. The latter was also used to examine the
relationship between poly-parasitism and SES. Logistic regression models were
used to examine relationships between infection with helminths and with Giardia
intestinalis for the following factors: aged 3.00-3.99 years and 4.00-5.65 years;

97
Table 5.3 Frequency (%) of intestinal parasite infection, deworming treatment and vitamin A supplementation
by daycare group (n=325)

Variable City centre Peri-urban favela Peri-urban gated


N=60 N=185 N=80

n (%) n (%) n (%) p

Presence of parasite 10 (16.7) 68 (36.8) 17 (21.3) 0.002

Helminth1 4 (6.7) 43 (23.2) 11 (13.8) 0.007


98

Giardia intestinalis 7 (11.7) 29 (15.7) 6 (7.5) 0.182

Entamoeba histolytica / dispar 0 (0) 10 (5.4) 2 (2.5) 0.162

Non-pathogenic protozoal cysts 6 (10.0) 56 (30.3) 18 (22.5) 0.004

Deworming treatment within 6 months 26 (43.3) 91 (49.2) 52 (65.0) 0.020

Vitamin A supplementation ever 35 (58.3) 121 (65.4) 25 (31.3) <0.001

1
Helminth includes T. trichiura, A. lumbricoides and hookworm
sex, SES category, deworming treatment and vitamin A supplementation. The
sandwich estimator was used to obtain robust standard errors, to account for the
sampling procedure. The two SES categories were extremely low and low SES,
based on scores ≤ 34 and ≥ 35, respectively (Issler et al. 1996). Statistical analyses
were carried out using STATA version 11 (Stata Corporation, College Station, TX
USA).

5.3 Results

5.3.1 Comparison of characteristics between participants providing and not


providing a stool sample

Of the 376 participants in the overall study, 86% provided stool samples. No
significant differences in age, anthropometric indices, sex, mother’s education,
SES, and deworming treatment were found between the participants who provided
a stool sample (n=325) and those participants who did not (n=51), with the
exception of vitamin A supplementation (Table 5.1). More children who provided
a stool sample had received a vitamin A supplement compared to those who did
not provide a stool sample (p=0.016).

5.3.2 Prevalence of parasitic infections

The prevalence of helminthic and protozoal intestinal parasites as


determined by microscopy and ELISA assay (for Giardia and Cryptosporidium
sp.) is presented in Table 5.2. Nearly 30% (95/325) of the participants were
infected with at least one intestinal parasite. No species of schistosomes were
found in the stool samples. One-quarter (80/325) of all the children had non-
pathogenic protozoal cysts in their stool.
Trichuris trichiura and Ascaris lumbricoides were the most common
helminths reported with a prevalence (95%CI) of 12.0% (8.7, 16.0) and 10.5%
(7.4, 14.3), respectively, compared to the very low rate of hookworm infestation
(0.9%; 0.2, 2.7). Male participants had a significantly higher frequency of
infection with T. trichiura (p=0.016), but not with A. lumbricoides infection, than

99
Table 5.4 Logistic regression models of risk factors (by odds ratio and
95% CI) for infection with helminths (n=58) and Giardia intestinalis (n=42)

Helminths G. intestinalis
Variable p p
(n=58) (n=42)

Age group 1.09 (0.59, 2.03) 0.776 0.92 (0.37, 2.33) 0.866
(older group)

Sex (male) 2.34 (1.14, 4.82) 0.021 1.15 (0.67, 1.97) 0.620

SES group 2.04 (1.33, 3.13) 0.001 2.02 (0.87, 4.69) 0.101
(extremely low)

Deworming 0.60 (0.40, 0.91) 0.015 1.13 (0.51, 2.52) 0.764


treatment

Vitamin A 0.67 (0.29, 1.56) 0.355 0.46 (0.20, 1.03) 0.061


supplementation

100
the females (results not shown). Of the children with T. trichiura infestation,
nearly half (49%; 19/39) had a light egg burden, 36% (14/39) had a moderate egg
load, and 15% (6/39) had a heavy egg burden. The egg burden of children infected
with A. lumbricoides infection was similar to the T. trichiura infection, with light,
moderate and heavy egg burdens of 50% (17/34), 26% (9/34), and 24% (8/34),
respectively (results not shown). Fifteen children (5%) were infected with both T.
trichiura and A. lumbricoides. Hookworm infestation was always in conjunction
with either T. trichiura (n=2) or A. lumbricoides (n=1) (results not shown).
Giardia intestinalis infection was reported in 12.9% (9.5, 17.1) of the stool
samples. Fourteen children (4%) had both G. intestinalis and a helminth, either T.
trichiura (n=8), A. lumbricoides (n=3) or all three parasites simultaneously (n=3)
(results not shown). Entamoeba histolytica / dispar cysts were found in 3.7% (1.9,
6.4) of the stool samples, with approximately half of these infections (n=5) in
conjunction with one or more helminths and / or G. intestinalis. Of the children
who provided a stool sample for examination 9.2% (6.3, 12.9) had two or more
parasites present and 2.5% (1.1, 4.9) were infected with three or more parasites.
The prevalence of poly-parasitism (i.e., ≥ two parasites) was modestly higher
(p=0.089) in the extremely low SES group than the low SES group (i.e., 63.3%
(19/30) vs. 36.7% (11/30), respectively) (results not shown).
The frequency (%) of parasite infections, deworming treatment, and
vitamin A supplementation by daycare group are shown in Table 5.3. Significant
differences existed among the three daycare groups for the frequency of infection
with an intestinal parasite, specifically helminths and non-pathogenic protozoal
cysts, as well as for deworming and vitamin A supplementation. In general, the
children attending the peri-urban favela daycare had the highest prevalence of
intestinal parasite infections. Deworming treatment was more frequent among the
children attending the peri-urban gated daycare (65%), although only about a third
of these children had received vitamin A supplementation.

101
5.3.3 Risk factors associated with helminth and Giardia infections

In the multivariate logistic regression analysis, being male (p=0.021) and


living in extremely low SES circumstances (p=0.001) were significant risk factors
for the child being infected with helminths. There was a strong inverse association
between deworming treatment and helminth infection (p=0.015) and vitamin A
supplements showed a modest inverse association with Giardia intestinalis
infection (p=0.061) (Table 5.4).

5.4 Discussion
Our results showed that Ascaris lumbricoides, Trichuris trichiura and
Giardia intestinalis were the most prevalent intestinal parasite infections in these
preschoolers attending daycare in Salvador, whereas the prevalence of hookworm
and schistosomes was low. Further, the intensity of infection from A. lumbricoides
and T. trichiura was moderate to heavy in 50% of the children, based on the egg
burden data, increasing their risk of morbidity (Crompton & Nesheim 2002). In
general, other studies conducted in preschool children in Salvador (Prado et al.
2003; Barreto et al. 2010) and elsewhere in Brazil (Newman et al. 2001; Teixeira
et al. 2007) have reported similar results in both preschool children and school
aged children (Tsuyuoka et al. 1999; Fonseca et al. 2010), two age groups
particularly at risk of helminthic and protozoan infections (WHO 2002; Harhay et
al. 2010). Infection with hookworm and schistosomes was low as might be
expected in urban areas, and among children who were probably too young to
play in rivers and natural bodies of water (Hotez 2008).

5.4.1 Risk factors associated with helminth infections

Our results have highlighted two significant risk factors for helminth
infections among the preschool children studied here: being male and living in
extremely impoverished circumstances. Age group had no significant effect,
probably because we studied such a narrow age range of children. Other
investigators have shown that infection with A. lumbricoides and T. trichiura

102
often starts early in life, usually reaching the most intense infections between ages
5–15 years (Bethony et al. 2006; Jardim-Botelho et al. 2008a; Harhay et al. 2010).
In our study the boys had more than a two-fold risk for helminth infection
than girls (Table 5.4), particularly T. trichiura infection, perhaps because boys are
more likely to play barefooted outside. A compromised zinc status may be an
additional factor because boys have a higher zinc requirement than girls as a result
of their faster growth rate (Ruz et al. 1997; Gibson et al. 2007). Certainly, an
inverse relationship between a zinc status and worm burdens has been identified
in mice-models (Scott & Koski 2000). Further, in preschool children living in a
slum in Bangladesh, low serum zinc levels were associated with T. trichiura
infection (Kongsbak et al. 2006b).
The significant positive association noted here between risk of helminth
infections and living in extremely low socio-economic status circumstances is
well documented (Bethony et al. 2006). Such low-income and population-dense
environments also contribute to an increased risk for polyparasitism (i.e., infection
with >1 parasite) in young children (Harhay et al. 2010), with significant adverse
consequences on growth, morbidity, and cognition (Sawaya et al. 1990; Jardim-
Botelho et al. 2008b). In our study, 30 children (9%) were infected with more than
one parasite, notably helminths with or without G. intestinalis, of whom nearly
two-thirds were from the extremely low SES group, a trend that has been reported
in Brazilian children by others (Saldiva et al. 1999; Carvalho-Costa et al. 2007).
The protective association between helminth infections and deworming
treatment reported here is not unexpected (Table 5.4), and emphasizes the
importance of providing such an inexpensive treatment (Sinuon et al. 2005) at
appropriate intervals to the preschoolers studied here (Crompton & Nesheim
2002; WHO 2002). Albendazole or Mebendazole is the standard deworming
treatment (Garcia 2001). Both drugs are beneficial for the treatment of A.
lumbricoides, although less efficacious against T. trichiura, which appears more
resistant (Stephenson et al. 1993; Stoltzfus et al. 1997a; Dossa et al. 2001; Olsen
et al. 2009). More children in the peri-urban gated daycare had received
deworming treatment within the past six months compared to the children

103
attending the city centre or peri-urban favela daycares (Table 5.3). Nevertheless,
the prevalence of helminth infections was not lowest in these children, perhaps
because T. trichiura and not A. lumbricoides was the predominant helminth
infection in the peri-urban gated daycare. Indeed, the peri-urban favela daycare
group had the highest frequency of parasitic infections and non-pathogenic
protozoal cysts, probably because of their close proximity to unpaved roads and
use of water from contaminated sources, together with poor drainage and sewage
connections (Moraes et al. 2004; Barreto et al. 2010). In view of the wide range of
sometimes co-existing intestinal parasites infecting these daycare preschool
children, new strategies are required for the management of polyparasitism in
such urban slum environments (Harhay et al. 2010).

5.4.2 Risk factors associated with Giardia intestinalis

Giardiasis has long been associated with outbreaks of infectious diarrhoea


and gastroenteritis in daycare settings in both high- and low-income countries
(Rauch et al. 1990; Thompson 1994; Sempértegui et al. 1995; Rodríguez-
Hernández et al. 1996). Giardia intestinalis, a water-borne protozoa, is most
commonly acquired through the ingestion of contaminated water (Lane & Lloyd
2002; Ali & Hill 2003), although person-to-person spread is also common,
particularly in environments like daycare settings studied here. Our prevalence
data for G. intestinalis infection was comparable to that observed in two studies
conducted in preschool daycare settings in São Paulo State, Brazil (Machado et
al. 1999; Tashima et al. 2009) (i.e., 13% vs. 10% and 15%, respectively), although
an even higher prevalence has been reported in public daycares (i.e., 61%)
(Machaodo et al. 1999).
Unlike the helminths, we did not identify any significant risk factors for
infection with G. intestinalis. However, vitamin A supplements tended to be
modestly protective against G. intestinalis infection (Table 5.4). We are not the
first to report an inverse association between vitamin A supplementation and G.
intestinalis infection in children living in an impoverished environment in NE
Brazil. Lima et al. (2010) also found G. intestinalis infections were lower in the

104
children who had received vitamin A supplementation. Vitamin A has a critical
role in both immune function (Semba 1994) and epithelial gut integrity (Chen et
al. 2003; Vieira et al. 2008), and hence may stimulate the immune responses
needed to reduce G. intestinalis infection. Associations between Giardia-infected
school-children and low serum retinol have been reported in Mexico (Quihui-Cota
et al. 2008), although the bidirectional interaction is complex (Filteau & Tomkins
1994).
Other investigations of intestinal parasite infections in public daycares in
both northeast (Correia do Nascimento et al. 2009) and southeast Brazil
(Carvalho-Almeida et al. 2006) have focused on the spread of Cryptosporidium
because, like G. intestinalis and E. histolytica, it is associated with diarrhoeal
illness in preschool children (Mondal et al. 2006). We only found one positive
stool sample for Cryptosporidium, although nearly 5% of the children had E.
histolytica / dispar cysts, with E. histolytica the principle cause of amoebiasis
(Stanley 2003; Garcia 2009). Importantly, a quarter of the children had non-
pathogenic protozoal cysts in their stools, which is indicative of faecal-oral
transmission and the possible presence of pathogenic protozoan species unable to
be detected by concentrate microscopy (e.g., an E. histolytica infection present
with low numbers of cysts or high numbers of trophozoites) (WHO 2004b).

5.4.3 Strengths and limitations

Our study has both some strengths and limitations. We used the faecal
concentration test which had a high reproducibility (i.e., > 95%), and is more
sensitive than the more frequently used Kato-Katz method (Harhay et al. 2010).
Constraints with available laboratory facilities prevented us from using both the
sedimentation and the flotation procedures for parasite examination. Hence, based
on standard recommendations, the sedimentation method was chosen. Further, we
also used the ELISA antigen detection test which is more sensitive and specific
for assessing the prevalence of G. intestinalis and Cryptosporidium sp. than
microscopy examination alone. However, we were only able to collect one stool
specimen per child, which provided a semi-quantitative assessment of the

105
intensity of helminth infections. Finally, we cannot establish causality from the
risk factors or protective associations identified for infection with helminths or G.
intestinalis because of the observational nature of our study. In future studies,
information on household domestic animals should also be collected because they
have been identified as a risk factor in the transmission of G. intestinalis infection
(Thompson 2000; Lane & Lloyd 2002; Sackey et al. 2003).

5.4.4 Conclusion

In conclusion, nearly one-third of the preschool participants presented with


an intestinal parasite infection, notably Ascaris lumbricoides, Trichuris trichiura
and / or Giardia intestinalis. Moreover, in about 50% of the children with
helminths, the intensity of infection from A. lumbricoides and T. trichiura was
rated as moderate to heavy, increasing the likelihood of morbidity. Our findings
that boys had a greater risk than girls for helminth infections, and that vitamin A
supplements was inversely associated with the risk for G. intestinalis infection,
suggest that an investigation of the biochemical zinc and vitamin A status of these
preschool children is warranted.

This chapter has been accepted for publication in a modified format as:
Factors associated with growth and intestinal parasite infections in
preschoolers attending philanthropic daycares in Salvador, Northeast Brazil
Rebecca L. Lander, Alastair G. Lander, Hugo C. Costa-Ribeiro,
Danile L. Barreto, Angela P. Mattos, Sheila M. Williams,
Lisa Houghton, Rosalind S. Gibson
Cadernos de Saúde Pública (in press).

106
6 Micronutrient status of disadvantaged
preschoolers attending daycares in
Salvador, NE Brazil

6.1 Introduction
The micronutrient status of preschoolers in Brazil has been largely
unexplored, with national estimates of key micronutrient deficiencies such as zinc,
lacking in this age group (Victora et al. 2011). An exception is vitamin A, with
deficiency of this micronutrient declared a public health problem in children
under-five years of age living in NE Brazil more than thirty years ago (WHO
1982). As a result, efforts have been made to increase the coverage of vitamin A
supplementation in the region (Ramalho et al. 2002; Martins et al. 2004; Milagres
et al. 2007). In small studies conducted more recently, the prevalence of vitamin
A deficiency in preschoolers has decreased from more than 30%, now ranging
from 7–15% (Fernandes et al 2005; Paiva et al. 2006; de Azevedo et al. 2010).
Nevertheless, preschoolers living in poor urban environments in the region may
be susceptible to other co-existing micronutrient deficiencies, as has been reported
elsewhere (Anderson et al. 2008; Nhien et al. 2008; Lander et al. 2008).
In recent times, Brazil has implemented numerous pro-poor social policies
with the aim of enhancing nutritional outcomes for children, specifically
improvements in birth weight, growth, and micronutrient status (Gakidou et al.
2007). These include the promotion of universal education (Barros et al. 2010)
and childhood immunization coverage (Barreto et al. 2011), as well as targeted
vitamin A supplementation in the Northeastern states (Martins et al. 2007; de
Almeida et al. 2010). Programmes to expand the water supply and sanitation
(Victora et al. 2011) and fortify wheat and corn flour with iron and folic acid
(Brazil Ministry of Health 2002) have also been introduced. Free childcare for the
poor through philanthropically run daycare centres has been promoted (Marins &
Rezende 2011), as well as CCT programmes, (e.g., Bolsa Familia) to provide

107
monetary transfers to low-income families who comply with certain programme
requirements, such as preventive health visits and school enrollment (Burlandy
2007).
Unfortunately, the poor have seldom been the first recipients to benefit from
these pro-poor policies in Brazil (Victora et al. 2000). Instead, it has been families
of higher socio-economic status, a trend that has also been reported in Argentina
and Colombia (Gwatkin et al. 2004). Furthermore, these policies often fail to
target specific disease-causing environmental factors, such as soil-transmitted
helminth infections among urban slums in Brazil (Korkes et al. 2009). Little
attention has been directed to the population-wide control of helminths, despite a
large reduction achieved through a mass helminth treatment campaign in very
poor preschoolers in an endemic area in the Amazon (Boia et al. 2006). This is
disappointing because helminth infection, together with pathogenic protozoa
Giardia intestinalis, have the potential to exacerbate risk for micronutrient
deficiencies (e.g., vitamin A) through poor absorption and excessive losses
(Solomons 1993; Koski & Scott 2001). Certainly, parasitic infections have
compromised the micronutrient status of young children in other low-income
environments (Olivares et al. 2002; Long et al. 2007; Quihui-Cota et al. 2008),
although their impact on disadvantaged children living in NE Brazil has not been
investigated.
Therefore, this present study conducted in a pre-selected group of
disadvantaged preschool children aged three to six years attending urban and peri-
urban daycares in Salvador, NE Brazil aims to: (1) assess the prevalence of
deficiencies of vitamin A, zinc, selenium, vitamin B-12 and folate; (2) examine
associations between the micronutrients; and (3) identify predictors of
micronutrient biomarkers with emphasis on parasitic infections and socio-
demographic factors. Such data, together with the average daily micronutrient
supply for the preschoolers from the daycare menus (Chapter 4), may provide
insight on whether some of these pro-poor policies have had a positive influence
on the micronutrient status of these preschoolers.

108
6.2 Participants and Methods

6.2.1 Study sites and participants

This cross-sectional study was conducted in Salvador, the capital city of the
state of Bahia, located in NE Brazil between August – November, 2010. Salvador
is a city with a population of three million with a high degree of social inequality
(Assis et al. 2007). Seven philanthropically funded preschool daycares located
within the city centre and in peri-urban areas were pre-selected for participation in
this study, as described elsewhere (see Chapter 3). The children (n=376) enrolled
in the daycares (maximum class size of 25 children) were from low-income
families and attended daycare five days per week, except holidays, until school
age. The study protocol was approved by the responsible Ethics committees and
overarching daycare organizations, as described in section 3.2.1, and informed
written consent to participate in the study was given by the parents or primary
guardians of the children.

6.2.2 Assessment of socio-demographic, health, and morbidity status

A culturally appropriate structured questionnaire was developed to collect


information on the socio-demographic and health characteristics of the
participants, as described earlier (Chapter 3). Information on health status
variables for each child included the child’s maternal reports of parasite
deworming within the past six months, and the number and reasons for previous
hospitalizations, such as acute respiratory infections and diarrhoea. The ethnicity
of the children was determined by skin colour, and hair and facial characteristics,
as performed in the 2010 national census (IBGE 2010a) (see section 2.4.2).
Review of the child’s health card, provided either by the caregiver or the daycare,
was used to verify and / or record information on vitamin A supplementation.
Information about smoking in the home, either by the mother or another adult,
was collected, and history of asthma in the mother or siblings was also recorded.

109
Table 6.1 Socio-economic and health characteristics of the participants
and prevalence of intestinal parasitic infections

n %

Socio-economic status (extremely low) 182/376 48.4

Ethnicity
Black 154/365 42.2
Brown 189/365 51.8
White 22/365 6.0

Vitamin A supplementation (ever received) 200/376 53.2

Deworming treatment within 6 months 192/376 51.1

Hospitalizations 187/376 49.7


1-2 times 139/187 74.3
3 or more times 48/187 25.7

Reasons for hospitalizations


Diarrhoea 16/187 8.6
Upper respiratory infection 93/187 49.7
Other reasons 78/187 41.7

Smoking
Mother smoking in the house 47/376 12.5
Adult smoking in the house 89/376 23.7
History of asthma in mother or sibling 68/376 18.1

Parasite present 95/325 29.2


Helminths 58/325 17.8
Trichuris trichiura 39/325 12.0
Ascaris lumbricoides 34/325 10.5
Hookworm 3/325 0.9
Giardia intestinalis 42/325 12.9

110
6.2.3 Assessment of parasite status

A stool sample was submitted for parasite analysis by 86% (325/376) of the
participants. A faecal concentrate was prepared as described in Chapter 5 and
examined for the presence of helminths and protozoal intestinal parasites by
microscopy. An ELISA assay was used to identify positive samples for Giardia
intestinalis, also described previously.

6.2.4 Biochemical assessment

Morning fasting blood samples were obtained by peripheral venipuncture by


trained phlebotomists with participants in the sitting position, either held by their
mother or caregiver or sitting alone. A topical local vasodilator anesthetic
amethocaine (AmetopTM) was applied to the venipuncture site 30 minutes prior to
blood sampling in order to minimize any discomfort. Blood was drawn into a
trace-element-(TE)-free evacuated tube (Becton Dickinson, Franklin Lakes, NJ,
USA) in accordance with the IZiNCG (Hotz & Brown 2004) procedures for serum
zinc. All blood samples were refrigerated immediately following collection
(Tamura et al. 1994), protected from ultra-violet light, and the serum separated
within two hours using TE-free techniques. Aliquots of serum were stored in TE-
free polyethylene vials and frozen immediately, initially at -30°C and later at
-70°C. An aliquot of EDTA anticoagulated whole blood drawn into a second
vacutainer was hemolyzed by a 1:10 dilution in 1% ascorbic acid and frozen in
preparation for erythrocyte folate analysis (Thurlow et al. 2005). Frozen samples
were shipped on dry ice to the Trace Element Laboratory of the Department of
Human Nutrition, University of Otago, New Zealand for analysis.
Serum zinc was analyzed by flame atomic absorption spectrophotometry
(AAS) (ContrAA 700, Analytik Jena, Germany) using a modified method of
Smith et al. (1979). Serum selenium was assayed via electrothermal AAS (AA-
800, Perkin Elmer 2690, Ebos Group Ltd., Auckland, New Zealand) based on a
modified method of Jacobson and Lockitch (1988). Serum retinol was analyzed
by high-pressure liquid chromatography (Thurnham et al. 1988). Serum vitamin
B-12 was determined using a commercial kit (Vitamin B-12 Elecsys reagent kit,

111
Table 6.2 Quality control data for the assessment of accuracy and precision
of the biomarkers

Measured Certified
CV (%)
Mean (SD) Range

Serum zinc (µmol/L)


UTAK control 9.7 (0.4) 7.8-13.0 4.1
NIST control 13.2 (0.8) 13.1-13.8 5.9
Human pooled 13.1 (0.6) 4.5

Serum selenium (µmol/L)


UTAK control 106.9 (6.2) 74-124 5.8
Seronorm 112.2 (6.2) 94-118 5.6
Pooled serum 96.5 (6.9) 7.2

Serum retinol (µmol/L)


UTAK low control 1.6 (0.05) 1.3-1.8 3.4
UTAK high control 5.4 (0.19) 4.7-6.4 3.6
Pooled serum 1.87 (0.04) 2.2

Serum vitamin B12 (pmol/L)


Low control 308 (12) 166–436 4.0
Medium control 571 (18) 401–653 3.2
High control 1203 (52) 885–1355 4.3

Serum folate (nmol/L)


NIST low control 5.4 (0.1) 4.4–6.8 2.7
NIST medium control 12.7 (0.5) 11–17 4.1
Pooled serum 10.4 (1.5) 14.2

α -1-glycoprotein (g/L)
Precinorm 86.0 (1.1) 63.3–102.9 1.3
Pooled serum 77.1 (2.3) 2.9

C-reactive protein (mg/L)


Precinorm 11.7 (0.2) 9.8–13.4 2.0
Pooled serum 2.3 (0.2) 10.0

CV = Coefficient of variation equation (sd*100) / mean) %

112
Roche Diagnostics GmbH, Mannheim, Germany) based on an electrochemi-
luminescence immunoassay (ELICIA) carried out on an auto-analyzer (Roche,
New Zealand). Serum and whole blood folate concentrations were measured by
microbiological assay (Molloy & Scott 1997) in 96-well microtiter plates using
chloramphenicol–resistant cryo–preserved Lactobacillus rhamnosus (ATCC
27773; American Type Culture Collection, Manassa, VA) and 5–
methyltetrahydrofolate as the calibrator (Merck & Cie, Schaffhausen,
Switzerland). Erythrocyte folate concentrations were calculated from whole blood
values by using individual packed cell volumes and correction for serum folate
concentration. Serum CRP and AGP concentrations were assayed by
immunoturbidimetry (Roche Diagnostics) on an automated analyser system
(Cobas Mira II). The controls and standards for CRP and AGP analyses were both
purchased from Roche Diagnostics. The reagents for CRP were also purchased
from Roche Diagnostics, and those for AGP from Randox.
The precision of all the biochemical assays was checked using a pooled
serum sample and their accuracy established using certified reference materials or
appropriate manufacturers’ controls. The between-assay coefficient of variations
(CV as %) for the pooled serum for zinc, retinol, selenium, folate, CRP, and AGP
were 4.5%, 2.2%, 7.2%, 14.2%, 10.0%, and 2.9%, respectively. Values for the
certified reference materials or manufacturers’ controls fell within the certified
ranges.
The following interpretive criteria were used for the biomarkers to define
micronutrient deficiencies: serum zinc < 9.9 μmol/L (Hotz et al. 2003), serum
selenium ≤ 0.82 μmol/L (Thomson 2004), and serum retinol < 0.70 μmol/L (de
Pee & Dary 2002). Marginal vitamin A status was defined as serum retinol ≥ 0.7
but < 1.05 μmol/L (Ballew et al. 2001). Deficient vitamin B-12 status was defined
as serum vitamin B-12 < 150 pmol/L (de Benoist 2008). Low serum folate
concentration was defined as < 6.8 nmol/L and low erythrocyte folate as < 317
nmol/L (Pfeiffer et al. 2007). Presence of acute and chronic inflammation was

113
Table 6.3 Mean (SD) values of micronutrient biomarkers, including adjusted
mean values for those impacted by infection

n Mean (SD) Adjusted Mean (SD)

Serum zinc (µmol/L) 358 12.8 (1.8) n/a

Serum selenium (µmol/L) 358 0.99 (0.14) n/a

Serum retinol (µmol/L) 358 1.06 (0.23) 1.09 (0.24)1

Serum vitamin B12 (pmol/L) 357 649 (224) n/a

Serum folate (nmol/L) 355 51 (21) n/a

Erythrocyte folate (nmol/L) 355 1253 (390) n/a

1
Adjusted for infection according to Thurnham et al. (2003)
Abbreviations used: SD, Standard deviation; n/a Not applicable

114
assessed by serum CRP > 5 mg/L (Thurnham et al. 2005) and serum AGP > 1.0
g/L (WHO/CDC 2007), respectively.

6.2.5 Statistical analysis

The mean (SD) micronutrient concentrations were calculated, and adjusted


where necessary for the effect of inflammation where applicable (i.e., retinol)
based on recommendations from Thurnham et al. (2003). The data were log
transformed if the variable had a strongly skewed distribution (e.g., retinol and
folate). Multiple regression models included variables if associations using
univariate regression analysis were p<0.2. The following variables were included
in the models: age, sex, micronutrients, chronic inflammation (i.e., AGP > 1 g/L)
and specific parasites (i.e., helminth and Giardia intestinalis infections). AGP was
used as a dichotomized variable because the sensitivity of the assay precluded
using the data as a continuous variable. The sandwich estimator was used to
obtain robust standard errors, to account for the sampling procedure. Statistical
analyses were carried out using STATA version 11 (Stata Corporation, College
Station, TX, USA).

6.3 Results

6.3.1 Socio-demographic / health characteristics and prevalence of parasitic


infections

A summary of the participants’ SES and health characteristics and the


prevalence of parasitic infections are shown in Table 6.1. Vitamin A
supplementation (based on ever received) and deworming treatment (within the
past six months) had been received by approximately 50% of the children. Nearly
half of the preschoolers had been hospitalized at least once, primarily for upper
respiratory infections. Approximately a quarter of the preschoolers lived in homes
with an adult who smoked; however, only 13% of the mothers smoked.

115
Table 6.4 Prevalence (% and 95% CI) of low micronutrient biomarkers and
elevated biomarkers of inflammation

n % (95% CI)

Serum zinc < 9.9 µmol/L 13/358 3.6 (1.9, 6.1)

Serum retinol1
deficient status < 0.7 µmol/L 10/358 2.8 (1.3, 5.1)
marginally deficient status 0.7–1.05 µmol/L 158/358 44.1 (38.9, 49.4)

Serum selenium ≤ 0.82 µmol/L 34/358 9.5 (6.7, 13.0)

Serum vitamin B-12


deficient status < 148 pmol/L 1/357 0.3 (0.01, 1.7)
marginally deficient status 148–221 pmol/L 5/357 1.4 (0.5, 3.3)

Serum folate
low status < 6.8 nmol/L 0/355 0 (0, 1.0)

Erythrocyte folate
low status < 317 nmol/L 0/355 0 (0, 1.0)

α -1-glycoprotein > 1.0 g/L 112/358 31.3 (26.5, 36.4)

C-reactive protein > 5 mg/L 46/358 12.9 (9.6, 16.8)

1
Adjusted for infection (Thurnham et al. 2003)

116
6.3.2 Assessment of biomarkers for micronutrients and inflammation

The quality control data for the micronutrient and inflammation biomarkers
are shown in Table 6.2. The mean (SD) concentrations of the micronutrient
biomarkers are presented in Table 6.3, with an adjusted mean (SD) value for
retinol because it was impacted by chronic inflammation (AGP > 1g/L). The
prevalence of micronutrient deficiencies was low, with deficiencies of zinc,
retinol, selenium, and vitamin B-12 less than 10% (Table 6.4). None of the
children had low folate status. Overall, nearly 13% (46/358) of the children had
one micronutrient deficiency, 1.7% (6/358) had two deficiencies; none had three
or more micronutrient deficiencies. Marginal retinol status was found in 44.1%
(158/258) of the children. Nearly a third of the children (112/358) had elevated
AGP levels, but only 13% (46/358) had high CRP concentrations.

6.3.3 Predictors of micronutrient biomarkers

Tables 6.5 and 6.6 show the associations among the micronutrient
biomarkers. For serum retinol, significant predictors included a positive
association with serum zinc (p=0.031) and an inverse association with chronic
inflammation (p=0.001). There was also a modest association with gender: boys
tended to have lower serum retinol concentrations than girls (p=0.076). The only
determinant of serum zinc was a modest and positive association with serum
selenium (p=0.088). For serum selenium, significant predictors were an inverse
association with helminths (p=0.006) and boys had a higher selenium status than
girls (p=0.018). A modest inverse association was also found with chronic
inflammation (p=0.080).
Parasitic infections were also inversely associated with serum vitamin B-12.
Infection with helminths was a strong determinant (p=0.038), whereas Giardia
intestinalis was modestly associated with vitamin B-12 (p=0.068), as was age,
with older children tending to have a lower vitamin B-12 status than younger
children (p=0.055). In contrast, parasitic infection was positively associated with
folate status; specifically Giardia intestinalis was significantly associated with

117
Table 6.5 Predictors of serum retinol, zinc, and selenium based on multiple
regression models using age, sex, other micronutrients, chronic inflammation and
parasite status, as shown by ß-coefficient (95% CI)

Micronutrient univariate ß multivariate ß (95% CI) p

Serum retinol log transformeda


Age -0.01
Sex (male) -0.041 -0.04 (-0.09, 0.01) 0.076
Zinc 0.021 0.02 (0.003, 0.04) 0.031
AGP > 1 g/L -0.122 -0.12 (-0.17, -0.07) 0.001
Helminths -0.08
Giardia intestinalis -0.061 -0.06 (-0.15, 0.03) 0.167
Vitamin A supplementation 0.00

Serum zincb
Age 0.03
Sex (male) 0.01
Selenium 1.881 1.99 (-0.38, 4.37) 0.088
AGP > 1 g/L -0.19
Helminths -0.611 -0.50 (-1.33, 0.32) 0.190
Giardia intestinalis -0.27

Serum seleniumc
Age 0.03
Sex (male) 0.041 0.05 (0.01, 0.09) 0.018
AGP > 1 g/L -0.031 -0.04 (-0.08, 0.01) 0.080
Helminths -0.052 -0.06 (-0.10, -0.03) 0.006
Giardia intestinalis -0.01

1
p < 0.2
2
p < 0.05
R2 for multivariate regression models: a 0.111; b 0.041; c 0.073

118
serum folate (p=0.041). Boys had modestly lower serum folate than girls
(p=0.088). No associations were found between micronutrient status and SES,
ethnicity, hospitalizations, supplements, deworming treatment, smoking in the
home, or family history of asthma.

6.4 Discussion
Our findings highlight the low prevalence of micronutrient deficiencies,
except for marginal vitamin A status, in these disadvantaged preschoolers
attending daycare. To our knowledge, our data are the first in Brazil to examine
concurrently biomarkers of vitamin A, zinc, selenium, vitamin B-12, and folate
status in this age group, and to investigate their relationship with parasitic
infections.

6.4.1 Vitamin A, zinc, and selenium status

Only 3% of the preschoolers had serum retinol concentrations indicative of


vitamin A deficiency (i.e., < 0.7 µmol/L), in contrast to much higher prevalence
rates (i.e., > 30%) reported in earlier studies in NE Brazil (da Silva Prado et al.
1995; Martins et al. 2004). These findings suggest that at least among these
disadvantaged daycare preschoolers in Salvador, vitamin A deficiency is no
longer a severe public health problem. Nevertheless, nearly 45% of the children
had marginal vitamin A status (0.70–1.05 µmol/L), which may account for the
lower serum retinol concentrations for the percentile values of the Brazilian
preschoolers compared with those for children of similar age who participated in
the US NHANES III survey (Figure 6.1 and Figure 6.2).
Vitamin A plays a vital role in erythropoiesis (Zimmerman 2007), as well as
mucosal immune function (Semba 1994) and epithelial integrity (Maggini et al.
2007), and even marginal biochemical vitamin A status has been positively
associated with low Hb concentrations (Thurlow et al. 2006). Other investigators
have also reported a high prevalence of marginal vitamin A status (but not
deficiency) both in preschoolers in daycares and in the community in recent

119
Table 6.6 Predictors of serum vitamin B-12, and serum and erythrocyte folate
based on multiple regression models using age, sex, other micronutrients, chronic
inflammation and parasite status, as shown by ß-coefficient (95% CI)

Micronutrient Univariate ß Multivariate ß (95% CI) p

Serum vitamin B-12a


Age -33.701 -33.06 (-67.02, 0.90) 0.055
Sex (male) -62.251 -49.44 (-144.02, 45.14) 0.256
AGP > 1 g/L 38.33
Helminths -116.012 -97.63 (-188.33, -6.93) 0.038
Giardia intestinalis -70.562 -49.12 (-102.88, 4.64) 0.068

Serum folate log transformedb


Age 0.02
Sex (male) -0.121 -0.12 (-0.26, 0.02) 0.088
Helminths 0.04
Giardia intestinalis 0.141 0.14 (0.01, 0.28) 0.041

Red blood cell folate


Age 26.62
Sex (male) -128.16
Helminths 56.33
Giardia intestinalis 68.23

1
p < 0.2
2
p < 0.05
R2 for multivariate regression model: a 0.063; b 0.036

120
studies in NE Brazil (Fernandes et al. 2005; Vieira et al. 2008; de Azevedo et al.
2010).
Coverage of vitamin A supplements in the daycare preschoolers was about
50%. Whether this same coverage extends to preschoolers who do not attend
daycare is unknown. We found no association between serum retinol
concentrations and vitamin A supplementation. This finding is not surprising
because the impact of periodic vitamin A supplementation on improvements in
serum retinol is transient, often lasting for less than six months, and sometimes
even less in settings where additional stresses of infectious diseases may induce
urinary excretion of retinol (Stephensen et al. 1994; Sommer & West 1997).
Unlike other studies (Jalal et al. 1998; Quihui-Cota et al. 2008), neither
parasite infections nor deworming had a significant impact on the vitamin A status
of the preschoolers of this study, even though nearly 30% were parasite-infected
and approximately 50% had received deworming treatment. However, few
children (i.e., 11%) were infected with Ascaris lumbricoides, the helminth
infection specifically associated with malabsorption of vitamin A (Taren et al.
1987; Koski & Scott 2001). Instead, the marginal vitamin A status noted among
nearly 50% of the daycare preschoolers was probably associated with both an
inadequate supply of vitamin A in the daycare meals, as reported in Chapter 4,
and poor coverage of vitamin A supplementation. Certainly, low intakes of
vitamin A have been reported in other daycare studies in NE Brazil (Fernandes et
al. 2005; de Azevedo et al. 2010).
Few children (i.e., < 4%) were classified here as zinc deficient, based on the
recommended cut-off (< 9.9 µmol/L) for serum zinc for non-fasting blood
samples of children less than ten years of age (Hotz & Brown 2004). Further, the
percentile values for serum zinc concentrations for the Brazilian preschoolers
were higher in comparison with those for children of the same age who
participated in the US NHANES II (1976-1980) study (Figure 6.3). However, we
recognize that our data were based on fasting blood samples known to generate
higher serum zinc concentrations than non fasting samples (Wallock et al. 1993),
whereas the data for children in US NHANES II were non-fasting samples, some

121
Figure 6.1 Comparison of serum retinol concentrations for male children
between Brazilian daycare study (n=188) and US NHANES III (1988-1994)
participants ages 4-8 years1

Figure 6.2 Comparison of serum retinol concentrations for female children


between Brazilian daycare study (n=170) and US NHANES III (1988-1994)
participants ages 4-8 years1

1
From Ballew et al. (2001)

122
of which were taken in the morning and some in the afternoon, when values are
even lower. Hence, differences in both fasting status and time of day may account
in part for the higher percentile values compared to the US NHANES II, as well
as the apparent low prevalence of zinc deficiency reported here (Arsenault et al.
2011). To date, specific recommendations for cut-offs for serum zinc
concentrations based on fasting blood samples for young children less than ten
years of age, particularly preschoolers, is lacking.
It is of interest that serum zinc was a significant determinant of serum
retinol (Table 6.5). A similar relationship has been reported in other studies of
children living in poor environments (Muñoz et al. 2000; Rahman et al. 2002;
Thurlow et al. 2006), although in those studies the prevalence of zinc deficiency
was much higher than that reported here. Zinc plays a role in the hepatic release,
transport, and tissue utilization of vitamin A via its role in the synthesis of RBP in
the liver, and through the action of retinol dehydrogenase, a zinc-dependent
enzyme necessary for the oxidative conversion of retinol to retinaldehyde, a
critical step in the visual cycle in the retina of the eye (Christian & West 1998;
Lönnerdal 2004).
Few other studies in Brazil have examined the zinc status of this age group.
Ferraz et al. (2007) reported a very low prevalence of zinc deficiency (i.e., 0.5%)
in healthy children aged 24 to 72 months in São Paulo, whereas in low income
children aged 12 to 59 months in Brasilia, all the children were classified as zinc
deficient (Silva et al. 2006). Problems of adventitious contamination during the
blood sample collection, uncertainties in the cutoff values used (Silva et al. 2006),
and differences in the age range of the children and study settings may account for
these discrepancies: younger children tend to be more at risk to zinc deficiency
(Krebs & Hambidge 2007; Villalpando et al. 2003b). Certainly in the national
survey in Mexico, children aged five to six years had a lower prevalence of zinc
deficiency than their younger counterparts aged six months to four years of age
(21% vs. 33%, respectively), with those from the poorest regions being more
affected.

123
Figure 6.3 Comparison of serum zinc concentrations between Brazilian
daycare study (n=358) and US NHANES II (1976-1980) participants ages 3-8
years1

1
From Pilch & Senti (1984)

124
Inadequate intakes of zinc are considered to be the most likely cause of zinc
deficiency, and arise from low levels in the diet and / or poor bioavailability (Hotz
2007). In the daycare menus, cellular animal protein provided daily (80 g/d) was
the major source of readily available zinc (30%) and as result, phytate-to-zinc
molar ratios were low (i.e., 7.9). Consequently, both the daily supply and potential
bioavailability of zinc in the daycare meals appeared adequate (Chapter 4).
Therefore, it is not surprising that risk of zinc deficiency among these
preschoolers, who were provided with all their weekday meals in the daycare, was
very low compared to the preschoolers in the Mexican national survey. The major
source of dietary zinc for the Mexican preschoolers was unrefined maize, which is
high in phytate, a potent inhibitor of zinc absorption (Villalpando et al. 2003b).
Serum zinc concentrations are known to be reduced by inflammation
(Brown 1998; Wieringa et al. 2002). However, we were unable to detect such an
effect, even though about 30% of the preschoolers had elevated AGP levels
indicative of chronic infection. Indeed, the only determinant of serum zinc noted
here was serum selenium, and the effect reported was modest (Table 6.5). Only
10% of the daycare preschoolers had serum selenium concentrations unlikely to
be adequate for the optimal activity of at least one of the selenoproteins (Thomson
et al. 2007). Selenoproteins, specifically glutathione peroxidases, have the
potential to impact on zinc status through their role in regulating the release of
redox-inert zinc from metallothionein to the zinc-dependent enzyme, copper-zinc
superoxide dismutase (Maret 2000; Lyons et al. 2004). Hence, this mechanism
might account for the modest positive association between serum selenium and
serum zinc documented here, as reported earlier in studies of children (Krittaphol
et al. 2006; Lander et al. 2008; Gibson et al. 2011) and the elderly (de Jong et al.
2001).
Despite previous reports (McLachlan et al. 2004; Rubin et al. 2004), neither
maternal smoking, exposure to household smoke, or family history of asthma
influenced selenium status. Instead, helminth infections, evident in 18% of the
preschoolers, and elevated AGP concentrations each had an independent and
negative impact on serum selenium, although the impact of AGP was modest.

125
Infestation with helminths (e.g., Ascaris lumbricoides and Trichuris
trichiura) can cause malabsorption of micronutrients by damaging the mucosal
wall of the small intestine and reducing its absorptive surface (Solomons 1993;
Crompton & Nesheim 2002). It is conceivable that the response by circulating
levels of selenium to malabsorption was greater than that for retinol or zinc,
biomarkers with poor sensitivity in populations where there is little evidence of
biochemical deficiencies (Gibson 2005). The relatively modest negative effect of
elevated AGP on serum selenium concentrations may be attributed to mobilization
of selenium from the liver and serum to muscle during an acute phase reaction
(Maehira et al. 2002). The reason for the higher selenium status of the boys
compared to the girls is uncertain.

6.4.2 Vitamin B-12 and folate status

Parasitic infections were predictors of both the vitamin B-12 and folate
status of the preschoolers, but the effect, although negative for B-12, was positive
for folate status. The lower vitamin B-12 status observed with infections of
helminths, and to a lesser extent Giardia intestinalis, is well recognized (Brasitus
1983; Hjelt et al. 1992; Allen et al. 1995; Olivares et al. 2002). Such infections are
not only associated with changes in small intestine morphology that lead to
reductions in the absorptive surface (Müller & von Allmen 2005), but also with
increased bacterial overgrowth that causes malabsorption, specifically of vitamin
B-12 (Allen et al. 1995). Risk of bacterial overgrowth is especially high among
these disadvantaged preschool children living in such poor urban slum
environments (dos Reis et al. 2007). In the present study, older children tended to
have a lower vitamin B-12 status, albeit within normal range. Although an inverse
relationship between serum B-12 and parasitic infections among all children was
found, it is possible that the daycare meals provided sufficient vitamin B-12 to
maintain normal status, despite potential vitamin B-12 malabsorption.
The positive association found between serum folate status and Giardia
infection in this study was unexpected but is not implausible. Serum folate levels
(Table 6.3), indicative of recent folate intake, were comparable to the high levels

126
reported for primary school children in North America who also consume flour
fortified with folic acid (Pfeiffer et al. 2007; Colapinto et al. 2011). It is possible
that such a folate-rich environment may have provided optimal conditions for the
adherence of Giardia trophozoites to the intestinal epithelium (Sousa et al. 2001;
Khademi et al. 2006). Alternatively, circulating unmetabolized folic acid may be
present in these children, as a result of the mandatory fortification of flour with
folic acid (Smith 2010). Higher levels of unmetabolized folic acid have been
associated with decreased innate immune function in a study of postmenopausal
women (Troen et al. 2006). This finding was attributed to reduced natural killer
cell cytotoxicity, a mechanism responsible for targeting invading pathogens
(Wintergerst et al. 2007); however, to date these findings have not been replicated
in other studies. Clearly, further research is warranted to confirm this association
and establish the underlying biological mechanisms.

6.4.3 Strengths and limitations

Our study has both strengths and limitations. We examined multiple


micronutrients in conjunction with socio-economic and health status, two
biomarkers of inflammation, and parasite infections in these preschoolers, which
together provide a greater understanding of non-nutritional factors affecting their
micronutrient status. Our blood samples were fasting and collected at the same
time of day in all the preschoolers, thus reducing the variance in serum zinc
concentrations because of sampling. Further, we followed the guidelines of
IZiNCG for the collection and analysis of blood samples for serum zinc to avoid
adventitious sources of contamination. Some of our study limitations relate to the
sampling and study design. Sampling was restricted to seven philanthropically
funded daycares so our findings cannot be extrapolated to other daycare settings.
Finally, our data are observational and hence preclude causal inferences from
being made.

127
6.4.4 Conclusion

In conclusion, these preschoolers had a low prevalence of micronutrient


deficiencies which was probably a reflection of their micronutrient-rich daycare
meals together with the practice of mandatory folic acid fortification of wheat and
corn flour in Brazil (Chapter 4). Hence, our findings suggest that at least two of
the recent pro-poor social policies in Brazil have had some positive influence on
the nutritional status of the disadvantaged preschoolers studied here.
Nevertheless, marginal vitamin A status persisted in 50% of the children, and
parasite infections, namely helminths and Giardia intestinalis, were associated
with lower serum selenium and vitamin B-12 status. These findings emphasize the
importance of continued efforts to improve coverage of vitamin A
supplementation, extend parasite control measures to include deworming, and
improve water supply and sanitation in poor urban settings. Finally, the positive
association between Giardia intestinalis and folate status noted here warrants
further investigation in view of the postulated mechanisms proposed to account
for this finding.
The existence of both parasitic infections and marginal vitamin A status in
these preschoolers has the potential to compromise their iron status. Further, as
94% of the children were of African or African-Brazilian descent, they may be
predisposed to certain genetic Hb disorders known to be associated with anaemia.
Clearly, the iron and anaemia status of these preschoolers also warrants
investigation.

128
7 Predictors of haemoglobin and iron status
in disadvantaged preschoolers attending
daycares in Salvador, NE Brazil

7.1 Introduction
The high prevalence of anaemia (i.e., ~ 50%) among Brazilian children
under-five years of age is assumed to be caused by iron deficiency and has been of
great concern in recent years (Assis et al. 2004b; de Almeida et al. 2004; Vieira et
al. 2007; Jordão et al. 2009; Carvalho et al. 2010). In response, the Brazilian
government has implemented mandatory iron fortification of wheat and corn flour
products at the national level (Brazil Ministry of Health 2002) and universal oral
iron supplementation for young children aged 6–18 months (Brazil Ministry of
Health 2005), in an effort to combat the negative health consequences of anaemia
and iron deficiency. Impairments in motor and cognition development have been
reported in children with iron deficiency anemia (Grantham-McGregor & Ani
2001; Akman et al. 2004; Carter et al. 2010), some of which have the potential to
be irreversible after two years of age (Lozoff 2007). Such adverse effects arise
because of the rapid neurological development in children from 0–4 years and the
key role of iron in the development and maturation of brain neurotransmitters
(Beard 2003; Yehuda et al. 2006).
There is some evidence that in certain circumstances iron supplementation
can have harmful consequences. For example, excessive iron supplements in iron-
replete infants and young children can increase the risk of morbidity from certain
infections by enhancing the growth of iron-dependent pathogens (Sazawal et al.
2006; Ratledge 2007; Prentice et al. 2008), and contribute to reductions in
cognition (Lozoff et al. 2012) and possibly growth (Dewey et al. 2002).
Furthermore, iron supplementation in populations at risk of genetic Hb disorders
can result in iron overload (Hillman et al. 2005) due to ineffective erythropoiesis
in the bone marrow, which stimulates iron absorption even if iron stores are

129
adequate (Zimmermann et al. 2008a). This is of particular concern in NE Brazil
where there is a high prevalence of some genetic Hb disorders because a large
proportion of the population has West African ancestry.
Other factors, such as overweight and coffee intake, can also negatively
affect iron status in young children. Overweight is associated with conditions of
chronic inflammation, leading to reduced iron absorption and functional iron
deficiency (Zimmermann et al. 2008b; Aeberli et al. 2009). Consumption of
coffee inhibits non-haem iron absorption because it contains polyphenols (Ruel
2001), which form insoluble iron-phenolic compounds in the gastro-intestinal
tract (Gibson & Ferguson 2008). Brazilian children living in low-income settings
with high food insecurity reportedly drink large amounts of coffee (Antunes et al.
2010), which has been associated with sub-optimal iron status in other Latin
American countries, such as Guatemala (Dewey et al. 1997).
In many low income countries in Latin America, deficiencies of other
micronutrients including vitamin A, vitamin B-12, and folate that play a key role
in the pathogenesis of anaemia, have been reported (Allen et al. 2000; López de
Romaña et al. 2005). Vitamin A deficiency adversely affects iron-dependent
erythropoiesis via compromising the absorption, storage, or transport of iron into
the marrow, reducing erythrocytosis by lowering erythropoietin production, or
inducing iron sequestration as a result of inflammation (Bloem 1995; Semba &
Bloem 2002; Zimmerman et al. 2006). Although the prevalence of vitamin A
deficiency in NE Brazil has been reduced by increased coverage of vitamin A
supplementation, even marginal vitamin A deficiency plays a role in anaemia
(Osório et al. 2004). Both vitamin B-12 and folate are involved in DNA synthesis,
so deficiency impairs cell replication, resulting in abnormally large red blood cell
precursors (megaloblasts) and megaloblastic anaemia (Scott 2007). To date, there
is a paucity of data on the vitamin B-12 and folate status of preschoolers in NE
Brazil.
Intestinal parasitic infestations with soil-transmitted helminths and
pathogenic protozoa have also been associated with anaemia in children in Brazil
(Brooker et al. 2007; Silva et al. 2009), and elsewhere (Stoltzfus et al. 1997b;

130
Jiménez et al. 1999; Le et al. 2007). For some of these parasitic infections (e.g., T.
trichiura and hookworm), anaemia is induced by blood loss, whereas in others
(e.g., A. lumbricoides and G. intestinalis) malabsorption is a contributing factor
because the parasites reside in the duodenum and jejunum where most of the iron
is absorbed (Müller & von Allmen 2005; Thurnham & Northrop-Clewes 2007).
In Salvador, Brazil’s third largest city, a high proportion (~80%) of the
population is of West African descent (Gonçalves et al. 2003) in whom there is a
high prevalence of certain genetic Hb disorders. However the extent to which
genetic Hb disorders contribute to anaemia in children in Salvador is uncertain.
Therefore, in this pre-selected group of disadvantaged preschoolers aged three to
six years attending daycares in Salvador we aim to: (1) assess the prevalence of
anaemia, iron deficiency, and selected Hb disorders; and (2) identify the major
predictors of Hb and iron biomarkers with emphasis on the role of socio-
demographic factors, BMI, parasitic infections and genetic Hb disorders. Data
from this study, together with the average daily supply of iron and other
micronutrients from the daycare menus (Chapter 4), will provide insight as to
whether the government strategies to prevent anaemia and iron deficiency are
appropriate for these disadvantaged preschoolers in Salvador.

7.2 Participants and Methods

7.2.1 Study sites and participants

This cross-sectional study was conducted in Salvador, the capital city of the
state of Bahia, located in NE Brazil between August – November, 2010. Salvador
is a city with a population of three million with ~85% of the residents from
African or African/Brazilian descent (IBGE 2010a). Seven philanthropically
funded preschool daycares located within the city centre and in peri-urban areas
were pre-selected for participation in this study, as described elsewhere (see
Chapter 3). The children (n=376) enrolled in the daycares (maximum class size of
25 children) were from low-income families and attended daycare five days per

131
Table 7.1 Socio-economic and health characteristics of the participants
and prevalence of intestinal parasitic infections

n %

Maternal age (years) 368/376 29.7 (7.0)1

Socio-economic status (extremely low) 182/376 48.4

Ethnicity
Black 154/365 42.2
Brown 189/365 51.8
White 22/365 6.0

Smoking
Mother smoking in the house 47/376 12.5
Adult smoking in the house 89/376 23.7
History of asthma in mother or sibling 68/376 18.1

Iron sryup within 6 months 70/376 18.6

Vitamin A supplementation (ever) 200/376 53.2

Coffee intake within past 24 hours 149/376 39.6

Deworming treatment within 6 months 192/376 51.1

Parasite present 95/325 29.2


Helminths 58/325 17.8
Trichuris trichiura 39/325 12.0
Ascaris lumbricoides 34/325 10.5
Hookworm 3/325 0.9
Giardia intestinalis 42/325 12.9

1
Mean (SD)

132
week, except holidays, until school age. The study protocol was approved by the
responsible Ethics committees and overarching daycare organizations, as
described in section 3.2.1, and informed written consent to participate in the study
was given by the parents or primary guardians of the children.

7.2.2 Assessment of socio-demographic and health status

A culturally appropriate structured questionnaire was developed to collect


information on the socio-demographic, health, and current dietary characteristics
of the participants, as described earlier (Chapter 3). The ethnicity of the children
was determined by skin colour, and hair and facial characteristics, as performed in
the 2010 national census (IBGE 2010a) (see section 2.4.2). Information on health
status variables for each child included maternal reports of their child’s coffee
intake within the past 24 hours, use of iron supplements and deworming treatment
within the past six months, and smoking by the mother or another adult in the
house. The child’s health card, provided either by the caregiver or the daycare,
was reviewed to provide birth weight.

7.2.3 Anthropometric assessment

Anthropometric measurements, including weight and height, were taken


with children wearing light clothes and no shoes using standardized techniques
and calibrated equipment (WHO 2004a), as described in Chapter 3. For all
children, z-scores were calculated for body-mass-index (BMIZ) using the updated
US CDC 2000 Growth Reference in the United States (Kuczmarski et al. 2000).
Overweight or obesity in the children was based on BMIZ > 1 to ≤ 2SD or > 2SD,
respectively.

7.2.4 Assessment of parasite status

A stool sample was submitted for parasite analysis by 86% (325/376) of the
participants. A faecal concentrate was prepared as described in Chapter 5 and
examined for the presence of helminths and protozoal intestinal parasites by

133
Table 7.2 Quality control data for the assessment of accuracy and precision
of the iron biomarkers

Measured Mean Certified Range CV


(SD) (%)

Serum ferritin (µg/L)


Control A1 25.0 (0.5) 19.7–30.1 2.0
Control A2 346.0 (8.3) 265–407 2.4
Control A3 698.6 (15.8) 546–836 2.3
Pooled serum 54.9 (1.8) 3.3

Serum TfR (mg/L)


Normal control 5.6 (0.4) 4.5–6.1 7.6
High control 17.1 (1.0) 14.0–18.5 6.0

CV = Coefficient of variation equation (sd*100) / mean) %

134
microscopy. An ELISA assay was used to identify positive samples for Giardia
intestinalis, also described previously.

7.2.5 Biochemical assessment

Morning fasting blood samples were obtained by peripheral venipuncture by


trained phlebotomists with participants in the sitting position, either held by their
mother or caregiver or sitting alone. A topical local vasodilator anesthetic
amethocaine (AmetopTM) was applied to the venipuncture site 30 minutes prior to
blood sampling in order to minimize any discomfort. Blood was drawn into a
trace-element (TE)-free evacuated tube (Becton Dickinson, Franklin Lakes, NJ,
USA) for serum ferritin and TfR assays and into a pediatric evacuated tube
containing EDTA as an anticoagulant (Becton Dickinson, Franklin Lakes, NJ) for
a complete blood count (CBC) and screening for genetic Hb disorders. All blood
samples were refrigerated immediately following collection, and the serum
separated within two hours using TE-free techniques. The CBC was determined
using an automatic electronic analyzer (Coulter LH 750 Hematology Analyzer) in
the Professor Edgar Santos Hospital haematology laboratory in Salvador.
Aliquots of washed red blood cells for screening of genetic Hb disorders were
stored in TE-free polyethylene vials and frozen immediately, initially at -30°C and
later at -70°C. Frozen samples were shipped on dry ice to the Trace Element
Laboratory of the Department of Human Nutrition and to the Department of
Pathology, University of Otago, New Zealand for analysis.
Serum ferritin was determined using a commercial kit (Ferritin Elecsys
reagent kit, Roche Diagnostics GmbH, Mannheim Germany) carried out on an
auto-analyzer (Roche, New Zealand). Serum soluble TfR was analyzed via an
enzyme immunoassay by using commercial kits (Ramco Laboratories Inc,
Houston, TX). Serum CRP and AGP concentrations were assayed by
immunoturbidimetry (Roche Diagnostics) on an automated analyser system
(Cobas Mira II), as described previously. The presence of haemoglobinopathies
were determined by alkaline and acid Hb electrophoresis analysis (Sebia Hydrasys

135
Table 7.3 Prevalence (% and 95% CI) of haemoglobinopathies and α3.7
thalassemia in the preschoolers

n % (95% CI)
Overall prevalence of 101/311 32.5 (27.3, 38.0)
selected Hb disorders

Haemoglobinopathies
Hb AS 20/358 5.6 (3.4, 8.5)
Hb AC 9/358 2.5 (1.2, 4.7)

α3.7 thalassemia
Heterozygous 70/306 22.9 (18.3, 28.0)
Homozygous 6/306 2.0 (0.7, 4.2)

Table 7.4 Prevalence of haemoglobinopathies and α3.7 thalassemia in the


preschoolers by ethnicity

Hb disorder Black Brown White p

n (%) n (%) n (%)


Haemoglobinopathies
Hb AS 12 (60.0) 6 (30.0) 2 (10.0) 0.084
Hb AC 4 (44.4) 5 (55.6) 0 (0.0) 1.00

α3.7 thalassemia
Heterozygous1 32 (48.5) 30 (45.5) 4 (6.0) 0.388
Homozygous 3 (50.0) 3 (50.0) 0 (0.0) 1.00

1
Participants with concomitant Hb AS or Hb AC were excluded (n=4)

136
electrophoresis analyzer), a process designed for the separation of normal
haemoglobins (A, A2 and F) and detection of major Hb variants, including the
heterozygous variant of Hb S (Hb AS) and of Hb C (Hb AC). The presence of α3.7
thalassemia was determined via DNA extraction and multiplex PCR reaction (Tan
et al. 2001), as described elsewhere (Alsaleh 2011).
The precision of the biochemical assays was checked using a pooled serum
sample and their accuracy established using certified reference materials or
appropriate manufacturers’ controls. The between-assay coefficient of variations
(CV as %) for the pooled serum for ferritin and TfR were 3.3% and 7.6%,
respectively. Values for the certified reference materials or manufacturers’
controls fell within the certified ranges for ferritin and TfR; the values for CRP
and AGP compared to the certified ranges are shown in Chapter 6. A positive Hb
control (Haemoglobin FASC Variant Control, Canterbury Scientific, NZ) was
used for the electrophoresis analysis. The DNA from a known New Zealand
heterozygote α3.7 thalassemia patient was used as a positive control for the PCR
reactions and a negative control was included to confirm the absence of
contamination for the PCR results, as described elsewhere (Alsaleh 2011).
Anaemia and other haematological disturbances were defined using the
following interpretive criteria: Hb < 110 g/L and < 115 g/L for children < 5 years
and ≥ 5 years, respectively (WHO 2001); mean cell volume (MCV) < 73 fL and <
74 fL for children < 5 years and ≥ 5 years, respectively (WHO 2001); mean cell
Hb (MCH) < 25 pg (WHO 2001); and red cell distribution width (RDW) > 14%
(Gibson 2005). For storage iron depletion in the absence of anaemia, serum
ferritin was defined as < 12 µg/L and < 15 ug/L for children < 5 years and ≥ 5
years, respectively (WHO 2001). Tissue iron deficiency was defined as serum
soluble TfR > 8.5 mg/L (Cook et al. 1993). Presence of acute or chronic
inflammation was assessed by serum CRP > 5 mg/L (Thurnham et al. 2005) or
serum AGP > 1.0 g/L (WHO/CDC 2007), respectively. The abnormal
haemoglobinopathies, Hb AS and Hb AC, and α3.7 thalassemia, homozygous or
heterozygous, were identified as present or absent.

137
Table 7.5 Mean (SD) values for haematology and iron biomarkers for normal Hb type (Hb AA) and the four major
abnormal Hb variants for the preschoolers

α3.7thal α3.7thal
Hb AA Hb AS Hb AC
heterozygous1 homozygous p
n=210 n=20 n=9
n=66 n=6

Hb (g/L) 124.1 (7.1) 122.5 (6.6) 121.9 (13.5) 119.4 (6.5) 111.8 (4.4) < 0.001

MCV (fL) 80.7 (3.2) 79.4 (4.1) 72.9 (5.1) 75.4 (3.4) 63.8 (2.4) < 0.001
138

MCH (pg) 27.4 (1.2) 26.7 (1.5) 24.6 (2.0) 25.0 (1.3) 20.7 (0.7) < 0.001

RDW (%) 13.9 (1.0) 13.9 (2.6) 15.2 (0.9) 14.4 (1.5) 16.4 (1.1) < 0.001

Ferritin2(µg/L) 41.3 (19.9) 43.6 (18.5) 34.1 (10.2) 43.6 (17.2) 38.3 (14.4) 0.084

sTfR (mg/L) 7.1 (1.5) 8.1 (2.6) 7.8 (1.5) 7.4 (1.7) 7.2 (1.4) 0.259

1
Participants with concomitant Hb AS or Hb AC were excluded (n=4)
2
Adjusted for infection (Thurnham et al. 2010)
Abbreviations used: Hb, haemoglobin; MCV, mean cell volume; MCH, mean cell haemoglobin; RDW, red cell
distribution width; sTfR, soluble transferrin receptor
7.2.6 Statistical analysis

The mean (SD) haematology and iron biomarker concentration were


calculated, and adjusted where necessary for the effect of inflammation (i.e.,
ferritin) based on recommendations from Thurnham et al (2010). Associations
between elevated AGP concentrations (> 1 g/L) and BMIZ > 1SD, and between
the presence of genetic Hb disorders and ethnicity were investigated using
Fisher’s exact test. AGP was used as a dichotomized variable because the
sensitivity of the assay precluded using the data as a continuous variable. The data
were log transformed if the variable had a strongly skewed distribution (e.g.,
ferritin and soluble TfR). Multiple regression models included variables if
associations using univariate regression analysis were p<0.2. The following
variables were included: age, sex, ethnicity, abnormal Hb variants, micronutrients,
chronic inflammation (i.e., AGP > 1 g/L), helminths, and BMIZ > 1SD. The
sandwich estimator was used to obtain robust standard errors, to account for the
sampling procedure. Statistical analyses were carried out using STATA version 11
(Stata Corporation, College Station, TX, USA).

7.3 Results

7.3.1 Socio-demographic and health characteristics

A summary of the participants’ SES and health characteristics and the


prevalence of parasitic infections are shown in Table 7.1, as described previously.
The ethnicity of the preschoolers was black (42.2%), brown (51.8%), and white
(6.0%). Of the children, 18.6% (70/376) and 51.1% (192/376) had received
supplementation with iron syrup and deworming treatment, respectively, within
the past six months. Coffee had been consumed by nearly 40% (149/376) of the
children within the past 24-hours, and nearly 25% of the households reported an
adult smoking in the house. Prevalence of overweight and obesity in the children
was 10.7% (39/364) and 2.5% (9/364), respectively.

139
Table 7.6 Prevalence (% and 95% CI) of selected haematological variables,
iron biomarkers, and elevated biomarkers of inflammation

Cut-off values n % (95% CI)


Hb
< 5 years of age < 110 g/L 11/319 3.4 (1.7, 6.1)
≥ 5 years of age < 115 g/L 3/40 7.5 (1.5, 21.9)

MCV
< 5 years of age < 73 fL 31/319 9.7 (6.7, 13.5)
≥ 5 years of age < 74 fL 3/40 7.5 (1.5, 21.9)

MCH < 25 pg 65/359 18.1 (14.3, 22.5)

RDW > 14% 168/359 46.8 (41.5, 52.1)

Serum ferritin1
< 5 years of age < 12 µg/L 6/318 1.9 (0.7, 4.1)
≥ 5 years of age < 15 µg/L 1/40 2.5 (0.1, 13.9)

Serum soluble TfR > 8.5 mg/L 54/358 15.1 (11.5, 19.2)

AGP > 1.0 g/L 112/358 31.3 (26.5, 36.4)

CRP > 5 mg/L 46/358 12.9 (9.6, 16.8)

1
Adjusted for infection (Thurnham et al. 2010)
Abbreviations used: Hb, haemoglobin; MCV, mean cell volume; MCH, mean cell
haemoglobin; RDW, red cell distribution width; TfR, transferrin receptor; AGP,
α-1-glycoprotein; CRP, C-reactive protein

140
7.3.2 Assessment of genetic Hb disorders, haematological and biochemical
iron indices

The quality control data for the iron biomarkers are shown in Table 7.2. The
overall prevalence (95% CI) of children with genetic Hb disorders was 32.5%
(27.3, 38.0) (Table 7.3). The presence of the haemoglobinopathies – Hb AS and
Hb AC – was low (i.e., 5.6% (3.4, 8.5) and 2.5% (1.2, 4.7), respectively).
Likewise, the prevalence of homozygous α3.7 thalassemia was low (i.e., 2.0% (0.7,
4.2), although 22.9% (18.3, 28.0) of the children were heterozygous α3.7
thalassemia. Only four children (i.e., 1.1%) had both a haemoglobinopathy and
α3.7thalassemia present, of whom two were ethnically black and two brown. The
prevalence of Hb AS was modestly higher (p=0.084) in the black group (i.e., 60%,
12/20) compared to the brown (i.e., 30%, 6/20) and white (10%, 2/20) groups;
however, no significant difference was noted between ethnicity and those with Hb
AC or α3.7 thalassemia (Table 7.4).
The mean (SD) concentrations of these biomarkers based on normal Hb (Hb
AA), haemoglobinopathies Hb AS and Hb AC, and α3.7 thalassemia (heterozygous
and homozygous) are presented in Table 7.5. Mean (SD) serum ferritin
concentrations were normally distributed after adjustment for acute and chronic
inflammation (CRP > 5mg/L and AGP > 1g/L, respectively). Significant
differences existed across the groups for Hb, MCV, MCH, and RDW (p<0.001 in
all cases), although for serum ferritin the differences were only modest (p=0.084).
The children with normal Hb type (Hb AA) had a greater mean Hb, MCV, and
MCH and a lower RDW compared to those with haemoglobinopathies Hb AS and
Hb AC and / or α3.7 thalassemia. Of the four abnormal genetic Hb disorders, the
children with homozygous α3.7 thalassemia had the lowest mean Hb, MCV, and
MCH and the highest RDW. The mean soluble TfR was highest in the Hb AS
group, whereas the adjusted mean serum ferritin was lowest in the Hb AC group,
although the serum ferritin values varied widely.
The prevalence of anaemia and storage iron depletion was low (i.e., ≤
7.5%), as indicated by a low Hb, and a low serum ferritin, respectively (Table

141
Table 7.7 Predictors of Hb based on multiple regression models using
abnormal Hb variants and micronutrient status, as shown by ß-coefficient
(95% CI)

Haemoglobin univariate ß multivariate ß (95% CI) p

α3.7 thalassemia homozygous -12.292 -12.19 (-14.78, -9.60) <0.001


α3.7 thalassemia heterozygous -4.842 -5.28 (-7.44, -3.11) 0.001
HbAS -0.44
HbAC -1.00
Selenium 11.712 11.71 (6.26, 17.17) 0.001
Retinol (log transformed) 7.622 5.08 (0.65, 9.51) 0.030
Ferritin (log transformed) 2.111 2.35 (-0.76, 5.46) 0.117
Zinc 0.712 0.42 (-0.09, 0.92) 0.092
AGP > 1 g/L 0.011 0.29 (-2.09, 2.68) 0.779

1
p < 0.2
2
p < 0.05
R2 for multivariate regression model: 0.242

142
7.6). As expected, the prevalence of anaemia among children with one of the
selected Hb disorders was significantly higher than children with a normal Hb
type (i.e., 69%, 9/13 vs. 31%, 4/13; p=0.006). An elevated soluble TfR indicative
of tissue iron deficiency was found in 15% (54/358) of the children, of whom
30% (16/54) had a genetic Hb disorder. Only 13% (46/358) of the children had
high CRP concentrations, whereas nearly a third of the children (112/358) had
elevated AGP levels. Furthermore, the prevalence of elevated AGP concentrations
(i.e., > 1.0 g/L) was significantly greater among the overweight / obese children
(i.e., BMIZ > 1SD) compared to those with BMIZ < 1 SD (i.e., 47%, 21/45 vs.
29%, 90/307; p=0.025), respectively.

7.3.3 Predictors of Hb and iron biomarker concentrations

Table 7.7 highlights the predictors of Hb, including the significant inverse
association observed between Hb and α3.7 thalassemia (p≤0.001 for both types).
The major positive predictors of Hb were serum selenium (p=0.001), retinol
(p=0.030), and zinc (0.092), although the latter association was only modest. Only
one positive predictor of serum ferritin was found: an elevated AGP (i.e., > 1 g/L)
(p=0.001) (Table 7.8). Serum soluble TfR was positively associated with Hb AS
(p=0.018) and BMIZ > 1SD (p=0.049), with a modest inverse association with
age (p=0.070) (Table 7.8). No associations were found between either Hb or iron
status with maternal age, birth weight, sex, SES, ethnicity, helminths,
supplementation with iron or vitamin A, deworming treatment, recent coffee
intake, or smoking in the home.

7.4 Discussion
Our findings highlight the low prevalence of anaemia, iron deficiency
anaemia, and iron deficiency in these disadvantaged preschoolers attending
daycare. To our knowledge, our data are the first in Brazil to examine
concurrently in this age group the complex inter-relationships between Hb and
iron status and other micronutrients, parasitic infections, and genetic Hb disorders.

143
Table 7.8 Predictors of serum ferritin and soluble TfR based on multiple
regression models using abnormal Hb variants, chronic inflammation, parasite
status and descriptive variables, as shown by ß-coefficient (95% CI)

univariate ß multivariate ß (95% CI) p

Serum ferritin log transformeda


Age 0.03
BMIZ >1SD 0.062 0.06 (-0.02, 0.14) 0.134
AGP > 1 g/L 0.312 0.30 (0.16, 0.44) 0.001
HbAS 0.07
HbAC -0.12
α3.7 thalassemia heterozygous 0.04
α3.7 thalassemia homozygous -0.05

Serum soluble TfR


log transformedb
Age -0.042 -0.03 (-0.06, 0.00) 0.070
Sex (male) 0.04
BMIZ >1SD 0.092 0.08 (0.00, 0.15) 0.049
Helminths -0.061 -0.06 (-0.14, 0.03) 0.155
HbAS 0.162 0.18 (0.04, 0.31) 0.018
HbAC 0.09
α3.7 thalassemia heterozygous 0.03
α3.7 thalassemia homozygous 0.01

1
p < 0.2
2
p < 0.05
R2 for multivariate regression model: a 0.098; b 0.083

144
It is of interest that the major significant predictors of Hb were α3.7 thalassemia,
serum selenium and serum retinol, but not serum ferritin.

7.4.1 Prevalence of anaemia, iron deficiency anaemia and iron deficiency

The prevalence of anaemia, iron deficiency anaemia, and iron deficiency


amongst these preschoolers (i.e., < 10%) was much lower than that reported
earlier in children under-five years of age in Salvador (Assis et al. 2004b) and
other cities in NE Brazil (Oliveira et al. 2006; Vieira et al. 2007; Carvalho et al.
2010; Leal et al. 2011). Indeed, some earlier reports suggested that as many as
40% of urban children in NE Brazil were anaemic (Jordão et al. 2009). Several
factors may account for these discrepancies. The children were older, with lower
iron requirements (per kg/body weight) than infants and toddlers (Yip &
Ramakrishnan 2002), which were most likely met by an adequate supply of
bioavailable iron from their daycare meals. The latter provided a daily source of
readily absorbable haem iron (0.75 mg/d) from animal-source foods (80 g/day),
and non-haem iron from legumes and cereals fortified with iron (and folate), both
usually served with a source of vitamin C to enhance non-haem iron absorption
(Chapter 4). Moreover, nearly 20% of the children had received iron supplements
within the past six months. Finally, in many of the earlier studies, the existence of
genetic Hb disorders was not taken into account when interpreting the iron
biomarkers, even though many of the children in NE Brazil are of mixed ethnicity.

7.4.2 Genetic Hb disorders as predictors of Hb

Genetic Hb disorders can lead to structurally abnormal Hb


(haemoglobinopathies) or to a significant decrease in the rate of synthesis of one
or more globin chains (thalassemias) (Bain 2006). Both types can have a negative
impact on Hb and red blood cell indices. As expected among persons mainly of
West African ancestry (Sonati et al. 1991; Borges et al. 2001), almost a third of
the preschoolers had at least one of the genetic Hb disorders studied, the most
frequent (i.e., 25%) being either a homozygous (n=6) or heterozygous state of α3.7

145
Figure 7.1 Comparison of serum ferritin concentrations between the Brazilian
daycare study (n=358) and US NHANES III (1988-1994) participants ages 3-5
years1

1
From Hollowell et al. (2005)

146
thalassemia (n=70). In contrast, none of the preschoolers were homozygous for
Hb S or Hb C, although a few were heterozygous for Hb S (Hb AS, n=20) (6%) or
Hb C (Hb AC, n=9) (3%). In earlier studies of children in Salvador, corresponding
prevalence rates for Hb AS and Hb AC were 10% and 7%, respectively, and 22%
for α3.7 thalassemia (Azevêdo et al. 1980; Adorno et al. 2005).
In Hb S, commonly known as sickle cell Hb, glutamic acid is replaced by
valine at position six on the ß globin chain. During deoxygenation, this defect
results in reduced solubility and polymerization, causing deformed and unstable
erythrocytes shaped like a sickle (Bain 2006). Hb C is caused by a mutation in the
ß globin gene at the same site as the mutation in the sickle cell Hb (Bain 2006). It
is of interest that neither of the Hb AS or Hb AC heterozygotes were significant
predictors of Hb (Table 7.7). This finding is not unexpected because Hb
concentrations of these heterozygotes are usually normal, and thus of little clinical
significance. However, both homozygous and heterozygous states of α3.7
thalassemia were significant negative predictors of Hb (Table 7.7). Indeed, even
the four children with Hb AS or Hb AC in association with α-thalassemia trait had
borderline anaemia and / or microcytosis.
In α3.7 thalassemia, impaired synthesis of the α globin chain arises from the
3.7 kb deletion of all or part of the α-2 globin gene that occurs at the 3’ end of the
α-2 gene and the 5’ end of the α-1 gene (Hillman et al. 2005). Hence, a microcytic
hypochromic state is usually observed with α3.7 thalassemia, with homozygotes
markedly more microcytic and hypochromic than heterozygotes (Rees et al.
1998), as observed here. Indeed, the preschoolers who were homozygous for α3.7
thalassemia (n=6) had the lowest mean values for Hb and MCV (Table 7.5),
despite a mean serum ferritin that was indicative of adequate storage iron and
comparable to that noted for children with a normal Hb type (i.e., Hb AA). Such
hematologic values are consistent with those for children homozygous for α3.7
thalassemia in other studies in Salvador (Adorno et al. 2005; Couto et al. 2003), as
well as elsewhere (Bowden et al. 1985).

147
7.4.3 Micronutrient predictors of Hb

It is of interest that serum selenium and serum retinol were both stronger
predictors of Hb in these preschoolers than serum ferritin. Low selenium status,
noted in 10% of the children (Chapter 6), has the potential to decrease Hb
concentrations (Van Nhien et al. 2008; Semba et al. 2009) via several plausible
mechanisms. Selenium is a potent antioxidant (Burk 2002). Reduced activity of
glutathione peroxidase, the selenoenzyme that protects Hb against oxidation in red
blood cells (Nagababu et al. 2003), results in increased inflammation and
oxidative stress, leading to higher levels of circulating interleukin-6, implicated in
the upregulation of hepcidin. The latter controls both the concentrations of
circulating iron in the serum and distribution of iron in the tissues by inhibiting
the absorption, recycling and mobilization of iron from hepatic stores (Roy &
Andrews 2005; Thurnham & Northrop-Clewes 2007; Young et al. 2009). Low
selenium status may also result in decreased activity of the selenoenzyme
thioredoxin reductase, postulated to be implicated in the upregulation of hepatic
haem oxygenase-1, known to play an initial role in haem catabolism (Mostert et
al. 2003).
Serum retinol was also a significant positive predictor of Hb (Table 7.7).
Nearly 50% of the preschoolers had marginal vitamin A status (i.e., < 1.05
µmol/L, Chapter 6), shown to be associated with low Hb concentrations in several
other studies (van Stuijvenberg et al. 1997; Thurlow et al. 2005). Lower vitamin A
status can impair mobilization of iron from spleen or liver stores into the
circulation (Fishman et al. 2000), a plausible mechanism in view of the apparently
low prevalence of storage iron depletion (i.e., < 3%) in these children. Additional
or alternative mechanisms may be involved, such as a decrease in haematopoiesis
arising from the reduction in the role of vitamin A in the regulation of
erythropoietin (Evans 2005).
The lack of association found between Hb and serum ferritin was
unexpected in view of the major role that iron deficiency has been assumed to
play in the aetiology of anaemia in children in Brazil. Under normal conditions,
serum ferritin levels generally parallel the amount of storage iron in the liver (Mei

148
et al. 2005). However, in this study several factors complicated the interpretation
of the serum ferritin concentrations, including the presence of genetic Hb
disorders, and acute and chronic inflammation.
Unlike some other reports (Gibson et al. 2008; Cole et al. 2010), we found
serum zinc was only modestly associated with Hb, perhaps because so few
children were anaemic or zinc deficient in this study. Several mechanisms have
been proposed that might explain the modest positive effect of zinc on Hb
concentrations noted here. Synthesis of Hb depends on several zinc-dependent
enzyme systems such as aminolevulinic acid dehydrase which has a role in haem
synthesis (Garnica 1981) and thymidine kinase and DNA polymerase which are
involved in DNA synthesis (Spivak et al. 1992). Furthermore, GATA-1, a zinc-
finger transcription factor, is essential for normal hematopoiesis (Labbaye et al.
1995). Besides these direct mechanisms, zinc may also play a role in stabilizing
red cell membranes (O’Dell 2000) and in increasing plasma IGF-1 levels, which
may in turn stimulate erythropoeisis (Nishiyama et al. 1999).
Not unexpectedly, neither serum vitamin B-12 nor serum folate was
associated with Hb concentrations. The daycare meals provided adequate amounts
of vitamin B-12 and folate (Chapter 4) and few children (i.e., < 2%, Chapter 6)
had low serum concentrations of these micronutrients, and no evidence of
megaloblastic anemia, based on an elevated MCV.

7.4.4 Predictors of iron status

The World Health Organization (2001) recommends serum ferritin as the


biomarker for detecting iron deficiency (based on storage iron depletion).
However, ferritin is a positive acute phase protein that increases in response to
chronic inflammation, as shown here (Table 7.8), as a result of the increased rate
of ferritin synthesis in the reticulo-endothelial system (Thurnham et al. 2010). No
other significant predictors of serum ferritin were observed.
In contrast, an abnormal genetic Hb disorder, namely Hb AS, as well as
overweight, were both significant positive predictors of soluble TfR, an indicator
of tissue iron deficiency (WHO/CDC 2007), together with age, although the latter

149
effect was negative and only modest (Table 7.8). Genetic Hb disorders, such as
Hb AS, are associated with ineffective erythropoiesis that stimulates an increase
in iron absorption, and as a consequence, elevated soluble TfR concentrations,
even when iron stores are adequate, as noted here. As a result, tissue iron overload
may occur, which can cause cardiac and hepatic damage (Bain 2006; Rechavi &
Rivella 2008). However, iron overload is unlikely in these preschoolers because
the affected children had the heterozygous form of sickle cell Hb and not the
homozygous form.
The positive relationship between overweight and soluble TfR indicative of
tissue iron deficiency (Table 7.8) was probably associated with the chronic
inflammation that often accompanies overweight and / or obesity, as found here.
Proinflammatory cytokines such as interleukin-6 and leptin up-regulate hepcidin
expression from both the liver and adipose tissue, which in turn down-regulates
intestinal iron absorption independent of iron status, leading to functional iron
deficiency (Table 7.8) (Yanoff et al. 2007). A reduction in the dietary iron supply
per se does not seem to be the primary mechanism, although the exact etiology is
not entirely clear (Aeberli et al. 2009; Cepeda-Lopez et al. 2011). Certainly, the
negative response of overweight children to iron fortification, as reported by
Zimmermann et al. (2008b), is a cause for concern, and may explain in part the
lack of response among anaemic Brazilian preschoolers to iron fortified flour
(Assunção et al. 2007). The negative trend observed between soluble TfR and
increasing age in these preschoolers has been reported elsewhere (Choi et al.
1999; Worwood 2002; Chouliaras et al. 2009). Several reasons may account for
this relationship, including the length of time the preschoolers had attended the
daycare and hence received an adequate supply of both bioavailable haem iron
and non-haem iron from daycare meals.
The absence of any significant negative relationship between serum ferritin
and overweight (Table 7.8), despite the apparent existence of functional iron
deficiency may be associated with several factors, including the effects of genetic
Hb disorders and chronic inflammation on serum ferritin levels, as noted earlier. It
is of interest that we did not find a positive association between serum ferritin and

150
genetic Hb disorders, particularly α3.7 thalassemia (Table 7.8), as might be
expected given the likelihood of ineffective erythropoeisis, and thus tendency
towards increased absorption of dietary iron (Earley et al. 1990; Hillman et al.
2005). However, other factors, not investigated here, may have contributed to the
low prevalence of storage iron depletion, including the presence of glucose-6-
phosphate dehydrogenase deficiency (Wong & Saha 1987). The latter is a genetic
enzyme deficiency commonly affecting African populations (Beutler 2008;
Nkhoma et al. 2009), which has been reported earlier in schoolchildren in
Salvador (Azevédo et al. 1980). Hence, although here we have only been able to
account for the effect of chronic inflammation on serum ferritin concentrations,
several other factors may have complicated their interpretation in this study. This
may have contributed, at least in part, to the seemingly higher serum ferritin
concentrations for the percentile values for these preschoolers compared to the
corresponding percentile values for US preschoolers in the NHANES III survey
(Figure 7.1).
It is notable that we did not find a negative association between Hb and
helminth infections, although others have reported such a relationship (Ramdath et
al. 1995; Stoltzfus et al. 1997b; Le et al. 2007), probably because the prevalence
of helminth infestation, particularly hookworm was low (i.e., < 1%). Moreover,
we did not find an association between iron status and coffee intake, as observed
elsewhere (Dewey et al. 1997), even though nearly 40% of the children had been
given coffee within the past 24 hours. However, the amount, strength, and
frequency of the coffee consumption may have been too low to impair iron status.

7.4.5 Strengths and limitations

Our study has both strengths and limitations. We examined the presence of
genetic Hb disorders, including haemoglobinopathies and thalassemias, as well as
multiple micronutrients, two biomarkers of inflammation, parasites, socio-
economic and health status variables, which together provides a greater
understanding of factors influencing Hb and iron status in these preschoolers.
Some of our study limitations relate to the sampling and study design. Sampling

151
was restricted to seven philanthropically funded daycares so our findings may not
apply to other daycare settings. Furthermore, our investigations were limited to
specific genetic Hb disorders, and did not include other genetic enzyme disorders
that may also affect this population. Finally, our data are observational and hence
preclude causal inferences from being made.

7.4.6 Conclusion

The findings reported here highlight the important role of both genetic Hb
disorders and other micronutrients, besides iron, on the Hb and iron status of these
preschoolers. The feasibility of improved low cost screening programmes for α3.7
thalassemia and sickle cell disease in NE Brazil should be considered. Our results
emphasize the need for caution in supplementing iron-replete children with iron in
this setting, in light of the possible danger of iron overload associated with these
genetic Hb disorders. Further, the existence of functional iron deficiency
associated with overweight in these preschoolers is of particular concern, as 13%
were overweight or obese (Chapter 3). Hence, efforts to prevent overweight in
preschoolers in NE Brazil should be strengthened. Many of the factors identified
here, such as ethnicity, micronutrient status, and genetic Hb disorders also have
the potential to influence the physical growth of these children. Clearly, an
investigation focusing on the inter-relationships among these factors and the
physical growth of these preschoolers is needed.

152
8 Factors associated with growth in
disadvantaged preschoolers attending
daycare in Salvador, NE Brazil using
structural equation modelling

8.1 Introduction
The double-burden of malnutrition (i.e., the presence of under- and
overnutrition both among individuals and within households) has escalated among
the urban poor in Brazil in recent years because of the vulnerability of food-
insecure households in the midst of the rapid nutrition transition (Caballero 2005).
Co-existing overweight and mild stunting have been reported among
disadvantaged children in several Brazilian studies (Sawaya et al. 1998; Hoffman
et al. 2000a; Hoffman et al. 2000b). This pattern has been linked with an increased
risk of adolescent and adult obesity, and the onset of non-communicable diseases
such as diabetes, hyperlipidemia, and hypertension (Stein et al. 2005; da Luz
Santos et al. 2010; Clemente et al. 2012). In addition, mild childhood stunting has
also been associated with an increased risk of morbidity, delayed motor
development, impaired cognition and attention deficits (Mendez & Adair 1999;
Fernald et al. 2006), some of which may persist into adulthood (Galler et al.
2012).
National reports suggest that nearly 10% of households in Brazil have both
under- and overnourished members (Caballero 2005). However, a much higher
prevalence (i.e., 30%) has been reported among the low-income population in the
under-resourced NE region (Florêncio et al. 2001), where poor Brazilian women
are susceptible to the extremes of underweight or obesity (Monteiro et al. 2004;
Velásquez-Melendez et al. 2011). Such trends may lead to foetal under- or
overnutrition during gestation, both of which may be associated with overweight
and obesity in later in life, albeit by different mechanisms (Barker 2001; Oken &
Gillman 2003; Levin 2006).

153
Maternal fatness has been identified as an important childhood predictor of
adult obesity. Parsons et al. (1999), in an extensive systematic review, reported
that children of an obese parent were at an increased risk of fatness in adulthood, a
finding confirmed in large cohort studies of young children in both southern
Brazil (Araújo et al. 2010) and the United Kingdom (Reilly et al. 2005). However,
whether this heightened risk was through genetic mechanisms or environmental
and nutritional factors could not be established.
Many of these same factors are also associated with linear growth in
childhood, although the separate effects of genetics, environment, and nutrition
are difficult to identify. For example, the genetic role of maternal height is well-
established (Hautvast et al. 2000; Espo et al. 2002). However, inter-generational
poverty may have an additional influence on genetic height potential (Hernández-
Díaz et al. 1999; Ramakrishnan et al. 1999), as reported in a population-based
survey in Brazil (Sichieri et al. 2000). Likewise, higher birth order (i.e., third- or
subsequent born) may negatively affect childhood linear growth via the effects of
poor maternal nutritional status and / or through the lack of available resources in
crowded low-income households, as reported in NE Brazil (de Menezes et al.
2011) and elsewhere (Bronte-Tinkew & DeJong 2004; Gribble et al. 2009; Singh
et al. 2009). In contrast, differences in growth patterns by ethnicity and sex are
less influenced by socio-economic status (Ulijaszek 2001; Torun 2005).
Of the nutritional factors, consumption of diets with a low nutrient density,
frequently encountered in impoverished environments, may compromise growth
and body composition during childhood. Additional exacerbating factors arise
from poor hygiene and sanitation and contaminated drinking water, including the
presence of parasitic infections and subtropical enteropathy. Together, these can
adversely affect growth as a result of increased intestinal permeability and
reduced nutrient absorption (Ashbolt 2004; Humphrey 2009).
Clearly, growth of preschool children living in poor urban areas of NE
Brazil can be affected by multiple factors within a complex aetiological pathway.
Structural equation modelling allows for the simultaneous examination of
multiple factors linked in a coherent framework. Such an approach is likely to

154
improve our understanding of the direct and indirect determinants of linear and
ponderal growth of disadvantaged preschool children from Salvador, NE Brazil.
Therefore, this present study aims to: (1) assess the prevalence of under- and
overnutrition among these preschoolers and within mother-child diads; and (2)
apply structural equation modelling to examine the direct and indirect effects of
relationships between biological, social, nutritional, and environmental factors on
growth of these disadvantaged daycare preschoolers in Salvador.

8.2 Participants and Methods

8.2.1 Study sites and participants

This cross-sectional study was conducted in Salvador, the capital city of


Bahia, NE Brazil and included 376 preschoolers (196 boys and 180 girls) aged
three to six years, all from low-income families, who were attending philanthropic
daycares (n=7) in urban and peri-urban areas of this large city. The study sites and
participants are described in detail in section 3.2.1.

8.2.2 Assessment of socio-demographic and health status

Socio-demographic and health status information was collected from all of


the mothers or caregivers (376/376) through the use of a culturally appropriate
structured questionnaire (Appendix A). Based on data collected from the mother
or guardian, an overall SES score was developed for each participant and divided
into extremely low or low SES, as described in sections 3.2.2 and 3.2.5. The
ethnicity of the children was determined by skin colour, and hair and facial
characteristics, as performed in the 2010 national census (IBGE 2010a) (see
section 2.4.2). Information on health status variables, including birth order and
maternal reports of deworming treatment within the past six months, was also
collected, as described earlier.

155
Table 8.1 Socio-demographic characteristics and prevalence of anaemia,
low zinc and selenium status of the preschool children

n %
Children
Sex (male) 196/376 52.1

Ethnicity
Black 154/365 42.2
Brown 189/365 51.8
White 22/365 6.0

Birth order (3rd or subsequent born) 127/376 33.8

Socioeconomic status (extremely low) 182/376 48.4

Helminth infection 58/325 17.8

Deworming treatment within 6 months 192/376 51.1

Anaemia
< 5 years (Hb < 110 g/L) 11/319 3.4
≥ 5 years (Hb < 115 g/L) 3/40 7.5

Low Zn status (serum Zn < 9.9 µmol/L) 13/358 3.6

Low Se status (serum Se ≤ 0.82 µmol/L) 34/358 9.5

156
8.2.3 Assessment of child and maternal anthropometry

Almost all of the children (96%; 364/376) had their weight and height
measured, wearing light clothing and no shoes. Standardized techniques and
calibrated equipment were used for the measurements (see section 3.2.4). Z-scores
for height-for-age and BMI were calculated using the updated US CDC 2000
Growth Reference (Kuczmarski et al. 2000), based on the method described by
Vidmar et al (2004). Nearly 85% (312/376) of the mothers had their weight and
height measured. Measurements were taken to the nearest 0.1 kg and 0.1 cm,
respectively, and BMI was calculated using standard equations (Frisancho 1990),
as described previously (section 3.2.4).
Children were classified as stunted or mildly stunted and wasted and mildly
wasted based on HAZ- and WHZ-scores of < -2 or < -1 to ≥ -2SD, respectively.
Overweight or obesity in the children was based on BMIZ > 1 to ≤ 2SD or > 2SD,
respectively. Mothers’ heights < 150 cm were classified as short, in accordance
with other studies in NE Brazil (Amorim et al. 2011; de Menezes et al. 2011).
Maternal underweight was based on BMI < 18.5 and overweight or obese based
on BMI ≥ 25 to < 30 or BMI ≥ 30, respectively (WHO 2000).

8.2.4 Assessment of parasite status, Hb and micronutrient biomarkers

Of the 376 participants, 86% (n=325) provided a stool sample for analysis.
A faecal concentrate was prepared and examined for the presence of helminths by
microscopy, as described in section 5.2.2. Morning fasting blood samples were
drawn from 95.7% (360/376) of the children (see Figure 3.2), and Hb, serum zinc,
and selenium assayed, as described earlier in sections 6.2.4 and 7.2.5.

8.2.5 Statistical analysis

The structural equation model (SEM) was based on the biological


plausibility of the observed variables to affect specific growth outcomes, namely
BMIZ and HAZ. Variables investigated included sex, ethnicity, birth order, SES,

157
Table 8.2 Anthropometric variables for the preschoolers and their mothers,
and mother-child diads

Children Mean (SD)


1
HAZ
Males 0.37 (1.08)
Females 0.60 (1.03)
BMIZ1
Males -0.33 (1.17)
Females -0.11 (1.12)

n %
Stunted (HAZ < -2SD) 7/364 1.9
Mildly stunted (HAZ < -1 ≥ -2SD) 20/364 5.5
Wasted (WHZ < -2SD) 19/363 5.2
Mildly wasted (WHZ < -1 ≥ -2SD) 57/363 15.7
Overweight (BMIZ > 1 ≤ 2SD) 39/364 10.7
Obese (BMIZ > 2SD) 9/364 2.5
Stunted / mildly stunted & overweight / obese 5/27 20.0

Mothers
Short (Height < 150cm) 25/312 8.0
Underweight (BMI < 18.5) 13/312 4.2
Overweight (BMI ≥ 25 to < 30) 100/312 32.1
Obese (BMI ≥ 30) 67/312 21.5
Short & overweight / obese 13/25 52.0

Child – Mother Diads


Stunted / mildly stunted children with
10/22 45.5
short mothers
Stunted / mildly stunted children with
11/22 50.0
overweight / obese mothers
Wasted / mildly wasted children with
23/64 35.9
overweight / obese mothers
Overweight / obese children with
31/43 72.1
overweight / obese mothers

1
Adjusted by age and daycare cluster

158
helminth infection, deworming treatment, maternal weight and height, serum
selenium, zinc, and Hb. Age and chronic inflammation (as indicated by serum
AGP) were not included in the model because of the narrow age range of the
children and the sensitivity of the AGP assay precluded using the data as a
continuous variable. Birth weight and genetic Hb disorders were also not included
because limited data were available for these specific variables.
Comparison between variables for the complete data set included in the
SEM and the excluded incomplete set for the categorical variables (i.e., sex,
ethnicity, birth order, SES, helminth infection and deworming treatment) was
based on the Fisher’s exact test, and for the continuous variables (i.e., maternal
weight and height, serum selenium and zinc, Hb, BMIZ and HAZ) the Student's
two-sample t-test was used. The SEM was used to simultaneously estimate the
regression equations identified in this study that pertained to growth, using
complete data sets adjusted by daycare clustering. The direct and indirect effects
reported are standardized regression weights. Standardized ß coefficients with p
values < 0.05 were considered significant. The sandwich estimator was used to
obtain robust standard errors, to account for the sampling procedure. Statistical
analyses were carried out using STATA version 12 (Stata Corporation, College
Station, TX, USA).

8.3 Results

8.3.1 Socio-demographic characteristics of the children

Table 8.1 highlights the socio-demographic characteristics of the children


that specifically pertain to the variables included in the SEM. Most of the children
were ethnically either black (42.2%) or brown (51.8%), and nearly 50% were
from extremely low SES households. No statistically significant difference was
found between SES and child race. A third of the children were of higher (3rd or
subsequent) birth order, of whom nearly 60% (75/127) were from extremely poor
households (p=0.003). Although 51% of the children had received deworming
treatment within the past six months, nearly 20% had a helminth

159
Table 8.3 Comparison between variables for the complete data set included in the structural equation model (SEM)
(n=264) and the excluded incomplete set

Variables n SEM complete case n Incomplete data set p


% %

Sex (male) 264 52.7 112 50.9 0.822


Child Ethnicity (black) 264 46.6 101 30.7 0.016
Child Ethnicity (white) 264 6.1 101 5.9
Birth order (3rd or subsequent) 264 32.2 112 37.5 0.341
SES (extremely low) 264 49.6 112 45.5 0.499
160

Helminth infection 264 18.6 61 14.8 0.580


Deworming treatment within 6 months 264 53.0 112 46.4 0.260

Mean (SD) Mean (SD)


Maternal weight (kg) 264 66.6 (15.4) 60 65.7 (15.5) 0.688
Maternal height (cm) 264 158.9 (6.9) 48 160.0 (5.7) 0.294
Serum selenium (umol/L) 264 0.99 (0.14) 94 1.00 (0.15) 0.338
Serum zinc (umol/L) 264 12.9 (1.8) 94 12.6 (1.8) 0.312
Hb (g/L) 264 122.8 (7.5) 95 123.0 (7.7) 0.838
BMIZ 264 -0.20 (1.18) 100 -0.29 (1.06) 0.517
HAZ 264 0.51 (1.05) 100 0.40 (1.09) 0.386
infection. Few children were anaemic or had low zinc status (i.e., < 8%), although
nearly 10% had low selenium status.

8.3.2 Anthropometric assessment of the children and mothers

Very few children (i.e., < 2%) were stunted; 5.5% were mildly stunted
(Table 8.2). Likewise, only 5.2% children were wasted, although 15.7% were
mildly wasted. Overweight among the children was 10.7%; 2.5% were obese.
Nearly 20% (i.e., n=5) of the stunted children were overweight or obese and of
these, four were from extremely low SES households (results not shown).
Of the mothers or primary caregivers, 8% (25/312) were short and less than
5% (13/312) were underweight (Table 8.2). However, more than 30% (100/312)
were overweight and over 20% (67/312) were obese. Investigation of the
anthropometric relationships between mothers and their children showed that
more than 40% (10/22) of the stunted and mildly stunted children had short
mothers. Moreover, 36% (23/64) of wasted and mildly wasted children had
overweight or obese mothers. In contrast, over 70% (31/43) of the children with
BMIZ > 1SD had mothers with a BMI ≥ 25.

8.3.3 Comparison between SEM and overall study data

Of the 376 study participants, 70.2% had complete data (n=264) for
inclusion in the SEM to examine plausible factors that may impact on HAZ and
BMIZ in these preschoolers. No significant differences were found between the
variables for those participants for whom complete data were available and those
with incomplete data available (n= 112), except for ethnicity (Table 8.3). The
SEM sample had a higher proportion of black children.

8.3.4 Correlation matrix of variables included in the SEM

The correlation matrix for the variables used in the SEM is shown in Table
8.4. The variables sex, ethnicity, birth order, and maternal weight and height
represent unmodifiable biological factors that may affect growth. Helminth
infection was included in the model because of its association with

161
Table 8.4 Correlation matrix of biological, social and environmental factors affecting the children’s BMIZ and HAZ (n=264)

1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sex Black White Birth SES Hel- De- Mat Mat Se Zn Hb BMIZ HAZ
(male) order (ex low) minth worm weight height
1. Sex (male) 1.00
2. Ethnicity (black) 0.131 1.00
3. Ethnicity (white) -0.06 . 1.00
4. Birth order (3rd or later) 0.09 0.10 -0.12 1.00
5. SES (extremely low) 0.03 0.02 0.091 0.09 1.00
6. Helminth infection 0.201 -0.01 -0.171 0.15 0.172 1.00
162

7. Deworming treatment -0.06 0.07 0.003 0.11 -0.02 -0.121 1.00


8. Maternal weight -0.02 -0.03 -0.007 -0.002 -0.161 0.01 0.04 1.00
9. Maternal height 0.02 -0.02 0.051 -0.242 -0.04 0.003 -0.01 0.352 1.00
10. Serum selenium 0.151 -0.005 0.04 -0.111 -0.05 -0.11 0.02 -0.09 0.08 1.00
11. Serum zinc 0.04 0.05 0.07 -0.13 0.003 -0.14 0.08 -0.07 0.05 0.201 1.00
12. Hb -0.03 -0.03 0.07 -0.13 -0.14 -0.161 0.06 0.10 0.02 0.272 0.161 1.00
13. BMIZ -0.091 -0.03 -0.11 -0.06 -0.242 -0.01 0.111 0.282 -0.03 -0.02 -0.09 0.171 1.00
14. HAZ -0.141 -0.08 -0.10 -0.262 -0.101 -0.102 0.03 0.282 0.392 0.03 -0.04 0.091 0.211 1.00

1
p < 0.05
2
p < 0.001
deworming treatment (Chapter 5) and serum selenium (Chapter 6), as noted
earlier. Serum selenium was a strong predictor of Hb concentration (Chapter 7),
with Hb showing a strong positive association with BMIZ in univariate regression
(results not shown). Additionally, serum selenium modestly predicted serum zinc
(Chapter 6), a well-established growth-limiting micronutrient. Lastly, socio-
economic status was included because of its association with the BMIZ of the
children and the BMI of the mothers (Chapter 3), as well as with risk of helminth
infection, as reported in Chapter 5.

8.3.5 Direct and indirect effects on HAZ and BMIZ

A structural equation model shown in Figure 8.1 presents a theoretical


causal pathway between the observed variables as they impact on HAZ and BMIZ
scores of the children. Standardized regression weights are shown beside the
arrows. The root mean square residual (RMSR) was 0.049, indicating that the
model was a good fit. The coefficient of determination, akin to R2 in linear
regression, was 0.362.
Table 8.5 shows the direct, indirect, and the combined total effects of each
observed variable on HAZ and BMIZ. Maternal height had the only significant
and positive direct effect on HAZ-scores with the largest magnitude of all the
observed variables (i.e., 0.37). Significant and negative direct associations on
HAZ, but all of smaller magnitude, were being male, white, the third or
subsequent child by birth order, and having a helminth infection. Sex and serum
selenium had significant and negative indirect effects on HAZ-scores, through the
action of sex on helminths and serum selenium. The figure shows the direct and
positive effect of sex (i.e., being male) on both helminths and serum selenium,
effects which in turn influenced serum zinc and stature. Extremely low SES and
deworming treatment had an indirect but non-significant effect on HAZ through
directly acting on helminths, which in turn had a negative influence on HAZ-
scores.

163
Figure 8.1 Structural equation model for growth with outcomes HAZ and BMIZ

Maternal weight 0.212

SES (extremely low) -0.181

0.131 BMIZ
Haemoglobin
0.161
-0.081
Sex (male)
0.272 -0.10
1
0.19 1 1 0.42
0.18 -0.12
164

Helminth -0.141 0.201 Serum zinc


Serum selenium
-0.08
1
infection -0.08

-0.161
HAZ
-0.101 Birth order
Ethnicity (white) -0.102
Deworming treatment
Maternal height 0.372

1
p < 0.05
2
p < 0.001
The significantly positive direct effects on BMIZ-scores were maternal
weight and Hb, whereas extremely low SES and being male were significant and
negative direct effects (Table 8.5). Only the presence of a helminth infection had a
significantly negative indirect effect on BMIZ-scores and thus ponderal growth
through serum selenium and zinc. In this model, HAZ and BMIZ were modestly
correlated with a correlation of 0.42 (p=0.058).

8.4 Discussion
This model illustrates the complex multiple pathways that determine HAZ
and BMIZ-scores in these disadvantaged preschoolers from Salvador, NE Brazil.
To our knowledge, this is the first study to use a SEM to examine the effects of
maternal influences, socio-economic and health factors, helminth infections,
haematological and micronutrient status on linear and ponderal growth among
Brazilian preschoolers.

8.4.1 Role of maternal and biological factors on HAZ scores

Not unexpectedly, maternal height was a strong positive and direct predictor
of linear growth in these children, as reported elsewhere (Simondon et al. 1998;
Hautvast et al. 2000; Espo et al. 2002). Clearly, the positive influence of parental
height is well-accepted. However, in this low-income environment, inter-
generational poverty may have contributed to reductions in maternal height as
well as linear growth during childhood through intra-uterine growth retardation,
independent of their genetic growth potential (Hernández-Díaz et al. 1999;
Ramakrishnan et al. 1999; Coly et al. 2006). Indeed, more than 45% of the
preschoolers classified here as stunted and mildly stunted had short mothers,
perhaps arising from a generational cycle of impaired growth associated with low
SES, as reported elsewhere in NE Brazil (Ferreira et al. 2009; Amorim et al. 2011;
de Menezes et al. 2011).
We also observed ethnic and birth order differences in the children’s growth
patterns. The white children were significantly shorter than their black and brown

165
Table 8.5 Direct and indirect effects of SEM variables on HAZ
and BMIZ (n=264)

HAZ Direct Indirect Total p

Sex (male) -0.121 -0.021 -0.13 < 0.001


Ethnicity (black) -0.04 -0.04 0.288
Ethnicity (white) -0.102 -0.10 < 0.001
Birth order -0.161 -0.16 0.002
SES (extremely low) -0.01 -0.01 0.120
1
Helminth infection -0.08 0.0021 -0.08 0.053
Deworming treatment 0.008 0.01 0.141
Maternal height 0.372 0.37 < 0.001
Serum selenium -0.021 -0.02 0.018
Serum zinc -0.08 -0.08 0.158

BMIZ Direct Indirect Total p

Sex (male) -0.081 0.003 -0.07 0.018


SES (extremely low) -0.181 -0.000 -0.18 0.001
Helminth infection -0.0021 -0.002 0.015
Deworming treatment 0.000 0.000 0.311
Maternal weight 0.212 0.21 < 0.001
Serum selenium 0.02 0.02 0.209
Serum zinc -0.10 -0.10 0.118
Hb 0.131 0.13 0.006

1
p < 0.05
2
p < 0.001

166
peers, a pattern reported by Ulijaszek (2001) and consistent with that of children
in the US NHANES III survey (Grummer-Strawn et al. 2001). Birth order also
impacted the children’s height; children of third or higher birth order were
significantly shorter than first- or second-born children, as reported elsewhere
(Gribble et al. 2009). Frequent reproductive cycling can result in depleted
maternal nutritional status during foetal development, which can in turn
negatively affect birth length and subsequent linear growth. In addition, in such
low-income settings, available social and economic resources in large families
will be limited, which together may induce poor post-natal nutrition resulting in
reduced linear growth (Bégin et al. 1999; Bronte-Tinkew & DeJong 2004).
Certainly, in an earlier study in Salvador, preschoolers living in crowded
households (i.e., more than five persons per bedroom) had nearly twice the risk of
mild-to-moderate height-for-age deficits compared to those from smaller
households (Assis et al. 2007). Comparable findings have been reported in other
studies in NE Brazil (de Menezes et al. 2011) and elsewhere in Latin America
(Pelto et al. 1991; Sereebutra et al. 2006).
The model shows that boys were significantly shorter and lighter than the
girls (Chapter 3), a pattern that is probably associated, at least in part, with the
higher energy and nutrient requirements for growth in young boys during early
childhood, as documented elsewhere (Torun 2005). Sub-optimal zinc status,
possibly through its association with selenium, may have also played a role. Boys
are said to have higher requirements for zinc than girls (Cavan et al. 1993; Gibson
et al. 2007; Gibson et al. 2011), and zinc has a critical role in linear growth
(Brown et al. 2009) through several mechanisms. They may include the enhancing
effects of zinc on vitamin D and IGF-1, and thus subsequent bone formation and
mineralization (Salgueiro et al. 2002; Yamaguchi & Fukagawa, 2005),
interactions of zinc with specific hormones involved in bone growth (e.g., thyroid
hormones, insulin, somatomedin-c) (Salgueiro et al. 2002), as well as the critical
role of zinc in the optimum functioning of the immune system (Black 2001).

167
8.4.2 Role of environmental factors and micronutrient status on HAZ scores

Sanitation services and a proper water supply are often unavailable in low-
income areas in NE Brazil (de Lima et al. 2010), particularly in the poor areas of
Salvador (Barreto et al. 2010). Crowded households lacking clean water and
sewage facilities foster infection with helminths, such as Trichuris trichiuris and
Ascaris lumbricoides (Korkes et al. 2009). Such infections cause diarrhoea and
reduced absorption of micronutrients (e.g., zinc) (Kongsbak et al. 2006b), by
increasing the permeability of the small intestine while simultaneously decreasing
its absorptive surface (Crompton & Nesheim 2002). Thus, it is not surprising that
helminth infections had a direct and significant negative effect on linear growth,
as reported by others in Brazilian preschoolers in the Northeast (Moore et al.
2001) and elsewhere (Saldiva et al. 1999; Jardim-Botelho et al. 2008a).
Helminth infections also had an indirect negative impact on the linear
growth of these children, which was also significant. Such an indirect effect may
be attributed to the association between helminth infections and low serum
selenium concentrations noted earlier in these preschoolers (Chapter 6), and their
potential to impact on zinc status. Selenoproteins, specifically glutathione
peroxidases, regulate the release of redox-inert zinc from metallothionein to
certain zinc enzymes, specifically copper-zinc superoxide dismutase (Maret 2000;
Lyons et al. 2004). Hence, an association between selenium and zinc may be
responsible in part for the observed indirect relationship between helminth
infections and linear growth.
It is of interest that serum selenium concentrations also had a significant
indirect effect on HAZ-scores. In addition to the selenium-zinc association,
selenium is also known to protect the host during inflammatory reactions
(Wintergerst et al. 2007). A low selenium status may contribute to a reduction in
immune competence and thus, heightened susceptibility to helminth infection,
which may in turn, indirectly impair linear growth. More studies are needed to
investigate the relationships between selenium status and helminth infections and
their potential effects on linear growth.

168
Our failure to find a positive association between deworming treatment and
linear growth may have arisen because the prevalence of helminth infections was
relatively low (i.e., < 20%) in the preschoolers and 50% had received deworming
treatment in the past six months (Chapter 5). Indeed, recent evidence suggests that
a single dose of antihelminthic drugs may lead to a significant increase in weight,
but not height, in children who live in endemic areas with a high worm prevalence
(i.e., > 50%), but not when a low worm prevalence or intensity exists (Hall et al.
2008; Taylor-Robinson et al. 2011). Most studies have been conducted in school-
aged children (Taylor-Robinson et al. 2011), in line with the WHO deworming
strategy (WHO, 2002), but the effect of deworming may be greater in younger
children (Bhan et al. 2001; Stoltzfus et al. 2004). Therefore, further research is
needed on the impact of deworming treatment on growth for children under-five
years of age.

8.4.3 Role of maternal factors and SES on BMIZ scores

Maternal BMI strongly and positively affected the BMIZ-scores of the


preschoolers. Such a direct association is likely to be linked to maternal
nutritional status, both pre- and post-natally. Both maternal under- and
overnutrition during gestation and complementary feeding may lead to long-term
obesity and metabolic disorders in children via different mechanisms (Levin
2006). For example, in response to poor growth in foetal- and postnatal-life
induced by maternal undernutrition, metabolic efficiency may be heightened
during childhood. Hence, when sufficient energy becomes available, adiposity is
promoted and insulin-resistance develops (Hales & Barker 2001; McMillen et al.
2009). In contrast, maternal obesity during gestation and lactation may result in
increased numbers of adipocytes and pancreatic ß-cell hyperplasia during
childhood, trends that also predispose children to obesity and the long-term
deleterious effects of insulin-resistance and other metabolic disorders (Levin
2006; O’Rahilly & Farooqi 2006). The finding that more than 70% of the
overweight and obese children in our study had overweight / obese mothers is
consistent with this possible mechanism.

169
Nevertheless, some investigators have criticized the foetal origins
hypothesis because it does not adequately account for the confounding effect of
factors, such as SES, on the peri- and post-natal growth and chronic disease
relationships (Adair & Kuzawa 2001; Joseph 2002). Certainly, it is notable that
the model showed that SES had a direct negative and significant impact on the
BMIZ-scores of the preschoolers which appeared to be of similar magnitude to
that of maternal weight on BMIZ-scores (Figure 8.1). This finding is consistent
with the reports that food-insecure households in Brazil are the most vulnerable to
the effects of the rapid nutrition transition and urbanization through the increased
availability of cheap, energy-dense, nutrient-poor foods (Monteiro et al. 2004;
Velásquez-Melendez et al. 2011). Such a dietary pattern has the potential to
contribute to excess maternal weight gain while undermining optimal growth
during childhood (Drewnowski & Specter 2004). This may account for our
finding that nearly 40% of the wasted or mildly wasted children (WHZ < -1SD)
had overweight or obese mothers (BMI ≥ 25). Thus, maternal influence on the
ponderal growth of these preschoolers may be modified by this low-income
setting, as has been reported among school children in NE Brazil (Amorim et al.
2011).
As anticipated, height and weight were modestly correlated in this model.
However, overweight was present in nearly 20% of the stunted or mildly stunted
children, a pattern associated with a heightened risk of adolescent and adult
obesity (Sawaya & Roberts 2003; Stein et al. 2005). Of these stunted-overweight
children, most (80%) were from extremely poor households, consistent with other
studies in Brazil (Sawaya et al. 1998; Hoffman et al. 2000a; Hoffman et al.
2000b). Several plausible mechanisms may explain this stunting-overweight
paradox. Low zinc status arising from diets with a low content and / or
bioavailability of zinc may impair linear growth and increase fatness (Prentice
1993; Perrone et al. 1998; Gibson et al. 2000b), as zinc plays an important role in
accrual of fat-free mass (Arsenault et al. 2008). However, few children in our
study were zinc deficient (Chapter 6), probably as a result of the zinc-rich low-
phytate diet of the daycares (Chapter 4). Whether these children were zinc-

170
deficient during infancy and early childhood, when their zinc requirements for
growth were higher (Krebs & Hambidge 2007), and prior to their regular
consumption of zinc-rich daycare meals, is unknown. Other postulated
mechanisms include an increased deposition of fat due to decreased synthesis of
IGF-1 and reduced lipolysis (Hoffman et al. 2000a; Veiga et al. 2010). This
condition arises from limited energy intake during growth, and possibly an
increased ratio of cortisol to insulin leading to cortisol-induced truncal obesity, a
characteristic of stunted-overweight children (Sawaya et al. 1998). Finally,
children from food-insecure households who are under-nourished during
intrauterine- and early life may opportunistically overeat because of a reduced
ability to self-regulate their energy intake as a result of underlying metabolic
disorders or behavioural factors (Hoffman et al. 2000b).

8.4.4 Role of Hb on BMIZ scores

We are not the first to report a positive relationship between Hb and weight
status among Brazilian children (Oliveira et al. 2010; Amorim et al. 2011),
although the physiological mechanisms are not entirely clear. Several studies have
reported improvements in appetite and weight gain in anaemic children receiving
iron-supplements because iron deficiency can depress appetite (Auckett et al.
1986; Lawless et al. 1994; Chwang et al. 1998). It is notable few children (i.e., <
10%) in our study were anaemic, probably due to the provision of iron-rich
daycare meals (Chapter 4). Certainly other studies have reported adequate intakes
of iron among those attending daycares in Brazil (Oliveira et al. 2010) and
elsewhere (Brotanek et al. 2007). Further, some positive correlations have been
reported between Hb and growth (HAZ and WAZ) among Brazilian daycare
children (Oliveira et al. 2010).

8.4.5 Strengths and limitations

The SEM has both strengths and limitations. The model shown here had a
good fit, as the RMSR was less than 0.08 (StataCorp 2011). Further, the
coefficient of determination explained a moderate amount of the total variance.

171
The comprehensive nature of our study provided a range of factors that are likely
to impact on both linear growth and body mass index of these preschoolers,
although limited by the number of complete data sets available. Nevertheless, the
differences between the complete model data set and those excluded were not
significant. Finally, the biological direction implied by the direct and indirect
effects found in this cross-sectional study must be interpreted cautiously, as the
model conveys causal assumptions and not necessarily validated conclusions.

8.4.6 Conclusion

In conclusion, our findings highlight the multi-factorial associations of the


maternal, biological, socioeconomic, nutritional, and environmental influences on
both linear growth and body mass indices of these preschoolers. Certainly, this
model stresses the need for effective deworming strategies for these children.
Treatment of helminth infections is a modifiable risk factor which has the
potential to enhance growth outcomes. In light of our findings and those of others
(Stoltzfus et al. 2004), we recommend that the deworming strategy of WHO
should also include younger children living in at-risk environments, as well as
school-aged children currently targeted.
The exploratory analyses showed the susceptibility of these preschoolers to
both under- and overnutrition. The long term consequences of mild stunting and /
or overweight and obesity in these low-income preschoolers are of great concern,
not only because of the adverse effects on the health of the present generation, but
also on future generations.
In our study, daycares played a very important role in promoting healthy
growth, by providing access to clean drinking water, hygienic facilities, and a
high quality diet. We recommend that all the daycares continue to supply
micronutrient-rich diets with adequate energy density on a regular basis to the
preschoolers (Chapter 4) and to emphasize physical activities for the children, as
well as provide nutrition education classes for the mothers. With the addition of
such strategies, the deleterious effects of poverty on the health of both the
preschoolers and their mothers could be further reduced. Finally, expansion of

172
these daycares to cover the needs of all at-risk preschoolers living in the urban and
peri-urban slum areas of Salvador should be encouraged.

173
174
9 Summary, Conclusions and
Recommendations

9.1 Summary and Conclusions


Preschool children living in urban slums in Salvador, NE Brazil may be
susceptible to malnutrition, as Salvador is the third most populated city in Brazil
and is located in the country’s poorest region. Disadvantaged children living in
food-insecure households are likely to be at risk of under- or overnutrition
because of inadequate dietary intakes, as the diets of the poor often have a low
micronutrient density coupled with poor bioavailability of key micronutrients
(e.g., iron, zinc, and calcium). Further, risk of parasitic infections, such as soil-
transmitted helminths and Giardia intestinalis, are likely in poor areas of Salvador
because of lack of access to hygiene and sanitation services (Barreto et al. 2010).
These infections can have a negative impact on micronutrient status, and
consequently impair growth. Moreover, as most (i.e., 80%) of the population of
Salvador is of African or African Brazilian descent (IBGE 2010a), genetic Hb
disorders, such as sickle cell disease and thalassemias, also have the potential to
negatively impact the nutritional status of young children living in Salvador.
Brazil has implemented several pro-poor policies over the past three
decades in an attempt to improve the nutritional status of young children. The
high prevalence of anaemia (i.e., ~50%) reported among Brazilian children under-
five years of age has been assumed to be caused by iron deficiency (Jordão et al.
2009). As a result, universal fortification of wheat and corn flour with iron and
folic acid, as well as supplementation with iron in children less than two years of
age, has been introduced. Further, efforts have been made to expand the coverage
of vitamin A supplementation programmes for preschool children living in the NE
region. This strategy has been associated with a marked reduction in the
prevalence of vitamin A deficiency, although marginal vitamin A deficiency is

175
still highly prevalent (i.e., > 20%). However, little else is known about the
micronutrient status of preschoolers living in this region.
The prevalence of undernutrition in preschool children in NE Brazil has
also been dramatically reduced, and again attributed to some of the pro-poor
policies, namely promotion of universal primary education, improvements in
water and sewage services, and increased purchasing power for the poor (de Lima
et al. 2010). Indeed, NE Brazil is likely to meet the first United Nations
Millennium Development Goal, which aims to reduce the number of underweight
children by half between 1990 and 2015. Nevertheless, the rising prevalence of
overweight in children in this region is of grave concern, particularly as energy-
dense micronutrient-poor foods are increasingly consumed by low-income
participants who are recipients of Brazil’s CCT programme, Bolsa Familia, and
who live in the poor NE region (de Bem Lignani et al. 2011).
Early childhood education legislation was passed nearly twenty years ago
to ensure all young children from birth to six years of age in Brazil have access to
preschool daycares because these programmes have been associated with
improved nutritional outcomes for children living in poverty (Engle et al. 2007).
However, the implementation of programmes remains incomplete and few poor
children have the opportunity to attend daycare (Marins & Rezende 2011).
Despite some reduction in the reportedly high prevalence of anaemia in children
under-five years of age attending daycares in southeastern Brazil (da Silva Pereira
et al. 2010), there have been few comprehensive studies of the nutritional status of
disadvantaged children attending daycares, especially in NE Brazil.
Therefore, the overall goal of this research was to examine the nutritional
status of a group of preschoolers aged three to six years from low-income families
who attended philanthropic urban and peri-urban daycares in Salvador, NE Brazil.
The comprehensive assessment included: (i) an extensive socio-demographic and
health status questionnaire, including early IYCF practices; (ii) calculation of the
adequacy of the energy and nutrient supply based on the daycare menus; (iii)
assessment of the growth and body composition of the preschoolers; (iv)
determination of the prevalence and risk factors associated with parasitic

176
infections; (v) analysis of Hb, zinc, vitamin A, selenium, folate, vitamin B-12, and
iron biomarkers (i.e., serum ferritin and soluble TfR), and markers of acute and
chronic inflammation; and (vi) detection of the presence of genetic Hb disorders,
specifically sickle cell disease and α3.7 thalassemia.
The results of the cross-sectional survey suggested that the children living
in extreme poverty were more susceptible to undernutrition, whereas those living
in less extreme poverty, although still classified as low-income, were more likely
to suffer from overnutrition (Chapter 3), despite receiving most of their daily food
at daycare. High quality diets were provided for these disadvantaged children, and
the daycares are commended for their efforts to meet the nationally recommended
food-based dietary guidelines for children over two years of age (Chapter 4).
Nevertheless, the daycare menus failed to meet the desirable energy requirements
for children of this age, particularly in relation to an adequate supply of fat;
deficits in intakes of calcium, vitamin A, and thiamine were also noted.
Of the 86% of the children who provided stool samples, nearly one-third
were infected with at least one parasite infection, predominantly helminths (i.e.,
Trichuris trichiura and Ascaris lumbricoides) and the pathogenic protozoa,
Giardia intestinalis (Chapter 5). We found boys and children from extremely low
SES had more than twice the risk of helminth infections, and vitamin A
supplementation was inversely associated with the risk of G. intestinalis infection.
These preschoolers had a low prevalence (i.e., < 10%) of vitamin A, zinc,
selenium, folate and vitamin B-12 deficiencies (Chapter 6), which was probably a
reflection of their micronutrient-rich daycare meals together with the practice of
mandatory iron and folic acid fortification of wheat and corn flour in Brazil.
However, marginal vitamin A status persisted in more than 40% of the children.
Parasite infections, namely helminths and Giardia intestinalis, were associated
with lower serum selenium and vitamin B-12 concentrations, whereas a positive
association was found between Giardia intestinalis and folate status (Chapter 6).
The findings reported here also highlight the role of genetic Hb disorders
and other micronutrients, besides iron, on the Hb and iron status of these
preschoolers (Chapter 7). The danger of iron overload associated with genetic Hb

177
disorders is well-known, and our results emphasize the need for caution in
supplementing iron-replete children with iron when such a large proportion of the
population is of African ancestry. Further, the positive association between
functional iron deficiency and overweight in these preschoolers is of particular
concern, especially in light of recent evidence that overweight children may have
a negative response to iron fortification (Zimmermann et al. 2008b).
Finally, the growth of these preschoolers, specifically their linear and ponderal
growth, was influenced by multiple factors, including biological, nutritional,
socioeconomic, and environmental factors, which were associated with both
under- and overnutrition (Chapter 8). Our structural equation model also
highlighted the need for effective deworming strategies for these children,
particularly as treatment of helminth infections is a modifiable risk factor which
could possibly improve growth outcomes.
Our sampling was restricted to the seven philanthropic daycares and we were
unable to compare these preschoolers with those who do not attend preschool
daycare. However, recent evidence suggests that it is likely the nutritional status
of these low-income children who attended daycare full-time was higher than
those not attending daycare or only attending daycare part-time (Marins &
Rezende 2011). Hence, our findings suggest that the good quality of the diet
provided by micronutrient-rich daycare meals, enhanced by iron and folic acid
fortification, together with access to clean drinking water and hygienic sanitation
facilities in this group of disadvantaged preschoolers was reflected in overall
adequate nutritional status.

9.2 Recommendations
 We recommend some modifications for the daycare menus to ensure that
each child is supplied with the desirable range of energy derived from fat,
together with the requirements for calcium, thiamine, and vitamin A. The
daycares should consider increasing the supply of solid food in the menus

178
instead of beverages, a practice that will increase both the energy density
and the satiety of the meals. Further, to overcome the observed deficits in
the supply of calcium, vitamin A, and thiamine, we recommend
substituting a daily fruit juice snack with cow’s milk, and providing larger
serving sizes of the dishes at lunch time, particularly the main composite
dish.

 In light of the positive association found here between Giardia intestinalis


and folate status, further research is needed to confirm these findings and
to determine the underlying biological mechanism(s) by which folate
status could affect intestinal parasite invasion or growth. This investigation
is particularly important in children from low-income environments
receiving folic acid fortified foods.

 As almost a quarter of the children had α3.7 thalassemia and nearly 10%
had a haemoglobinopathy (i.e., Hb AS or Hb AC), the feasibility of
improved low cost screening programmes for these genetic Hb disorders in
NE Brazil should be considered.

 We suggest continued efforts to improve coverage of vitamin A


supplementation and extend parasite control measures to include
deworming in preschoolers. Although the WHO deworming strategy tends
to focus on school-aged children, our findings and those of others
(Stolzfus et al. 2004), suggest that regular deworming programmes in
younger children living in low-income environments may be advantageous
for improved micronutrient status and growth outcomes.

 Efforts to prevent overweight in preschoolers should be strengthened, with


continuing emphasis on physical activities for the children and nutrition
education classes for the mothers as part of the daycare programmes.

179
 Finally, we recommend that the provision of high-quality daycare
programmes, such as those found here, be expanded in other low-income
settings in Salvador, and to include younger children from six months of
age. Daycares such as these have the potential to reduce the risks of
malnutrition in urban slum environments and ensure primary school
attendance, which together may contribute positive benefits for achieving
the first two United Nations Millenium Development Goals of reduction of
poverty and hunger, and achieving universal primary education.

180
References

Abrams SA. (2002). Nutritional rickets: an old disease returns. Nutrition Reviews
60: 111-115.

Adair LS, Kuzawa CW. (2001). Early growth retardation and syndrome X:
conceptual and methodological issues surrounding the programming
hypothesis. In Martorell R, Haschke F (eds): Nutrition and Growth. Nestle
Nutrition Workshop Series Pediatric Program. Nestle Ltd., Vevey,
Nestec/Philadelphia, Lippincott Williams and Wilkins. Vol. 47: 333-346.

Adish AA, Esrey SA, Gyorkos TW, Johns T. (1999). Risk factors for iron
deficiency anaemia in preschool children in northern Ethiopia. Public
Health Nutrition 2: 243-252.

Adorno EV, Couto FD, Neto JP, Menezes JF, Rêgo M, dos Reis MG, Gonçalves
MS. (2005). Hemoglobinopathies in newborns from Salvador, Bahia,
Northeast Brazil. Cadernos de Saúde Pública 21: 292-298.

Aeberli I, Hurrell RF, Zimmermann MB. (2009). Overweight children have higher
circulating hepcidin concentrations and lower iron status but have dietary
iron intakes and bioavailability comparable with normal weight children.
International Journal of Obesity 33: 1111-1117.

Akman M, Cebeci D, Okur V, Angin H, Abah O, Akman AC. (2004). The effects
of iron deficiency on infants’ developmental test performance. Acta
Pædiatrica 93: 1391-1396.

Ali SA, Hill DR. (2003). Giardia intestinalis. Current Opinion in Infectious
Diseases 16: 453-460.

Allen LH. (1994). Nutritional influences on linear growth: a general review.


European Journal of Clinical Nutrition 48: S75-89.

Allen LH, Backstrand JR, Stanek EJ, Pelto G, Chávez A, Molina E, Castillo JB,
Mata A. (1992). The interactive effects of dietary quality on the growth and
attained size of young Mexican children. American Journal of Clinical
Nutrition 56: 353-364.

Allen LH, Rosado JL, Casterline JE, Martinez H, Lopez P, Muñoz E, Black AK.
(1995). Vitamin B-12 deficiency and malabsorption are highly prevalent in
rural Mexican communities. American Journal Clinical Nutrition 62: 1013-
1019.

181
Allen LH, Rosado JL, Casterline JE, Lopez P, Muñoz E, Garcia OP, Martinez H.
(2000). Lack of hemoglobin response to iron supplementation in anemic
Mexican preschoolers with multiple micronutrient deficiencies. American
Journal of Clinical Nutrition 71: 1485-1494.

Alloway BJ. (2004). Zinc in Soils and Crop Nutrition. International Zinc
Association: Brussels, Belgium.

Alsaleh AA. (2011). Haemoglobinopathies and thalassemia among Brazilian


children: cross sectional study. Unpublished Masters thesis. Department of
Pathology, University of Otago, Dunedin New Zealand.

Amato A, Santoro N, Calabro P, Grandone A, Swinkels DW, Perrone L, del


Giudice EM. (2010). Effect of body mass index reduction on serum
hepcidin levels and iron status in obese children. International Journal of
Obesity 34: 1772-1774.

Amorim RM, Lima MC, de Lira PIC, Emond AM. (2011). Does low birthweight
influence the nutritional status of children at school age? A cohort study in
northeast Brazil. Maternal and Child Nutrition 7: 295-306.

Anderson VP, Jack S, Monchy D, Heam N, Hok P, Bailey KB, Gibson RS.
(2008). Co-existing micronutrient deficiencies among stunted Cambodian
infants and toddlers. Asia Pacific Journal of Clinical Nutrition 17: 72-79.

Antunes MM, Sichieri R, Salles-Costa R. (2010). Food intake among children


under three years of age in an area with high food insecurity. Cadernos de
Saúde Pública 26: 1642-1650.

Araújo CL, Dumith SC, Menezes AM, Hallal PC, Vieira MA, Madruga SW,
Victora CG. (2010). Nutritional status of adolescents: the 11-year follow-up
of the 1993 Pelotas (Brazil) birth cohort study. Cadernos de Saúde Pública
26: 1895-1903.

Arsenault JE, López de Romaña D, Penny ME, Van Loan MD, Brown KH.
(2008). Additional zinc delivered in a liquid supplement, but not in a
fortified porridge, increased fat-free mass accrual among young Peruvian
children with mild-to-moderate stunting. Journal of Nutrition 138: 108-114.

Arsenault JE, Mora-Plazas M, Forero Y, López-Arana S, Marín C, Baylin A,


Villamor E. (2009). Provision of a school snack is associated with vitamin
B12 status, linear growth, and morbidity in children from Bogotá,
Columbia. Journal of Nutrition 139: 1744-1750.

182
Arsenault JE, Wuehler SE, López de Romaña D, Penny ME, Sempértegui F,
Brown KH. (2011). The time of day and the interval since previous meal are
associated with plasma zinc concentrations and affect estimated risk of zinc
deficiency in young children in Peru and Ecuador. European Journal of
Clinical Nutrition 65: 184-190.

Arthur JR, McKenzie RC, Beckett GJ. (2003). Selenium in the immune system.
Journal of Nutrition 133: 1457-1459.

Ashbolt NJ. (2004). Microbial contamination of drinking water and disease


outcomes in developing regions. Toxicology 198: 229-238.

Asobayire FS, Adou P, Davidsson L, Cook JD, Hurrell RF. (2002). Prevalence of
iron deficiency with and without concurrent anemia in population groups
with high prevalences of malaria and other infections: a study in Côte
d’Ivoire. American Journal of Clinical Nutrition 74: 776-782.

Assis AM, Prado MS, Barreto ML, Reis MG, Conceição Pinheiro SM, Parraga
IM, Blanton RE. (2004a). Childhood stunting in northeast Brazil: the role of
Schistosoma mansoni infection and inadequate dietary intake. European
Journal of Clinical Nutrition 58: 1022-1029.

Assis AM, Barreto ML, Gomes GS, Prado M, Santo NS, Sanot LMP, Sampaio
LR, Ribeiro R, de Oliveria LPM, de Oliveira VA. (2004b). Childhood
anemia prevalence and associated factors in Salvador, Bahia, Brazil.
Cadernos de Saúde Pública 20: 1633-1641.

Assis AM, Barreto ML, de Oliveira LP, de Oliveira VA, Prado M, Gomes GS,
Pinheiro SM, dos Santos NS, da Silva R, Sampaio LR, Santos LM. (2007).
Determinants of mild-to-moderate malnutrition in preschoolers in an urban
area of Northeastern Brazil: a hierarchical approach. Public Health Nutrition
11: 387-394.

Assunção MC, Santos IS, Barros AJ, Gigante DP, Victora CG. (2007). Effect of
iron fortification of flour on anemia in preschool children in Pelotas, Brazil.
Revista de Saúde Pública 41: 539-548.

Auckett MA, Parks YA, Scott PH, Wharton BA. (1986). Treatment with iron
increases weight gain and psychomotor development. Archives of Disease
in Childhood 61: 849-857.

Azevêdo ES, Alves AFP, da Silva MCB, Souza MG, Lima AMV, Azevêdo WC.
(1980). Distribution of abnormal hemoglobins and glucose-6-phosphate
dehydrogenase variants in 1200 school children of Bahia, Brazil. American
Journal of Physical Anthropology 53: 509-512.

183
Bain BJ. (2006). Haemoglobinopathy Diagnosis, 2nd edition. Blackwell
Publishing: Oxford, UK.

Baird J, Fisher D, Lucas P, Kleijnen J, Roberts H, Law C. (2005). Being big or


growing fast: systematic review of size and growth in infancy and later
obesity. British Medical Journal 331: 929-934.

Balaban G, Motta MEFA, Silva GAP. (2010). Early weaning and other potential
risk factors for overweight among preschool children. Clinics 65: 181-187.

Ballew C, Bowman BA, Sowell AL, Gillespie C. (2001). Serum retinol


distributions in residents of the United States: third National Health and
Nutrition Examination Survey, 1988-1994. American Journal of Clinical
Nutrition 73: 586-593.

Bandeira F, Griz L, Dreyer P, Eufrazino C, Bandeira C, Freese E. (2006). Vitamin


D deficiency: a global perspective. Brazilian Archives of Endocrinology and
Metabolism 50: 640-646.

Barbarich BN, Willow ND, Wang L, Clandinin MT. (2006). Polyunsaturated fatty
acids and anthropometric indices of children in rural China. European
Journal of Clinical Nutrition 60: 1100-1107.

Barden EM, Kawchak DA, Ohene-Frempong K, Stallings VA, Zemel BS. (2002).
Body composition in children with sickle cell disease. American Journal of
Clinical Nutrition 76: 218-225.

Barker DJP. (2001). The malnourished baby and infant. British Medical Bulletin
60: 69-88.

Barnett WS. (1998). Long-term cognitive and academic effects of early childhood
education on children in poverty. Preventive Medicine 27: 204-207.

Barr SI, Murphy SP, Poos ML. (2002). Interpreting and using the dietary
references intakes in dietary assessment of individuals and groups. Journal
of the American Dietetic Association 102: 780-788.

Barreto ML, Genser B, Strina A, Teixeira MG, Assis AM, Rego RF, Teles CA,
Prado MS, Matos SMA, Santos DM, dos Santos LA, Cairncross S. (2007).
Effect of city-wide sanitation programme on reduction in rate of childhood
diarrhea in northeast Brazil: assessment of two cohort studies. Lancet 370:
1622-1628.

Barreto ML, Genser B, Strina A, Teixeira MG, Assis AM, Rego RF, Teles CA,
Prado MS, Matos SMA, Alcântara-Neves NM, Cairncross S. (2010). Impact
of a citywide sanitation program in northeast Brazil on intestinal parasites
infection in young children. Environmental Health Perspectives 118: 1637-
1642.

184
Barreto ML, Teixeira MG, Bastos FI, Ximenes RAA, Barata RB, Rodrigues LC.
(2011). Success and failures in the control of infectious diseases in Brazil:
social and environmental context, policies, interventions, and research
needs. Lancet 377: 1877-1889.

Barros FC, Matijasevich A, Requejo JH, Giugliani E, Maranhão AG, Monteiro


CA, Barros AJ, Bustreo F, Merialdi M, Victora CG. (2010). Recent trends
in maternal, newborn and child health in Brazil: progress toward millennium
development goals 4 and 5. American Journal of Public Health 100: 1877-
1889.

Baum MK, Shor-Posner G. (1998). Micronutrient status in relationship to


mortality in HIV-1 disease. Nutrition Reviews 56: S135-139.

Beard JL. (2001). Iron biology in immune function, muscle metabolism and
neuronal functioning. Journal of Nutrition 131: 568-580.

Beard JL. (2003). Iron deficiency alters brain development and functioning.
Journal of Nutrition 133: S1468-1472.

Beard JL, Murray-Kolb LE, Haas JD, Lawrence F. (2007). Iron absorption
prediction equations lack agreement and underestimate iron absorption.
Journal of Nutrition 137: 1741-1746.

Beck MA. (1999). Selenium and host defence towards viruses. Proceedings of the
Nutrition Society 58: 707-711.

Bégin F, Frongillo EA, Delisle H. (1999). Caregiver behaviors and resources


influence child height-for-age in rural Chad. Journal of Nutrition 129: 680-
686.

Beinner MA, Menezes MA, Silva JB, Amorim FR, Jansen AK, Lamounier JA.
(2010). Plasma zinc and hair zinc levels, anthropometric status and food
intake of children in a rural area of Brazil. Revista de Nutrição 23: 75-83.

Belkacemi L, Nelson DM, Desai M, Ross MG. (2010). Maternal undernutrition


influences placental-fetal development. Biology of Reproduction 83: 325-
331.

Bethony J, Brooker S, Alonico M, Geiger SM, Loukas A, Diemert D, Hotez PJ.


(2006). Soil-transmitted helminth infections: ascariasis, trichuriasis and
hookworm. Lancet 367: 1521-1532.

Beutler E. (2008). Glucose-6-phosphate dehydrogenase deficiency: a historical


perspective. Blood 111: 16-24.

185
Bhan MK, Bahl R, Bhandari N. (2001). Infection: how important are its effects on
child nutrition and growth? In Martorell R, Haschke F (eds): Nutrition and
Growth. Nestle Nutrition Workshop Series Pediatric Program. Nestle Ltd.,
Vevey, Nestec/Philadelphia, Lippincott Williams and Wilkins. Vol. 47: 197-
221.

Bhandari N, Bahl R, Hambidge KM, Bhan MK. (1996). Increased diarrhoeal and
respiratory morbidity in association with zinc deficiency – a preliminary
report. Acta Pædiatrica 85: 148-150.

Bhargava A, Bouis HE, Scrimshaw NS. (2001). Dietary intakes and


socioeconomic factors are associated with the haemoglobin concentration of
Bangladeshi women. Journal of Nutrition 131: 758-764.

Bhaskaram P. (2001). Immunobiology of mild micronutrient deficiencies. British


Journal of Nutrition 85: S75-80.

Bhutta ZA. (2006). Effect of infections and environmental factors on growth and
nutritional status in developing countries. Journal of Pediatric
Gastroenterology and Nutrition 43: 13-21.

Biesalski HK. (1997). Bioavailability of vitamin A. European Journal of Clinical


Nutrition 51: S71-75.

Black RE. (2001). Zinc deficiency, immune function, and morbidity and mortality
from infectious disease among children in developing countries. Food and
Nutrition Bulletin 22: 155-162.

Black RE, Sazawal S. (2001). Zinc and childhood infectious disease morbidity
and mortality. British Journal of Nutrition 85: S125-129.

Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, Mathers C,
Rivera J. (2008). Maternal and child undernutrition: global and regional
exposures and health consequences. Lancet 371: 243-60.

Bloem MW. (1995). Interdependence of vitamin A and iron: an important


association for programmes of anaemia control. Proceedings of the
Nutrition Society 54: 501-508.

Bloem MW, Wedel M, Egger RJ, Speek AJ, Schrijver J, Saowakontha S, Schreurs
WHP. (1990). Mild vitamin A deficiency and risk of respiratory tract
diseases and diarrhea in preschool and school children in northeastern
Thailand. American Journal of Epidemiology 131: 332-339.

186
Bóia MN, Carvalho-Costa FA, Sodré FC, Eyer-Silva WA, Lama CC, Rosadinski
M, Pino VL, Cantalice Filho JP, Oliveira ALL, Carvalho LMA, Gross JB,
Souza ALS, Moraes TI, Burmudez-Aza EH, Martins EB, Coura JR. (2006).
Mass treatment for intestinal helminthiasis control in an Amazonian
endemic area in Brazil. Revista do Instituto de Medicina Tropical de São
Paulo 48: 189-195.

Borges E, Wenning MRSC, Kimura EM, Gervásio SA, Costa FF, Sonati MF.
(2001). High prevalence of α-thalassemia among individuals with
microcytosis and hypochromia without anemia. Brazilian Journal of
Medical and Biological Research 34: 759-762.

Bortonlini GA, Vitolo MR. (2010). Relationship between iron deficiency and
anemia in children younger than 4 years. Jornal de Pediatria 86: 488-492.

Bowden DK, Hill AVS, Higgs DR, Weatherall DJ, Cleg JB. (1985). Relative roles
of genetic factors, dietary deficiency, and infection in anaemia in Vanuatu,
South-west Pacific. Lancet 8463: 1025-1028.

Braga AB, Fialho AM, Rodrigues MN, Queiroz DM, Rocha AM, Braga LL.
(2007). Helicobacter pylori colonization among children up to 6 years:
results of a community-based study from Northeastern Brazil. Journal of
Tropical Pediatrics 53: 393-397.

Brasitus TA. (1983). Parasites and malabsorption. Clinical Gastroenterology 12:


495-510.

Brazil Ministry of Health. (1992). Perfil estatistico de crianças e mães no Brasil:


aspectos nutricionais, 1989. Instituto Brasileiro de Geografia e Estatistica.
Rio de Janeiro.

Brazil Ministry of Health. (1996a). Public law on the rights and basis of
education. Law 9.394, December 20, 1996. Brasilia: Ministry of Health.

Brazil Ministry of Health. (1996b). National demographic and health survey,


1996: final report. Rio de Janeiro.

Brazil Ministry of Health. (2002). Brazil legislation for the mandatory iron and
folic acid fortification. RDC 344, December 13. www.anvisa.gov.br

Brazil Ministry of Health. (2005). National programme of iron supplementation:


operations manual. Brasilia: Ministry of Health.

Brazil Ministry of Health. (2006). Brazil guidelines for the promotion of healthy
eating in schools of education: kindergarten, elementary and middle level of
public and private schools, nationwide. Ordinance No 1010, May 8.
www.anvisa.gov.br

187
Brazil Ministry of Health. (2007a). National woman and child demographic health
survey. Departamento de Ciência e Tecnologia. Brasília, Brazil.

Brazil Ministry of Health. (2007b). The ten steps to healthy eating for children
over two years of age.
www.portal.saude.giv.br/portal/arquivos/pdf/caderneta_2007_25.pdf

Bronte-Tinkew J, DeJong G. (2004). Children’s nutrition in Jamaica: do


household structure and household economic resources matter? Social
Science & Medicine 58: 499-514.

Brooker S, Jardim-Botelho A, Quinnell RJ, Geiger SM, Caldas IR, Fleming F,


Hotez PJ, Correa-Oliveira R, Rodrigues LC, Bethony JM. (2007). Age-
related changes in hookworm infection, anaemia and iron deficiency in an
area of high Necator americanus hookworm transmission in south-eastern
Brazil. Transactions of the Royal Society of Tropical Medicine and Hygiene
101: 146-154.

Brotanek JM, Gosz J, Weitzman M, Flores G. (2007). Iron deficiency in early


childhood in the United States: risk factors and racial/ethnic disparities.
Pediatrics 120: 568-575.

Brown KH. (1998). Effect of infections on plasma zinc concentrations and


implications for zinc status assessment in low-income countries. American
Journal of Clinical Nutrition 68: S425-429.

Brown KH, Lanata CF, Yuen ML, Peerson JM, Butron B, Lönnerdal B. (1993).
Potential magnitude of the misclassification of a population’s trace element
status due to infection: example from a survey of young Peruvian children.
American Journal of Clinical Nutrition 58: 549-554.

Brown KH, Rivera JA, Bhutta Z, Gibson RS, King JC, Lönnerdal B, Ruel MT,
Sandstrom B, Wasantwisut E (International Zinc Nutrition Consultative
Group). (2004). Assessment of the risk of zinc deficiency in populations and
options for its control. Food and Nutrition Bulletin 25: S95-203.

Brown KH, Peerson JM, Baker SK, Hess SY. (2009). Preventive zinc
supplementation among infants, preschoolers, and older prepubertal
children. Food and Nutrition Bulletin 30: S12-40.

Bunch S, Murphy SP. (1997). User’s Guide to the Operation of the WorldFood
Dietary Assessment System, version 2.0. Office of Technology Licensing,
University of California, Berkeley.

Burk RF. (2002). Selenium, an antioxidant nutrient. Nutrition in Clinical Care 5:


75-79.

188
Burlandy L. (2007). Conditional cash transfer programs and food and nutrition
security. Ciência Saúde Coletiva 12: 1441-1451.

Butte NF, Wong WW, Hopkinson JM, Heinz CJ, Mehta NR, Smith EOB (2000).
Energy requirements derived from total energy expenditure and energy
deposition during the first two years of life. American Journal of Clinical
Nutrition 72: 1558−1569.

Caballero B. (2005). A nutrition paradox – underweight and obesity in developing


countries. New England Journal of Medicine 352: 1514-1516.

Carmel R. (1997). Cobalamin, the stomach and ageing. American Journal of


Clinical Nutrition 66: 750-759.

Carter RC, Jacobson JL, Burden MJ, Armony-Sivan R, Dodge NC, Angelilli ML,
Lozoff B, Jacobson SW. (2010). Iron deficiency anemia and cognitive
function in infancy. Pediatrics 126: 3427-3434.

Carvalho AG, de Lira PIC, Barros M, Aléssio MLM, Lima M, Carbonneau MA,
Berger J, Lãger CL. (2010). Diagnosis of iron deficiency anemia in children
of northeast Brazil. Revista de Saúde Pública 44: 513-519.

Carvalho-Almeida TT. Pinto PLS, Quadros CMS, Torres DMAGV, Kanamura


HY, Casimiro AM. (2006). Detection of Cryptosporidium sp. in non
diarrheal faeces from children in a daycare centre in the city of Sao Paulo,
Brazil. Revista do Instituto de Medicina Tropical de São Paulo 48: 27-32.

Carvalho-Costa FA, Gonçalves AQ, Lassance SL, da Silva Neto LM, Salmazo
CAA, Bóia MN. (2007). Giardia lamblia and other intestinal parasitic
infections and their relationships with nutritional status in children in
Brazilian Amazon. Revista do Instituto de Medicina Tropical de São Paulo
49: 147-153.

Castenmiller JM, West CE. (1998). Bioavailability and bioconversion of


carotenoids. Annual Review of Nutrition 18: 19-38.

Caudill MA. (2010). Folate bioavailability: implications for establishing dietary


recommendations and optimizing status. American Journal of Clinical
Nutrition 91: S1455-1460.

Caughy MO, DiPietro JA, Strobino DM. (1994). Day-care participation as a


protective factor in the cognitive development of low-income children.
Child Development 65: 457-471.

Cavan KR, Gibson RS, Grazioso CF, Isalgue AM, Ruz M, Solomons NW. (1993).
Growth and body composition of periurban Guatemalan children in relation
to zinc status: a cross-sectional study. American Journal of Clinical
Nutrition 57: 334-343.

189
Cepeda-Lopez AC, Osendarp SJM, Melse-Boonstra A, Aeberli I, Gonzalez-
Salazar F, Feskens E, Villalpando S, Zimmermann MB. (2011). Sharply
higher rates of iron deficiency in obese Mexican women and children are
predicted by obesity-related inflammation rather than by differences in
dietary iron intake. American Journal of Clinical Nutrition 93: 975-983.

César JA, Mendoza-Sassi R, Horta BL, Ribeiro PR, D’Avila AC, Santos FN,
Martins PB, Brandolt RR. (2006). Basic indicators of child health in an
urban area in southern Brazil: estimating prevalence rates and evaluating
differentials. Jornal de Pediatria 82: 437-444.

Chen P, Soares AM, Lima AAM, Gamble MV, Schorling JB, Conway M, Barrett
LJ, Blander WS, Guerrant RL. (2003). Association of vitamin A and zinc
status with altered intestinal permeability: analyses of cohort data from
northeastern Brazil. Journal of Health, Population and Nutrition 21: 309-
315.

Choi JW, Pai SH, Im MW, Kim SK. (1999). Change in transferrin receptor
concentrations with age. Clinical Chemistry 45: 1562-1563.

Chouliaras GL, Premetis E, Tsiftis G, Drosatou P, Papassotiriou I, Stamoulakatou


A, Lycopoulou L. (2009). Serum transferrin receptors: distribution and
diagnosis in pre-school children. Blood Cells, Molecules and Diseases 43:
163-168.

Christian P, West KP. (1998). Interactions between zinc and vitamin A: an update.
American Journal of Clinical Nutrition 68: S435-441.

Chwang L, Soemantri AG, Pollitt E. (1988). Iron supplementation and physical


growth of rural Indonesian children. American Journal of Clinical Nutrition
47: 496-501.

Cipolotti R, Caskey MFB, Franco RP, Mello EV. (2000). Childhood and
adolescent growth of patients with sickle cell disease in Aracaju, Sergipe,
North-east Brazil. Annals of Tropical Paediatrics 20: 109-113.

Claster S, Wood JC, Noetzli L, Carson SM, Hofstra TC, Khanna R, Coates TD.
(2009). Nutritional deficiencies in iron overloaded patients with
hemoglobinopathies. American Journal of Hematology 84: 344-348.

Clemente APG, Santos CD, Silva AAB, Martins VJ, Marchesano AC, Fernandes
MB, Albuquerque MP, Sawaya AL. (2012). Mild stunting is associated with
higher blood pressure in overweight adolescents. Arquivos Brasileiros de
Cardiologia 98: 6-12.

Clements MR, Johnson L, Fraser DR. (1987). A new mechanism for induced
vitamin D deficiency in calcium deprivation. Nature 325: 62-65.

190
CMI (Centro de Midia Independente) Brasil. (2007). Bairro da Paz 25 anos de
resistência.
http://www.midiaindependente.org/pt/blue/2007/04/380475.shtml

Colapinto CK, O’Connor DL, Tremblay MS. (2011). Folate status of the
population in the Canadian Health Measures Survey. Canadian Medical
Association Journal 183: 100-106.

Cole CR, Lifshitz F. (2008). Zinc nutrition and growth retardation. Pediatric
Endocrinology Reviews 5: 889-896.

Cole CR, Grant FK, Swaby-Ellis ED, Smith JL, Jacques A, Northrop-Clewes CA,
Caldwell KL, Pfeiffer CM, Ziegler TR. (2010). Zinc and iron deficiency and
their interrelations in low-income African American and Hispanic children
in Atlanta. American Journal of Clinical Nutrition 91: 1027-1034.

Coly AN, Milet J, Diallo A, Ndiaye T, Bénéfice E, Simondon F, Wade S,


Simondon KB. (2006). Preschool stunting, adolescent migration, catch-up
growth, and adult height in young Senegalese men and women of rural
origin. Journal of Nutrition 136: 2412-2420.

Cook JD, Skikne BS, Baynes RD. (1993). Serum transferrin receptor. Annual
Review of Medicine 44: 63-74.

Cook JD, Flowers CH, Skikne BS. (2003). The quantitative assessment of body
iron. Blood 101: 3359-3364.

Correia do Nascimento WR, Cavalcanti IMF, Irmão JI, Rocha FJS. (2009).
Presence of Cryptosporidium spp in children with acute diarrhea in a public
daycare center in Recife, State of Pernambuco. Revista da Sociedade
Brasileira de Medicine Tropical 42: 175-178.

Corvalán C, Uauy R, Flores R, Kleinbaum D, Martorell R. (2008). Reductions in


the energy content of meals served in the Chilean National Nursery School
Council Program did not consistently decrease obesity among beneficiaries.
Journal of Nutrition 138: 2237-2243.

Coutinho GG, Boloni-Bertollo EM, Bertelli EC. (2005). Iron deficiency anemia in
children: a challenge for public health and for society. São Paulo Medical
Journal 132: 88-92.

Couto FD, de Albuquerque ABL, Adorno EV, de Moura Neto JP, de Freitas
Abbehusen L, de Oliveira JLB, dos Reis MG, de Souza Gonçalves M.
(2003). Alpha-thalassemia 2, 3.7 kb deletion and hemoglobin AC
heterozygosity in pregnancy: a molecular and hematological analysis.
Clinical and Laboratory Haematology 25: 29-34.

191
Crompton DWT, Nesheim MC. (2002). Nutritional impact of intestinal
helminthiasis during the human life cycle. Annual Review of Nutrition 22:
35-59.

Cunningham-Rundles S, McNeeley DF, Moon A. (2005). Mechanisms of nutrient


modulation of the immune response. Journal of Allergy and Clinical
Immunology 115: 1119-1128.

da Luz Santos, Clemente APG, Martins VJB, Albuquerque MP, Sawaya AL.
(2010). Adolescents with mild stunting show alterations in glucose and
insulin metabolism. Journal of Nutrition and Metabolism 943070: 1-6.

da Silva Pereira A, Lanzillotti JS, Soares EA. (2010). Day care centers attendance
and preschool children’s nutritional status: a systematic review. Revista
Paulista de Pediatria 28: 366-372.

da Silva Prado M, Assis AM, Martins MC, Nazaré PA, Rezende IF, Conceiçao
ME. (1995). Vitamin A deficiency in children of rural areas of the semi-arid
region in Bahia. Revista de Saúde Pública 29: 295-300.

da Veiga GV, da Cunha AS, Sichieri R. (2004). Trends in overweight among


adolescents living in the poorest and richest regions of Brazil. American
Journal of Public Health 94: 1544-1548.

Darnton-Hill I, Mora JO, Weinstein H, Wilbur S, Nalubola PR. (1999). Iron and
folate fortification in the Americas to prevent and control micronutrient
malnutrition: an analysis. Nutrition Reviews 57: 25-31.

David L. (1991). Common vitamin D-deficiency rickets. In Glorieux FH (ed):


Rickets. Nestle Nutrition Workshop Series. Nestle Ltd., Vevey/Raven, New
York. Vol. 21: 107-122.

Davidsson L, Almgren A, Sandström B, Hurrell RF. (1995). Zinc absorption in


adult humans: the effect of iron fortification. British Journal of Nutrition 74:
417-425.

de Almeida CAN, Ricco RG, Ciampo LAD, Souza AM, Phino AP, de Oliveira
JED. (2004). Factors associated with iron deficiency anemia in Brazilian
preschool children. Jornal de Pediatria 80: 229-234.

de Almeida ER, de Carvalho AT, Nilson EAF, Coutinho JG, Ubarana JA. (2010).
Participatory evaluation of the National Program for vitamin A
supplementation in a municipality in Northeast Brazil. Cadernos de Saúde
Pública 26: 949-960.

de Azevedo MM, Cabral PC, Diniz AS, Fisberg M, Fisberg RM, Arrudat IK.
(2010). Vitamin A deficiency in preschool children of Recife, Northeast of
Brazil. Archivos Latinoamericanos de Nutricion 60: 36-41.

192
de Bem Lignani J, Sichieri R, Burlandy L, Salles-Costa R. (2011). Changes in
food consumption among the Programa Bolsa Família participant families in
Brazil. Public Health Nutrition 14: 785-792.

de Benoist B. (2008). Conclusions of a WHO technical consultation on folate and


vitamin B-12 deficiencies. Food and Nutrition Bulletin 29: S238-244.

de Graaf C. (2011). Why liquid energy results in overconsumption. Proceedings


of the Nutrition Society 70: 162-170.

de Jong N, Gibson RS, Thomson CD, Ferguson EL, McKenzie JE, Green TJ,
Horwath CC. (2001). Selenium and zinc status are suboptimal in a sample
of older New Zealand women in a community-based study. Journal of
Nutrition 131: 2677-2684.

de Lima ALL, da Silva ACF, Konno SC, Conde WL, Benicio MH, Monteiro CA.
(2010). Cause of the accelerated decline in child undernutrition in
Northeastern Brazil (1986-1996-2006). Revista de Saúde Pública 44: 17-27.

de Lucia M, Mitnick ME, Carpenter TO. (2003). Nutritional rickets with normal
circulating 25-hydroxyvitamin D: a call for re-examining the role of dietary
intake in North American infants. Journal of Clinical Endocrinology and
Metabolism 88: 3539-3545.

de Menezes RCE, de Lira PIC, Leal VS, Oliveira JS, Santana SC, Sequeira LA,
Rissin A, Filho MB. (2011). Determinants of stunting in children under five
in Pernambuco, Northeastern Brazil. Revista de Saúde Pública 45: 1079-
1087.

de Novaes Oliveira M, Martorell R, Nguyen P. (2010). Risk factors associated


with hemoglobin levels and nutrition status among Brazilian children
attending daycare centers in São Paulo City, Brazil. Archivos
Latinoamericanos de Nutricion 60: 23-29.

de Onis M, Blössner M. (2000). Prevalence and trends of overweight among


preschool children in developing countries. American Journal of Clinical
Nutrition 72: 1032-1039.

de Onis M, Blössner M, Borghi E, Frongillo EA, Morris R. (2004a). Estimates of


global prevalence of childhood underweight in 1990 and 2015. Journal of
American Medical Association 291: 2600-2606.

de Onis M, Onyango AW, Van den Broeck J, Chumlea Martorell R (WHO


Multicentre Growth Reference Study Group). (2004b). Measurement and
standardization protocols for anthropometry used in the construction of a
new international growth reference. Food and Nutrition Bulletin 25: S27-36.

193
de Pee S, Bloem MW. (2007). The bioavailability of (pro) vitamin A carotenoids
and maximizing the contribution of homestead food production to
combating vitamin A deficiency. International Journal for Vitamin and
Nutrition Research 77: 182-192.

de Pee S, Dary O. (2002). Biochemical indicators of vitamin A deficiency: serum


retinol and serum retinol binding protein. Journal of Nutrition 132: S2895-
2901.

de Silva A, Atukorala S, Weerasinghe I, Ahluwalia N. (2003). Iron


supplementation improves iron status and reduces morbidity in children
with or without upper respiratory tract infections: a randomized controlled
study in Colombo, Sri Lanka. American Journal of Clinical Nutrition 77:
234-241.

Dewey KG, Romero-Abal ME, de Serrano JQ, Bulux J, Peerson JM, Engle P,
Solomons NW. (1997). Effects of discontinuing coffee intake on iron status
on iron-deficient Guatemalan toddlers: a randomized intervention study.
American Journal of Clinical Nutrition 66: 168-176.

Dewey KG, Domellöf M, Cohen RJ, Landa RL, Hernell O, Lönnderdal B. (2002).
Iron supplementation affects growth and morbidity of breast-fed infants:
results of a randomized trial in Sweden and Honduras. Journal of Nutrition
132: 3249-3255.

Dijkhuizen MA, Wieringa FT, West CE, Muherdiyantiningsih, Muhilal. (2001).


Concurrent micronutrient deficiencies in lactating mothers and their infants
in Indonesia. American Journal of Clinical Nutrition 73: 786-791.

do Amaral RS, Tauil PL, Lima DD, Engels D. (2006). An analysis of the impact
of the Schistosomiasis control programme in Brazil. Memórias do Instituto
Oswaldo Cruz 101: 79-85.

dos Reis JC, de Morais MB, Oliva CAG, Fagundes-Neto U. (2007). Breath
hydrogen tests in the diagnosis of environmental enteropathy in children
living in an urban slum. Digestive Diseases and Sciences 52: 1253-1258.

Dossa RAM, Ategbo EAD, de Koning FLHA, van Raaij JMA, Hautvast JGAJ.
(2001). Impact of iron supplementation and deworming on growth
performance in preschool Beninese children. European Journal of Clinical
Nutrition 55: 223-228.

Drewnowski A, Specter SE. (2004). Poverty and obesity: the role of energy
density and energy costs. American Journal of Clinical Nutrition 79: 6-16.

194
Du S, Zhai F, Wang Y, Popkin BM. (2000). Current methods for estimating
dietary iron bioavailability do not work in China. Journal of Nutrition 130:
193-198.

Earley A, Valman HB, Altman DG, Pippard MJ. (1990). Microcytosis, iron
deficiency, and thalassemia in preschool children. Archives of Disease in
Childhood 65: 610-614.

Engle PL, Black MM, Behrman JR, de Mello MC, Gertler PJ, Kapirii L, Martorell
R, Young ME and the International Child Development Steering Group.
(2007). Strategies to avoid the loss of developmental potential in more than
200 million children in the developing world. Lancet 369: 229-242.

Espo M, Kulmala T, Maleta K, Cullinan T, Salin ML, Ashorn P. (2002).


Determinants of linear growth and predictors of severe stunting during
infancy in rural Malawi. Acta Pædiatrica 91: 1364-1370.

Evans T. (2005). Regulation of hematopoiesis by retinoid signaling. Experimental


Hematology 33: 1055-1061.

Fairweather-Tait SJ, Wharf SG, Fox TE. (1995). Zinc absorption in infants fed
iron-fortified weaning food. American Journal of Clinical Nutrition 62: 785-
789.

FAO (Food and Agriculture Organization of the United Nations). (2010). Fats
and fatty acids in human nutrition, FAO food and nutrition paper 91. Rome.

FAO/WHO/UNU (Food and Agriculture Organization of the United Nations/


World Health Organization/United Nations University). (2001). Human
energy requirements: report of a joint FAO/WHO/UNU expert consultation.
Rome.

Fawzi WW, Herrera MG, Willett WC, Nestel P, El Amin A, Mohamed KA.
(1997). Dietary vitamin A intake in relation to child growth. Epidemiology
8: 402-407.

Ferenčík M, Ebringer L. (2003). Modulatory effects of selenium and zinc on the


immune system. Folia Microbiologica 48: 417-426.

Fernald LC, Neufeld LM, Barton LR, Schnaas L, Rivera J, Gertler PJ. (2006).
Parallel deficits in linear growth and mental development in low-income
Mexican infants in the second year of life. Public Health Nutrition 9: 178-
186.

Fernald LC, Neufeld LM. (2007). Overweight with concurrent stunting in very
young children from rural Mexico: prevalence and associated factors.
European Journal of Clinical Nutrition 61: 623-632.

195
Fernandes TF, Diniz A, Cabral PC, Oliveira RS, Lola MM, Silva SMM, Kolsteren
P. (2005). Vitamin A deficiency among preschool children attending public
day care centres in Recife: biochemical and dietetic indicators. Revista de
Nutrição Campinas 18: 471-480.

Ferraz IS, Daneluzzi JC, Vannucchi H, Jordão AA, Ricco RG, Del Ciampo LA,
Martinelli CE, D’Angio Engelber AA, Bonhilha LRC, Custódio VIC.
(2007). Zinc serum levels and their association with vitamin A deficiency in
preschool children. Jornal de Pediatria 83: 512-517.

Ferreira HS, Moura FA, Cabral CR, Florêncio TM, Vieira RC, Assunção ML.
(2009). Short stature of mothers from an area endemic for undernutrition is
associated with obesity, hypertension and stunted children: a population-
based study in the semi-arid region of Alagoas, Northeast Brazil. British
Journal of Nutrition 101: 1239-1245.

Ferreira H, Luciano SC. (2010). Prevalence of extreme anthropometric


measurements in children from Alagoas, Northeastern Brazil. Revista de
Saúde Pública 44: 377-380.

Field CJ, Johnson IR, Schley PD. (2002). Nutrients and their role in host
resistance to infection. Journal of Leukocyte Biology 71: 16-32.

Filteau SM, Tomkins AM. (1994). Micronutrients and tropical infections.


Transactions of the Royal Society of Tropical Medicine and Hygiene 88: 1-
3.

Fishman SM, Christian P, West KP. (2000). The role of vitamins in the prevention
and control of anaemia. Public Health Nutrition 3: 125-150.

Florêncio TM, Ferreira HS, de França AP, Cavalcante JC, Sawaya AL. (2001).
Obesity and undernutrition in a very-low-income population in the city of
Maceió, northeastern Brazil. British Journal of Nutrition 86: 277-283.

Fonseca EOL, Teixeira MG, Barreto ML, Carmo EH, Costa M. (2010).
Prevalence and factors associated with geohelminth infections in children
living in municipalities with low HDI in North and Northeast Brazil.
Cadernos de Saúde Pública 26: 143-152.

Forbes RM, Erdman JW Jr., Parker HM, Kondo H, Ketelson SM. (1983).
Bioavailability of zinc in coagulated soy protein (tofu) to rats and effect of
dietary calcium at a constant phytate:zinc ratio. Journal of Nutrition 113:
205-210.

Fraker PJ, King LE, Laakko T, Vollmer TL. (2000). The dynamic link between
the integrity of the immune system and zinc status. Journal of Nutrition 130:
S1399-1406.

196
Freitas LBL, Shelton TL, Tudge JRH. (2008). Conceptions of US and Brazilian
early childhood care and education: a historical and comparative analysis.
International Journal of Behavioural Development 32: 161-170.

Frisancho AR. (1990). Anthropometric standards for the assessment of growth


and nutritional status. The University of Michigan Press: Ann Arbor,
Michigan.

Fung EB. (2010). Nutritional deficiencies in patients with thalassemia. Annals of


the New York Academy of Sciences 1202: 188-196.

Gakidou E, Oza S, Fuertes CV, Li AY, Lee DK, Sousa A, Hogan MC, Hoorn SV,
Ezzati M. (2007). Improving child survival through environmental and
nutritional interventions. Journal of the American Medical Association 298:
1876-1887.

Galler JR, Bryce CP, Zichlin ML, Fitzmaurice G, Eaglesfield GD, Waber DP.
(2012). Infant malnutrition is associated with persisting attention deficits in
middle adulthood. Journal of Nutrition 142: 788-794.

Garcia LS. (2001). Diagnostic Medical Parasitology, 4th edition. American


Society of Microbiology Press: Washington DC.

Garcia LS. (2009). Practical Guide to Diagnostic Parasitology, 2nd edition.


American Society of Microbiology Press: Washington DC.

Garnica AD. (1981). Trace metals and hemoglobin metabolism. Annals of


Clinical and Laboratory Science 11: 220-228.

Genser B, Strina A, Teles CA, Prado MS, Barreto ML. (2006). Risk factors for
childhood diarrhea incidence. Epidemiology 17: 658-667.

Gerster H. (1997). Vitamin A – functions, dietary requirements and safety in


humans. International Journal of Vitamin and Nutrition Research 67: 71-90.

Gibson RS. (2005). Principles of Nutritional Assessment, 2nd edition. Oxford


University Press: Oxford, UK.

Gibson RS. (2007). The role of diet and host-related factors on nutrient
bioavailability and thus in nutrient-based dietary requirement estimates.
Food and Nutrition Bulletin 28: S77-100.

Gibson RS, Ferguson EL. (2008). An interactive 24-hour recall for assessing the
adequacy of iron and zinc intakes in developing countries. HarvestPlus
Technical Monograph 8. HarvestPlus: Washington DC.

197
Gibson RS, Hotz C. (2001). Nutritional causes of linear growth faltering in infants
during the complementary feeding period. In Martorell R, Haschke F (eds):
Nutrition and Growth. Nestle Nutrition Workshop Series Pediatric Program.
Nestle Ltd., Vevey, Nestec/Philadelphia, Lippincott Williams and Wilkins.
Vol. 47: 159-196.

Gibson RS, Heywood A, Yaman C, Sohlstrom A, Thompson LU, Heywood P.


(1991). Growth in children from the Wosera subdistrict, Papua New Guinea,
in relation to energy and protein intakes and zinc status. American Journal
of Clinical Nutrition 53: 782-789.

Gibson RS, Yeudall R, Drost N, Mtitimuni B, Cullinan T. (1998). Dietary


interventions to prevent zinc deficiency. American Journal of Clinical
Nutrition 68: S484-487.

Gibson RS, Hotz C, Temple L, Yeudall F, Mtitimuni B, Ferguson E. (2000a).


Dietary strategies to combat deficiencies of iron, zinc, and vitamin A in
developing countries: development, implementation, monitoring and
evaluation. Food and Nutrition Bulletin 21: 219-231.

Gibson RS, Skeaff M, Williams S. (2000b). Interrelationship of indices of body


composition and zinc status in 11-yr-old New Zealand children. Biological
Trace Element Research 75: 65-77.

Gibson RS, Yeudall F, Drost N, Mtitimuni BM, Cullinan TR. (2003). Experiences
of a community-based dietary intervention to enhance micronutrient
adequacy of diets low in animal source foods and high in phytate: a case
study in rural Malawian children. Journal of Nutrition 133: S3992-3999.

Gibson RS, Manger MS, Krittaphol W, Pongcharoen T, Gowachirapant S, Bailey


KB, Winichagoon P. (2007). Does zinc deficiency play a role in stunting
among primary school children in NE Thailand? British Journal of Nutrition
97: 167-175.

Gibson RS, Abebe Y, Stabler S, Allen RH, Westcott JE, Stoecker BJ, Krebs NF,
Hambidge KM. (2008). Zinc, gravida, infection, and iron, but not vitamin
B-12 or folate status, predict hemoglobin during pregnancy in Southern
Ethiopia. Journal of Nutrition 138: 581-586.

Gibson RS, Bailey KB, Parnell WR, Wilson N, Ferguson EL. (2011). Higher risk
of zinc deficiency in New Zealand Pacific school children compared with
their Mãori and European counterparts: a New Zealand national survey.
British Journal of Nutrition 105: 436-446.

198
Golden MH. (2009). Proposed recommended nutrient densities for moderately
malnourished children. Food and Nutrition Bulletin 30: S267-342.

Gomes RCF, da Costa THM, Schmitz BAS. (2010). Dietary assessment of pre-
school children from Federal District Brazil. Archivos Latinoamericanos de
Nutricion 60: 168-174.

Gonçalves MS, Bomfim GC, Cerqueira I, Lyra I, Zanette A, Bomfim G, Adorno


EV, Albuquerque AL, Pontes A, Dupuit MF, Fernandes GB, dos Reis MG.
(2003). ßS-haplotypes in sickle cell anaemia patients from Salvador, Bahia,
Northeastern Brazil. Brazilian Journal of Medical and Biological Research
36: 1283-1288.

Goulet O, Lebenthal E, Branski D, Martin A, Antoine JM, Jones P. (2006).


Nutritional solutions to major health problems of preschool children: how to
optimize growth and development. Journal of Pediatric Gastroenterology
and Nutrition 43: S1-3.

Grantham-McGregor S, Ani C. (2001). A review of studies on the effect of iron


deficiency on cognitive development in children. Journal of Nutrition 131:
S649-668.

Grantham-McGregor S, Ani CC. (2002). Nutritional influences on child


development: cross-national perspectives. In Black RE, Michaelson KF
(eds): Public Health Issues in Infant and Child Nutrition. Nestle Nutrition
Workshop Series, Pediatric Program. Nestle Ltd., Vevey,
Nestec/Philadelphia, Lippincott Williams and Wilkins. Vol. 48: 53-69.

Grantham-McGregor S, Powell CA, Walker SP, Himes JH. (1991). Nutritional


supplementation, psychosocial stimulation, and mental development of
stunted children: the Jamaican Study. Lancet 338: 1-5.

Grantham-McGregor S, Cheung YB, Cueto S, Glewwe P, Richter L, Strupp B,


International Child Development Steering Group. (2007). Developmental
potential in the first 5 years for children in developing countries. Lancet
369: 60-70.

Gribble JN, Murray NJ, Menotti EP. (2009). Reconsidering childhood


undernutrition: can birth spacing make a difference? An analysis of the
2002-2003 El Salvador National Family Health Survey. Maternal and Child
Nutrition 5: 49-63.

Grillenberger M, Neumann CG, Murphy SP, Bwibo NO, Weiss RE, Jiang L,
Hautvast J, West CE. (2006). Intake of micronutrients high in animal-source
foods is associated with better growth in rural Kenyan school children.
British Journal of Nutrition 95: 379-390.

199
Grummer-Strawn LM, Ogden CL, Mei Z, Flegal KM, Johnson CL, Kuczmarski
RJ, Curtin LR. (2001). Scientific and practical issues in the development of
the US childhood growth reference. In Martorell R, Haschke F (eds):
Nutrition and Growth. Nestle Nutrition Workshop Series Pediatric Program.
Nestle Ltd., Vevey, Nestec/Philadelphia, Lippincott Williams and Wilkins.
Vol. 47: 21-33.

Gurgel RQ, Cardoso GS, Silva AM, Santos LN, Oliveira RCV. (2005). Children
daycare center: exposition or protection environment to intestinal parasites
infestation in Aracaju, NE Brazil. Revista de Sociedade Brasileira de
Medicina Tropical 38: 267-269.

Gwatkin DR, Bhuiya A, Victora CG. (2004). Making health systems more
equitable. Lancet 364: 1273-1280.

Hadi H, Stoltzfus RJ, Dibley MJ, Moulton LH, West KP, Kjolhede CL, Sadjimin
T. (2000). Vitamin A supplementation selectively improves the linear
growth of Indonesian preschool children: results from a randomized
controlled trial. American Journal of Clinical Nutrition 71: 507-513.

Hadi H, Dibley MJ, West KP. (2004). Complex interactions with infection and
diet may explain seasonal growth responses to vitamin A in preschool aged
Indonesian children. European Journal of Clinical Nutrition 58: 990-999.

Hales CN, Barker DJ. (2001). The thrifty phenotype hypothesis. British Medical
Bulletin 60: 5-20.

Hall A, Hewitt G, Tuffrey V, de Silva N. (2008). A review and meta-analysis of


the impact of intestinal worms on child growth and nutrition. Maternal and
Child Nutrition 4: 118-236.

Hallberg L. (2001). Perspectives on nutritional iron deficiency. Annual Review of


Nutrition 21: 1-21.

Hallberg L, Hulthén L. (2000). Prediction of dietary iron absorption: an algorithm


for calculating absorption and bioavailability of dietary iron. American
Journal of Clinical Nutrition 71: 1147-1160.

Hallberg L, Brune M, Rossander L. (1989). Iron absorption in man: ascorbic acid


and dose-dependent inhibition by phyate. American Journal of Clinical
Nutrition 49: 140-144.

Hambidge KM. (1992). Zinc and diarrhea. Acta Paediatrica 381: 82-86.

Hambidge KM. (1997). Zinc deficiency in young children. American Journal of


Clinical Nutrition. 65: 160-161.

200
Hambidge KM. (2003). Human zinc homeostasis: good but not perfect. Journal of
Nutrition 133: S1438-1442.

Hambidge KM, Krebs NF (2001). Interrelationships of key variables of human


zinc homeostasis: relevance to dietary zinc requirements. Annual Review of
Nutrition 21: 429-452.

Hambidge KM, Miller LV, Westcott JE, Krebs NF. (2008). Dietary reference
intakes for zinc may require adjustment for phytate intake based upon model
predictions. Journal of Nutrition 138: 2363-2366.

Hambidge KM, Miller LV, Westcott JE, Sheng X, Krebs NF. (2010). Zinc
bioavailability and homeostasis. American Journal of Clinical Nutrition 91:
S1478-1483.

Harhay MO, Horton M, Olliaro PL. (2010). Epidemiology and control of human
gastrointestinal parasites in children. Expert Review of Anti-infective
Therapy 8: 219-234.

Hasegawa H, Suzuki K, Suzuki K, Nakaji S, Sugawara K. (2000). Effects of zinc


on the reactive oxygen species generating capacity of human neutrophils
and on the serum opsonic activity in vitro. Luminescence 15: 321-327.

Hautvast JLA, Tolboom JJM, Kafwembe EM, Musonda RM, Mwanakasale V,


van Staveren WA, van ‘t Hof MA, Sauerwein RW, Willems JL, Monnens
LAH. (2000). Severe linear growth retardation in rural Zambian children:
the influence of biological variables. American Journal of Clinical Nutrition
71: 550-559.

Hernández-Diaz S, Peterson KE, Dixit S, Hernández B, Parra A, Barquera S,


Sepúlveda J, Rivera JA. (1999). Association of maternal short stature with
stunting in Mexican children: common genes vs common environment.
European Journal of Clinical Nutrition 53: 938-945.

Hess SY, Peerson JM, King JC, Brown KH. (2007). Use of serum zinc
concentration as an indicator of population zinc status. Food and Nutrition
Bulletin 28: 403-429.

Hillman RS, Ault KA, Rinder HM. (2005). Hematology in clinical practice, a
guide to diagnosis and management: 4th edition. McGraw-Hill: London.

Hjelt K, Pærregaard A, Krasilnikoff PA. (1992). Giardiasis: haematological status


and the absorption of vitamin B12 and folic acid. Acta Pædiatrica 81: 29-34.

Hoffman DJ, Sawaya AL, Verreschi I, Tucker KL, Roberts SB. (2000a). Why are
nutritionally stunted children at increased risk of obesity? Studies of
metabolic rate and fat oxidation in shantytown children from São Paulo,
Brazil. American Journal of Clinical Nutrition 72: 702-707.

201
Hoffman DJ, Roberts SB, Verreschi I, Martins PA, de Nascimento C, Tucker KL,
Sawaya AL. (2000b). Regulation of energy intake may be impaired in
nutritionally stunted children from the shantytowns of São Paulo, Brazil.
Journal of Nutrition 130: 2265-2270.

Hollowell JG, van Assendelft OW, Gunter EW, Lewis BG, Najjar M, Pfeiffer C.
(2005). Hematological and iron-related analytes – reference data for persons
aged 1 year and over: United States 1988-1984. Vital Health Statistics
Series 11, Number 247. Centers for Disease Control and Prevention,
National Center for Health Statistics, USA.

Horta BL, Victora CG, Menezes AM, Halpern R, Barros FC. (1997). Low
birthweight, preterm births and intrauterine growth retardation in relation to
maternal smoking. Paediatric and Perinatal Epidemiology 11: 140-151.

Horton S, Miloff A. (2010). Iodine status and availability of iodized salt: an


across-country analysis. Food and Nutrition Bulletin 31: 214-220.

Hotez PJ. (2008). Forgotten People, Forgotten Diseases: the neglected tropical
diseases and their impact on global health and development. American
Society of Microbiology Press: Washington DC.

Hotz C. (2007). Dietary indicators for assessing the adequacy of population zinc
intakes: Report of a WHO/UNICEF/IAEA/IZiNCG Interagency Meeting on
Zinc Status Indicators. Food and Nutrition Bulletin 28: S430-453.

Hotz C, Brown KH. (2004). Assessment of the risk of zinc deficiency in


populations and options for its control: Technical Document #1. Food and
Nutrition Bulletin 25: S91-203.

Hotz C, Peerson JM, Brown KH. (2003). Suggested lower cutoffs of serum zinc
concentrations for assessing zinc status: reanalysis of the second National
Health and Nutrition Examination Survey data (1976-1980). American
Journal of Clinical Nutrition 78: 756-764.

Humphrey JH. (2009). Child undernutrition, tropical enteropathy, toilets, and


handwashing. Lancet 374: 1032-1035.

Hunt JR. (2005). Dietary and physiological factors that affect the absorption and
bioavailability of iron. International Journal of Vitamin Nutrition Research
75: 375-384.

Hunt JR, Beiseigel JM (2009). Dietary calcium does not exacerbate phytate
inhibition of zinc absorption by women from conventional diets. American
Journal of Clinical Nutrition 89: 839-843.

202
Hunt JR, Beiseigel JM, Johnson LK (2008). Adaption in human zinc absorption as
influenced by dietary zinc and bioavailability. American Journal of Clinical
Nutrition 87: 1336-1345.

Hurrell R, Egli I. (2010). Iron bioavailability and dietary reference values.


American Journal of Clinical Nutrition 91: 1461-1467.

Hurrell RF, Reddy MB, Juillerat M, Cook JD. (2003). Degradation of phytic acid
in cereal porridges improves iron absorption by human subjects. American
Journal of Clinical Nutrition 77: 1213-1219.

Hurrell RF, Reddy MB, Juillerat M, Cook JD. (2006). Meat protein fractions
enhance nonheme iron absorption in humans. Journal of Nutrition 136:
2808-2812.

Iannotti LL, Tielsch JM, Black MM, Black RE. (2006). Iron supplementation in
early childhood: health benefits and risks. American Journal of Clinical
Nutrition 84: 1261-1276.

IBGE (Instituto Brasileiro de Geografia e Estatística). (2008). População residente


por cor ou raça, situação, e sexo. Rio de Janeiro.

IBGE (Instituto Brasileiro de Geografia e Estatística). (2010a). Census 2010. Rio


de Janeiro.

IBGE (Instituto Brasileiro de Geografia e Estatística). (2010b). Pesquisa de


Orçamentos Familiares 2008-2009: Antropometria e estado nutricional de
crianças, adolescents e adultos no Brasil. Rio de Janeiro.

Iqbal TH, Lewis KO, Cooper BT. (1994). Phytase activity in the human and rat
small intestine. Gut 35: 1233-1236.

Issler RMS, Giugliani ERJ, Kreutz GT, Meneses CF, Justo EB, Kreutz VM, Pires
M. (1996). Poverty levels and children’s health status: study of risk factors
in an urban population of low socioeconomic level. Revista de Saúde
Pública 30: 506-511.

Jacobson BH, Lockitch G. (1988). Direct determination of selenium in serum by


graphite-furnace atomic absorption spectrometry with deuterium
background detection and a reduced palladium modifier: age specific
reference ranges. Clinical Chemistry 34: 709–14.

Jalal F, Nesheim MC, Agus Z, Sanjur D, Habicht JP. (1998). Serum retinol
concentrations in children are affected by food sources of ß-carotene, fat
intake, and antihelmintic drug treatment. American Journal of Clinical
Nutrition 68: 623-629.

203
Jardim-Botelho A, Brooker S, Geiger SM, Fleming F, Lopes ACS, Diemert DJ,
Corrêa-Oliveira R, Bethony JM. (2008a). Age patterns in undernutrition and
helminth infection in a rural area of Brazil: associations with ascariasis and
hookworm. Tropical Medicine and International Health 13: 458-467.

Jardim-Botelho A, Raff S, Rodrigues R, Hoffman JH, Diemert DJ, Corrêa-


Oliveira R, Bethony JM, Gazzinelli MF. (2008b). Hookworm, Ascaris
lumbricoides infection and polyparasitism associated with poor cognitive
performance in Brazilian schoolchildren. Tropical Medicine and
International Health 13: 994-1004.

Jiménez JC, Rodríguez N, di Prisco MC, Lynch NR, Costa V. (1999).


Haemoglobin concentrations and infection by Giardia intestinalis in
children: effect of treatment with secnidazole. Annals of Tropical Medicine
and Parasitology 93: 823-827.

Jordão RE, Bernardi JL, Filho A. (2009). Prevalence of iron-deficiency anaemia


in Brazil: a systematic review. Revista Paulista de Pediatria 27: 90-98.

Joseph KS. (2002). Validating the fetal origins hypothesis: an epidemiologic


challenge. In Black RE, Michaelsen KF (eds): Public Health Issues in Infant
and Child Nutrition. Nestle Nutrition Workshop Series Pediatric Program.
Nestle Ltd., Vevey, Nestec/Philadelphia, Lippincott Williams and Wilkins.
Vol. 48: 295-313.

Kapil U, Bhavna A. (2002). Adverse effects of poor micronutrient status during


childhood and adolescence. Nutrition Reviews 60: S84-S90.

Katz N. (1998). Schistosomiasis control in Brazil. Memórias do Instituto Oswaldo


Cruz 93: 33-35.

Kawchak DA, Schall JI, Zemel BS, Ohene-Frempong K, Stallings VA. (2007).
Adequacy of dietary intake declines with age in children with sickle cell
disease. Journal of the American Dietetic Association 107: 843-848.

Khademi R, Ghaffarifar F, Asl HD. (2006). In vitro effect of folic acid and
cobalamin (vitamin B12) on adhesion and growth of Giardia lamblia.
Iranian Journal of Parasitology 1: 47-52.

Khor GK. (2008). Food-based approaches to combat the double burden among the
poor: challenge in the Asian context. Asia Pacific Journal of Clinical
Nutrition 17: 111-115.

Kongsbak K, Wahed MA, Friis H, Thilsted SH. (2006a). Acute-phase protein


levels, diarrhoea, Trichuris trichiura and maternal education are predictors
of serum retinol: a cross-sectional study of children in a Dhaka slum,
Bangladesh. British Journal of Nutrition 96: 725-734.

204
Kongsbak K, Wahed MA, Friis H, Thilsted SH. (2006b). Acute phase protein
levels, T. trichiura, and maternal education are predictors of serum zinc in a
cross-sectional study in Bangladeshi children. Journal of Nutrition 136:
2262-2268.

Korkes F, Kumagai FU, Belfort RN, Szejnfeld D, Abud TG, Kleinman A, Florez
GM, Szejnfeld T, Chieffi PP. (2009). Relationship between intestinal
parasitic infection in children and soil contamination in an urban slum.
Journal of Tropical Pediatrics 55: 42-45.

Koski KG, Scott ME. (2001). Gastrointestinal nematodes, nutrition and immunity:
breaking the negative spiral. Annual Review of Nutrition 21: 297-321.

Kral TVE, Stunkard AJ, Berkowitz RI, Stallings VA, Brown DD, Faith MS.
(2007). Daily food intake in relation to dietary energy density in the free-
living environment: a prospective analysis of children born at different risk
of obesity. American Journal of Clinical Nutrition 86: 41-47.

Krebs NF, Hambidge KM. (2007). Complementary feeding: clinically relevant


factors affecting timing and composition. American Journal of Clinical
Nutrition 85: S639-645.

Krittaphol W, Bailey KB, Pongcharoen T, Winichagoon P, Thomson C, Gibson


RS. (2006). Primary school children from northeast Thailand are not at risk
of selenium deficiency. Asia Pacific Journal of Clinical Nutrition 15: 474-
481.

Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Guo SS, Wei R,
Mei Z, Curtin LR, Roche AF, Johnson CL. (2000). CDC growth charts:
United States. Advance Data 314: 1-27.

Labbaye C, Valtiere M, Barberi T, Meccia E, Pelosi B, Condorelli GL. (1995).


Differential expression and functional role of GATA-2, NF-E2, and GATA-
1 in normal adult hematopoiesis. Journal of Clinical Investigation 95: 2346-
2358.

Lander RL, Enkhjargal T, Batjargal J, Bailey KB, Diouf S, Green TJ, Skeaff CM,
Gibson RS. (2008). Multiple micronutrient deficiencies persist during early
childhood in Mongolia. Asia Pacific Journal of Clinical Nutrition 17: 429-
440.

Lane S, Lloyd D. (2002). Current trends in research into the waterborne parasite
Giardia. Critical Reviews in Microbiology 28: 123-147.

Lanou AM, Berkow SE, Barnard ND. (2005). Calcium, dairy products and bone
health in children and young adults: a re-evaluation of the evidence.
Pediatrics 115: 736-743.

205
Lawless JW, Latham MC, Stephenson LS, Kinoti SN, Pertet AM. (1994). Iron
supplementation improves appetite and growth in anemic Kenyan primary
school children. Journal of Nutrition 124: 645-654.

Le HT, Brouwer ID, Verhoef H, Nguyen KC, Kok FJ. (2007). Anemia and
intestinal parasite infection in school children in rural Vietnam. Asia Pacific
Journal of Clinical Nutrition 16: 716-723.

Leahy KE, Birch LL, Rolls BJ. (2008). Reducing the energy density of multiple
meals decreases the energy intake of preschool-age children. American
Journal of Clinical Nutrition 88: 1459-1468.

Leal LP, Filho MB, de Lira PIC, Figueiroa JN, Osório MM. (2011). Prevalence of
anemia and associated factors in children aged 6-59 months in Pernambuco,
Northeastern Brazil. Revista de Saúde Pública 45: 457-466.

Lee WK, Leung SS, Lui SS, Lau J. (1993). Relationship between long-term
calcium intake and bone mineral content of children aged from birth to 5
years. British Journal of Nutrition 70: 235-248.

Leonard MB, Zemel BS, Kawchak DA, Ohene-Frempong K, Stallings VA.


(1998). Plasma zinc status, growth, and maturation in children with sickle
cell disease. Journal of Pediatrics 132: 467-471.

Leroy JL, García-Guerra A, García R, Dominguez C, Rivera J, Neufeld LM.


(2008). The Oportunidades program increases the linear growth of children
enrolled at young ages in urban Mexico. Journal of Nutrition 138: 793-798.

Leroy JL, Ruel M, Verhofstadt E. (2009). The impact of conditional cash transfer
programmes on child nutrition: a review of evidence using a programme
theory framework. Journal of Development Effectiveness 1: 103-129.

Levin BE. (2006). Metabolic imprinting: critical impact of the perinatal


environmental on the regulation of energy homeostasis. Philosophical
Transactions: Biological Sciences 361: 1107-1121.

Likuksi HJA, Forbes RM (1965). Mineral utilization in the rat. IV. Effects of
calcium and phytic acid on the utilization of dietary zinc. Journal of
Nutrition 85: 230-234.

Lima AAM, Soares AM, Lima NL, Mota RMS, Maciel BLL, Lvalsund MP,
Barrett LJ, Fitzgerald RP, Blaner WS, Guerrant RL. (2010). Effects of
vitamin A supplementation on intestinal barrier function, growth, total
parasitic, and specific Giardia spp infections in Brazilian children: a
prospective randomized, double-blind, placebo-controlled trial. Journal of
Pediatric Gastroenterology and Nutrition 50: 309-315.

206
Lindsay AC, Machado MT, Sussner KM, Hardwick CK, Peterson KE. (2008).
Infant-feeding practices and beliefs about complementary feeding among
low-income Brazilian mothers: a qualitative study. Food and Nutrition
Bulletin 29: 15-24.

Lindsay AC, Machado MT, Sussner KM, Hardwick CK, Kerr LR, Peterson KE.
(2009). Brazilian mothers’ beliefs, attitudes and practices related to child
weight status and early feeding within the context of nutrition transition.
Journal of Biosocial Science 41: 21-37.

Lira PI, Ashworth A, Morris SS. (1998). Effect of zinc supplementation on the
morbidity, immune function, and growth of low-birth-weight, full-term
infants in northeast Brazil. American Journal of Clinical Nutrition 68: S418-
424.

Long KZ, Estrada-Garcia T, Rosado JL, Santos JI, Haas M, Firestone M, Bhagwat
J, Young C, DuPont HL, Hertzmark E, Nanthakumar NN. (2006). The effect
of vitamin A supplementation on the intestinal immune response in Mexican
children is modified by pathogen infections and diarrhea. Journal of
Nutrition 136: 1365-1370.

Long KZ, Rosado JL, Montoya Y, Solano M, Hertzmark E, DuPont HL, Santos
JI. (2007). Effect of vitamin A and zinc supplementation on gastrointestinal
parasitic infections among Mexican children. Pediatrics 120: 846-855.

Lönnerdal B. (2000). Dietary factors influencing zinc absorption. Journal of


Nutrition 130: 1378-1383.

Lönnerdal B. (2004). Interactions between micronutrients: synergies and


antagonisms. In Pettifror JM, Zlotkin S (eds): Micronutrient Deficiencies
during the Weaning Period and the First Years of Life. Nestle Nutrition
Workshop Series Pediatric Program. Nestle Ltd., Vevey/Basel, S. Karger
AG. Vol. 54: 67-81.

Lönnerdal B, Cederbald Å, Davidsson L, Sandström B. (1984). The effect of


individual components of soy formula and cow’s milk formula on zinc
bioavailability. American Journal of Clinical Nutrition 40: 1064-1070.

López de Romaña GL, Cusirramos S, López de Romaña D, Gross R. (2005).


Efficacy of multiple micronutrient supplementation for improving anemia,
micronutrient status, growth, and morbidity of Peruvian infants. Journal of
Nutrition 135: S646-652.

Lozoff B. (2007). Iron deficiency and child development. Food and Nutrition
Bulletin 28: S560-571.

207
Lozoff B, Castillo M, Clark KM, Smith JB. (2012). Iron-fortified vs. low-iron
infant formula: developmental outcome at 10 years. Archives of Pediatrics
and Adolescent Medicine 166: 208-215.

Lynch SR. (1997). Interaction of iron with other nutrients. Nutrition Reviews 55:
102-110.

Lynch S. (2007). Iron metabolism. In Kraemer K, Zimmermann MB (eds):


Nutritional Anemia. Sight and Life Press: Switzerland. Pages 59-76.

Lyons GH, Stangoulis JCR, Graham RD. (2004). Exploiting micronutrient


interaction to optimize biofortification programs: the case for inclusion of
selenium and iodine in the HarvestPlus program. Nutrition Reviews 62:
247-252.

MacDonald RS. (2000). The role of zinc in growth and cell proliferation. Journal
of Nutrition 130: S1500-1508.

Machado RC, Marcari EL, Cristante S, Carareto CMA. (1999). Giardiasis and
helminthiasis in children of both public and private day-care centers and
junior and high schools in the city of Mirassol, São Paulo State, Brazil.
Revista da Sociedade Brasileiera de Medicina Tropical 32: 697-704.

Maehira F, Luyo GA, Miyagi I, Oshiron M, Yamane N, Kuba M, Nakazato Y.


(2002). Alterations of serum selenium concentrations in the acute phase of
pathological conditions. Clinica Chimica Acta 316: 137-146.

Maggini S, Wintergerst ES, Beveridge S, Hornig DH. (2007). Selected vitamins


and trace elements support immune function by strengthening epithelial
barriers and cellular and humoral immune response. British Journal of
Nutrition 98: S29-35.

Manary MC, Abrams SA, Griffin IJ, Quimper MM, Shulman RJ, Hamzo MG,
Chen Z, Maleta K, Manary MJ. (2010). Perturbed zinc homeostasis in rural
3-5-y-old Malawian children is associated with abnormalities in intestinal
permeability attributed to tropical enteropathy. Pediatric Research 67: 671-
675.

Maret W. (2000). The function of zinc metallothionein: a link between cellular


zinc and redox state. Journal of Nutrition 130: S1455-1458.

Marins S, Rezende A. (2011). Perceptions and practices of mothers of Brazilian


pre-school children, with regard to their children’s eating practices.
Enfermeria Global 21: 1-11.

208
Marquis GS, Habicht JP, Lanata CF, Black RE, Rasmussen KM. (1997). Breast
milk or animal-product foods improve linear growth of Peruvian toddlers
consuming marginal diets. American Journal of Clinical Nutrition 66: 1102-
1109.

Marriott BP, White AJ, Hadden L, Davies JC, Wallingford JC. (2010). How well
are infant and young child World Health Organization (WHO) feeding
indicators associated with growth outcomes? An example from Cambodia.
Maternal and Child Nutrition 6: 358-373.

Martins MC, Santos LMP, Assis AMO. (2004). Prevalence of hypovitaminosis A


among preschool children from northeastern Brazil, 1998. Revista de Saúde
Pública 38: 537-542.

Martins MC, Santo LMP, dos Santos SMC, Araújo MPN, Lima AMP, Santana
LAA. (2007). Evaluation of public policies to promote food security and the
fight against hunger, 1995-2002. 3 – The Brazilian national program to
control vitamin A deficiency. Cadernos de Saúde Pública 23: 2081-2093.

Masmas TN, Garly ML, Lisse ID, Rodriques A, Petersen PT, Birgens H. (2006).
Inherited hemoglobin disorders in Guinea-Bissau, West Africa: a population
study. Hemoglobin 30: 355-364.

Matos SMA, Assis AM, Prado M, Strina A, dos Santos LA, de Jesus SR, Barreto
ML. (2008). Giardia duodenalis infection and anthropometric status in
preschoolers in Salvador, Bahia State, Brazil. Cadernos de Saúde Pública
24: 1527-1535.

Matthews R, Garrison Y. (1975). Food yields summarized by different stages of


preparation. Agriculture Handbook No. 102. US Department of Agriculture,
Washington, D.C.

McLachlan SL, Thomson CD, Ferguson EL, McKenzie JE. (2004). Dietary and
biochemical selenium status of urban 6- to 24-month old South Island New
Zealand children and their postpartum mothers. Journal of Nutrition 134:
3290-3295.

McMillen IC, Rattanatray L, Duffield JA, Morrison JL, MacLaughlin SM, Gentili
S, Mühlhäusler BS. (2009). The early origins of later obesity: pathways and
mechanisms. Advances in Experimental Medicine and Biology 646: 71-81.

McNulty H, Pentieva K. (2004). Folate bioavailability. Proceedings of the


Nutrition Society 63: 529-536.

Medeiros-Neto G. (2009). Iodine nutrition in Brazil: where do we stand?


Arquivos Brasileiros de Endocrinoligia e Metabologia 53: 470-474.

209
Mei Z, Cogswell ME, Parvanta I, Lynch S, Beard JL, Stoltfus RJ, Grummer-
Strawn G. (2005). Hemoglobin and ferritin are currently the most efficient
indicators of population response to iron interventions: an analysis of nine
randomized controlled trials. Journal of Nutrition 135: 1974-1980.

Mendez MA, Adair LS. (1999). Severity and timing of stunting in the first two
years of life affect performance on cognitive tests in late childhood. Journal
of Nutrition 129: 1555-1562.

Menzies IS, Zuckerman MJ, Nukajam WS, Somasundaram SG, Murphy B,


Jenkins AP, Crane RS, Gregory GG. (1999). Geography of intestinal
permeability and absorption. Gut 44: 483-489.

Michaelsen KF, Dewey KG, Perez-Exposito AB, Nurhasa M, Lauritzen L, Roos


N. (2011). Food sources and intake of n-6 and n-3 fatty acids in low-income
countries with emphasis on infants, young children (6-24 months), and
pregnant and lactating women. Maternal and Child Nutrition 7: 124-140.

Milagres RCRM, Nunes LC, Pinheiro-Sant’Ana HM. (2007). Vitamin A


deficiency among children in Brazil and worldwide. Ciência Saúde Coletiva
12: 1253-1264.

Miller LV, Krebs NF, Hambidge M. (2007). A mathematical model of zinc


absorption in humans as a function of dietary zinc and phytate. Journal of
Nutrition 137: 135-141.

Miller M, Humphrey J, Johnson E, Marinda E, Brookmeyer R, Katz J. (2002).


Why do children become vitamin A deficient? Journal of Nutrition 132:
S2867-2880.

Minihane AM, Fairweather-Tait SJ. (1998). Effect of calcium supplementation on


daily nonheme-iron absorption and long-term iron status. American Journal
of Clinical Nutrition 68: 96-102.

Molloy AM. (2002). Folate bioavailability and health. International Journal of


Vitamin Nutrition Research 72: 46-52.

Molloy AM, Scott JM. (1997). Microbiological assay for serum, plasma, and red
cell folate using cryopreserved, microtitre plate method. Methods in
Enzymology 281: 43-53.

Mondal D, Peetri WA, Sack RB, Kirkpatrick BD, Haque R. (2006). Entamoeba
histolytica-associated diarrheal illness is negatively associated with the
growth of preschool children: evidence from a prospective study.
Transactions of the Royal Society of Tropical Medicine and Hygiene 100:
1032-1038.

210
Monsen ER, Balintfy JL. (1982). Calculating dietary iron bioavailability:
refinement and computerization. Journal of the American Dietetic
Association 80: 307-311.

Monteiro CA, Benicio MH, Iunes R, Gouveia NC, Taddei JA, Cardoso MA.
(1992). Nutritional status of Brazilian children: trends from 1975-1989.
Bulletin of the World Health Organization 70: 657-666.

Monteiro CA, Conde WL, Popkin BM. (2002). Is obesity replacing or adding to
undernutrition? Evidence from different social classes in Brazil. Public
Health Nutrition 5: 105-112.

Monteiro CA, Conde WL, Popkin BM. (2004). The burden of disease from
undernutrition to overnutrition in countries undergoing rapid nutrition
transition: a view from Brazil. American Journal of Public Health 94: 433-
434.

Monteiro CA, Benicio MH, Konno SC, da Silva AC, de Lima AL, Conde WL.
(2009). Causes for the decline in child under-nutrition in Brazil, 1996-2007.
Revista de Saúde Pública 43: 1-8.

Moore SR, Lima AAM, Conaway MR, Schorling JB, Soares AM, Guerrant RL.
(2001). Early childhood diarrhea and helminthiases associate with long-term
linear growth faltering. International Journal of Epidemiology 30: 1457-
1464.

Moraes LRS, Cancio JA, Cairncross S. (2004). Impact of drainage and sewerage
on intestinal nematode infections in poor urban areas in Salvador, Brazil.
Transactions of the Royal Society of Tropical Medicine and Hygiene 98:
197-204.

Morris SS, Olinto P, Flores R, Nilson EAF, Figueiró AC. (2004). Conditional
cash transfers are associated with a small reduction in the rate of weight
gain of preschool children in Northeast Brazil. Journal of Nutrition 134:
2336-2341.

Mostert V, Hill KE, Burk RF. (2003). Loss of activity of the selenoenzyme
thioredoxin reductase causes induction of hepatic heme oxygenase-1.
Federation of European Biochemical Societies Letters 541: 85-88.

Mukuria AG, Kothari MT, Abderrahim N. (2006). Infant and young child feeding
update. ORC Macro. Maryland, USA.

Müller N, von Allmen N. (2005). Recent insights into the mucosal reactions
associated with Giardia lamblia infections. International Journal for
Parasitology 35: 1339-1347.

211
Muñoz EC, Rosado JL, López P, Furr HC, Allen LH. (2000). Iron and zinc
supplementation improves indicators of vitamin A status of Mexican
preschoolers. American Journal of Clinical Nutrition 71: 789-794.

Murphy S, Beaton G, Calloway D. (1992). Estimated mineral intakes of toddlers:


predicted prevalence of inadequacy in village populations in Egypt, Kenya
and Mexico. American Journal of Clinical Nutrition 56: 565-572.

Nagababu E, Chrest FJ, Rifkind JM. (2003). Hydrogen-peroxide-induced heme


degradation in red blood cells: the protective roles of catalase and
glutathione peroxidase. Biochimica et Biophysica Acta 1620: 211-217.

NCSDH (National Commission on the Social Determinants of Health). (2008).


The Social Causes of Health Inequities in Brazil. Fundação Oswaldo Cruz,
Rio de Janeiro.

Nemeth E, Tuttle MS, Powelson J, Vaughn MB, Donovan A, Ward DM, Ganz T,
Kaplan J. (2004). Hepcidin regulates cellular iron efflux by binding to
ferroportin and inducing its internalization. Science 306: 2090-2093.

NEPA/UNICAMP (Núcleo de Estudos e Pesquisas em Alimentação),


Universidade Estadual de Campinas). (2006). Tabela Brasileira de
Composição de Alimentos (TACO), version 2. Grafica e Editora Flamboyant
Ltda, São Paulo.

Neumann CG, Bwibo NO, Murphy SP, Sigman M, Whaley S, Allen LH, Guthrie
D, Weiss RE, Demment MW. (2003). Animal source foods improve dietary
quality, micronutrient status, growth and cognitive function in Kenyan
school children: background, study design and baseline findings. Journal of
Nutrition 133: S3941-3949.

Neumann CG, Murphy SP, Gewa C, Grillenberger M, Bwibo NO. (2007). Meat
supplementation improves growth, cognitive, and behavioral outcomes in
Kenyan children. Journal of Nutrition 137: 1119-1123.

Newman RD, Moore SR, Lima AAM, Nataro JP, Guerrrant RL, Sears CL. (2001).
A longitudinal study of Giardia lamblia infection in north-east Brazilian
children. Tropical Medicine and International Health 6: 624-634.

Nhien NV, Khan NC, Ninh NX, Huan PV, Hop LT, Lam NT, Ota F, Yabutani T,
Hoa VQ, Motonaka J, Nishikawa T, Nakaya Y. (2008). Micronutrient
deficiencies and anemia among preschool children in rural Vietnam. Asia
Pacific Journal of Clinical Nutrition 17: 48-55.

212
Nishiyama S, Kiwaki K, Miyazaki Y, Hasuda T. (1999). Zinc and IGF-I
concentrations in pregnant women with anemia before and after
supplementation with iron and/or zinc. Journal of the American College of
Nutrition 18: 261-267.

Nkhoma ET, Poole C, Vannappagari V, Hall SA, Beutler E. (2009). The global
prevalence of glucose-6-phosphate dehydrogenase deficiency: a systematic
review and meta-analysis. Blood Cells, Molecules and Diseases 42: 267-
278.

O’Dell BL. (2000). Role of zinc in plasma membrane function. Journal of


Nutrition 130: S1432-1436.

Oken E, Gillman MW. (2003). Fetal origins of obesity. Obesity Research 11: 496-
506.

Olivares JL, Fernandez R, Fleta J, Ruiz MY, Clavel A. (2002). Vitamin B-12 and
folic acid in children with intestinal parasitic infection. Journal of the
American College of Nutrition 21: 109-113.

Oliveira MA, Osório MM, Raposo MCF. (2006). Hemoglobin level and anemia
in children in the State of Pernambuco, Brazil: association with
socioeconomic and food consumption factors. Cadernos de Saúde Pública
22: 2169-2178.

Oliveira MA, Osório MM, Raposo MCF. (2007). Socioeconomic and dietary risk
factors for anemia in children aged 6 to 59 months. Jornal de Pediatria 83:
39-46.

Oliveira M, Martorell R, Nguyen P. (2010). Risk factors associated with


hemoglobin levels and nutritional status among Brazilian children attending
daycare centers in São Paulo city, Brazil. Archivos Latinoamericanos de
Nutricion 60: 23-29.

Olsen A, Namwanje H, Nejsum P, Roepstorff A, Thamsbor SM. (2009).


Albendazole and mebendazole have low efficacy against Trichuris trichiura
in school-age children in Kabale District, Uganda. Transactions of the Royal
Society of Tropical Medicine and Hygiene 103: 443-446.

Oppenheimer SJ. (2001). Iron and its relation to immunity and infectious disease.
Journal of Nutrition 131: 616-635.

O’Rahilly S, Farooqi IS. (2006). Genetics of obesity. Philosophical Transactions


of the Royal Society of London Biological Sciences 361: 1095-1105.

Orioli IM, do Nascimento RL, López-Camelo JS, Castilla EE. (2011). Effects of
folic acid fortification on spina bifida prevalence in Brazil. Birth Defects
Research (Part A): Clinical and Molecular Teratology 91: 831-835.

213
Osório MM, Lira PIC, Ashworth A. (2004). Factors associated with Hb
concentration in children aged 6-59 months in the State of Pernambuco,
Brazil. British Journal of Nutrition 91: 307-314.

Pacheco SS, Braga C, de Souza AI, Figueiroa JN. (2009). Effects of folic acid
fortification on the prevalence of neural tube defects. Revista de Saúde
Pública 43: 565-571.

Paes-Sousa R, Santos LM, Miazaki ES. (2011). Effects of a conditional cash


transfer programme on child nutrition in Brazil. Bulletin of the World
Health Organization 89: 496-503.

PAHO (Pan American Health Organization). (2004). Flour fortification with iron,
folic acid and vitamin B12. Regional meeting report. PAHO: Washington
DC.

PAHO/INCAP (Pan American Health Organization/Institute of Nutrition of


Central America and Panama). (1999). Food based dietary guidelines and
health promotion in Latin America, edited by M Pena and V Molina.
PAHO: Washington DC.

Paim J, Travassos C, Almeida C, Bahia L, Macinko J. (2011). The Brazilian


health system: history, advances, and challenges. Lancet 377: 1778-1797.

Paiva A, Rondó PH, Gonçalves-Carvalho CMR, Illison VK, Pereira JA, Vaz-de-
Lima LRA, de Oliveira CA, Ueda M, Bergamaschi DP. (2006). Prevalence
and factors associated with vitamin A deficiency in preschool children from
Teresina, Piaui, Brazil. Cadernos de Saúde Pública 22: 1979-1987.

Parente JM, da Silva BB, Palha-Dias MP, Zaterka S, Nishimura NF, Zeitune JM.
(2006). Helicobacter pylori infection in children of low and high
socioeconomic status in Northeastern Brazil. American Journal of Tropical
Medicine and Hygiene 75: 509-512.

Parraga IM, Assis AM, Prado MS, Barreto ML, Reis MG, King CH, Blanton RE.
(1996). Gender differences in growth of school-aged children with
schistosomiasis and geohelminth infection. American Journal of Tropical
Medicine and Hygiene 55: 150-156.

Parsons TJ, Power C, Logan S, Summerbell CD. (1999). Childhood predictors of


adult obesity: a systematic review. International Journal of Obesity 23: S1-
107.

Pelto GH, Urgello J, Allen LH, Chave A, Martinez H, Meneses L, Capacchione C,


Backstrand J. (1991). Household size, food intake and anthropometric status
of school-age children in a highland Mexican area. Social Science in
Medicine 33: 1135-1140.

214
Perrone L, Gialanella G, Moro R, Feng SL, Boccia E, Palombo G, Carbone MT,
di Toro R. (1998). Zinc, copper, and iron in obese children and adolescents.
Nutrition Research 18: 183-189.

Pettifor JM. (1991). Dietary calcium deficiency. In Glorieux FH (ed): Rickets.


Nestle Nutrition Workshop Series. Nestle Ltd., Vevey/Raven, New York.
Vol. 21: 123-143.

Pettifor JM. (2004). The epidemiology of vitamin D and calcium deficiency. In


Pettifror JM, Zlotkin S (eds): Micronutrient Deficiencies during the
Weaning Period and the First Years of Life. Nestle Nutrition Workshop
Series Pediatric Program. Nestle Ltd., Vevey/Basel, S. Karger AG. Vol. 54:
21-35.

Pfeiffer CM, Johnson CL, Jain RB, Yetley EA, Picciano MF, Rader JI, Fisher KD,
Mulinare J, Osterloh JD. (2007). Trends in blood folate and vitamin B-12
concentrations in the United States, 1988-2004. American Journal of
Clinical Nutrition 86: 718-727.

Philippi ST. (2002). Tabela de Composicao de Alimentos: suporte para decisão


nutricional, 2nd edition. Editora Metha Ltda: São Paulo.

Pilch SM, Senti FR (eds). (1984). Assessment of the zinc nutritional status of the
U.S. population based on data collected in the second National Health and
Nutrition Examination Survey, 1976-1980. Life Sciences Research Office,
Federation of American Societies for Experimental Biology, Bethesda, MD.

Pinhas-Hamiel O, Newfield RS, Koren I, Agmon A, Lilos P, Phillip M. (2003).


Greater prevalence of iron deficiency in overweight and obese children and
adolescents. International Journal of Obesity 27: 416-418.

Popkin BM, Richards MK, Monteiro CA. (1996). Stunting is associated with
overweight in children of four nations that are undergoing the nutrition
transition. Journal of Nutrition 126: 3009-3016.

Popkin BM. (2002). An overview on the nutrition transition and its health
implications: the Bellagio meeting. Public Health Nutrition 5: 93-103.

Popkin BM. (2011). Contemporary nutritional transition: determinants of diet and


its impact on body composition. Proceedings of the Nutrition Society 70:
82-91.

Post C, Victora CG. (2001). The low prevalence of weight-for-height deficits in


Brazilian children is related to body proportions. Journal of Nutrition 131:
1290-1296.

215
Prado MS, Strina A, Barreto ML, Assis AM, Paz LM, Cairncross S. (2003). Risk
factors for infection with Giardia duodenalis in pre-school children in the
city of Salvador, Brazil. Epidemiology and Infection 131: 899-906.

Prado MS, Cairncross S, Strina A, Barreto ML, Assis AM, Rego S. (2005).
Asymptomatic giardiasis and growth in young children: a longitudinal study
in Salvador, Brazil. Parasitology 131: 51-56.

Prasad AS. (1983). Clinical, biochemical and nutritional spectrum of zinc


deficiency in human subjects: an update. Nutrition Reviews 41: 197-208.

Prentice A. (1993). Does mild zinc deficiency contribute to poor growth


performance? Nutrition Reviews 51: 268-277.

Prentice AM, Gershwin ME, Schaible UE, Keusch GT, Victora CG, Gordon JI.
(2008). New challenges in studying nutrition-disease interactions in the
developing world. Journal of Clinical Investigation 118: 1322-1329.

Prentice AM, van der Merwe L. (2011). Impact of fatty acid status on immune
function of children in low-income countries. Maternal and Child Nutrition
7: 89-98.

Quihui-Cota L, Astiazaran-Garcia H, Valencia ME, Morales-Figueroa GG,


Lopez-Mata MA, Ortiz FV. (2008). Impact of Giardia intestinalis on
vitamin A status in schoolchildren from northwest Mexico. International
Journal of Vitamin and Nutrition Reviews 78: 51-56.

Rahman MM, Mahalanabis D, Hossain S, Wahed MA, Alvarez JO, Siber GR,
Thompson C, Santosham M, Fuchs GJ. (1999). Simultaneous vitamin A
administration at routine immunization contact enhances antibody response
to diphtheria vaccine in infants younger than six months. Journal of
Nutrition 129: 2192-2195.

Rahman MM, Wahed MA, Fuchs GL, Baqui AH, Alvarez JO. (2002). Synergistic
effect of zinc and vitamin A on the biochemical indexes of vitamin A
nutrition in children. American Journal of Clinical Nutrition 75: 92-98.

Rahmathulla L, Underwood BA, Thulasiraj RD, Milton RC. (1991). Diarrhea,


respiratory infections, and growth are not affected by a weekly low-dose
vitamin A supplement: a masked, controlled field trial in children in
southern India. American Journal of Clinical Nutrition 54: 568-577.

Ramakrishnan U, Martorell R, Schroeder G, Flores R. (1999). Role of


intergenerational effects on linear growth. Journal of Nutrition 129: S544-
549.

216
Ramalho RA, Flores H, Saunders C. (2002). Hipovitaminose A no Brasil: um
problema de saude publica. Pan American Journal of Public Health 12: 117-
123.

Ramdath DD, Simeon DT, Wong MS, Grantham-McGregor SM. (1995). Iron
status of school children with varying intensities of Trichuris trichiura
infection. Parasitology 110: 347-351.

Ratledge C. (2007). Iron metabolism and infection. Food and Nutrition Bulletin
28: S515-523.

Rauch AM, Van R, Bartlett AV, Pickering LK. (1990). Longitudinal study of
Giardia lamblia infection in a day care center population. Pediatric
Infectious Disease Journal 9: 186-189.

Rechavi G, Rivella S. (2008). Regulation of iron absorption in


hemoglobinopathies. Current Molecular Medicine 8: 646-662.

Reddy M, Hurrell RF, Cook JD. (2000). Estimation of nonheme-iron


bioavailability from meal composition. American Journal of Clinical
Nutrition 71: 937-943.

Reed JD, Redding-Lallinger R, Orringer EP. (1987). Nutrition and sickle cell
disease. American Journal of Hematology 24: 441-455.

Rees DC, Williams TN, Maitland K, Clegg JB, Weatherall DJ. (1998). Alpha
thalassemia is associated with increased soluble transferrin receptor. British
Journal of Haematology 103: 365-369.

Reilly JJ, Armstrong J, Dorosty AR, Emmett PM, Ness A, Rogers I, Steer C,
Sheriff A, the Avon Longitudinal Study of Parents and Children Study
Team. (2005). Early life risk factors for obesity in childhood: cohort study.
British Medical Journal 330: 1357-1359.

Rink L. Gabriel P. (2000). Zinc and the immune system. Proceedings of the
Nutrition Society 59: 541-552.

Rivera JA, Hotz C, González-Cossío T, Neufeld L, García-Guerra A. (2003). The


effect of micronutrient deficiencies on child growth: a review of results
from community-based supplementation trials. Journal of Nutrition 133:
4010-4020.

Rivera JA, Sotres-Alvarez D, Habicht JP, Shamah T, Villalpando S. (2004).


Impact of the Mexican program for education, health, and nutrition
(Progresa) on rates of growth and anemia in infants and young children.
Journal of the American Medical Association 291: 2563-2570.

217
Rodríguez-Hernández J, Canut-Blasco A, Martín-Sánchez AM. (1996). Seasonal
prevalences of Cryptosporidium and Giardia infections in children
attending day care centers in Salamanca (Spain) studied for a period of 15
months. European Journal of Epidemiology 12: 291-295.

Romulus-Nieuwelink JJ, Doak C, Albernaz E, Victora CG, Haisma H. (2011).


Breastmilk amd complementary food intake in Brazilian infants according
to socio-economic position. International Journal of Pediatric Obesity 6:
e508-514.

Rosado JL. (1999). Separate and joint effects of micronutrient deficiencies on


linear growth. Journal of Nutrition 129: S531-533.

Rosado JL, Diaz M, Gonzalez K, Griffin I, Abrams SA, Preciado R. (2005). The
addition of milk or yogurt to a plant-based diet increases zinc bioavailability
but does not affect iron bioavailability in women. Journal of Nutrition 135:
465-468.

Rosado JL, Caamaño MC, Montoya YA, Solano M, Santos JI, Long KZ. (2009).
Interaction of zinc or vitamin A supplementation and specific parasite
infections on Mexican infants’ growth: a randomized clinical trial. European
Journal of Clinical Nutrition 63: 1176-1184.

Roy CN, Andrews NC. (2005). Anemia of inflammation: the hepcidin link.
Current Opinion in Hematology 12: 107-111.

Rubin RN, Navon L, Cassano PA. (2004). Relationship of serum antioxidants to


asthma prevalence in youth. American Journal of Respiratory and Critical
Care Medicine 169: 393-398.

Ruel MT. (2001). Can food-based strategies help reduce vitamin A and iron
deficiencies? A review of recent evidence. International Food Policy
Research Institute: Washington D.C.

Russell RM. (2000). The aging process as a modifier of metabolism. American


Journal of Clinical Nutrition 72: S529-532.

Russell RM. (2001). Factors in aging that effect the bioavailability of nutrients.
Journal of Nutrition 131: 1359-1361.

Ruz M, Castillo-Duran C, Lara X, Codoceo J, Rebolledo A, Atalah E. (1997). A


14-mo zinc supplementation trial in apparently healthy Chilean preschool
children. American Journal of Clinical Nutrition 66: 1406-1413.

Sackey ME, Weigel MM, Armijos RX. (2003). Predictors and nutritional
consequences of intestinal parasitic infections in rural Ecuadorian children.
Journal of Tropical Pediatrics 49: 17-23.

218
Sakti H, Nokes C, Hertanto WS, Hendratno S, Hall A, Bundy DAP, Satoto.
(1999). Evidence for an association between hookworm infection and
cognitive function in Indonesian school children. Tropical Medicine and
International Health 4: 322-334.

Saldiva SR, Silveira AS, Philippi ST, Torres DM, Mangini AC, de Souza Dias
RM, da Silva RM, Buratini MN, Massad E. (1999). Ascaris-Trichuris
association and malnutrition in Brazilian children. Paediatric and Perinatal
Epidemiology 13: 89-98.

Salgueiro MJ, Zubillaga MB, Lysionek AE, Caro RA, Weill R, Boccio JR.
(2002). The role of zinc in the growth and development of children.
Nutrition 18: 510-519.

Salles-Costa R, Barroso G, de Mello MA, Antunes MM, Yokoo EM. (2010).


Sources of variation in energy and nutrient intakes among children from six
to thirty months old in a population-based study. Cadernos de Saúde Pública
26: 1175-1186.

Sandström B, Davidsson L, Cederblad Å, Lönnerdal B. (1985). Oral iron, dietary


ligands and zinc absorption. Journal of Nutrition 115: 411-414.

Sawaya AL, Amigo H, Sigulem D. (1990). The risk approach in preschool


children suffering malnutrition and intestinal parasitic infection in the city
of São Paulo, Brazil. Journal of Tropical Pediatrics 36: 184-188.

Sawaya AL, Dallal G, Solymos G, Siuza MH, Ventura ML, Roberts SB, Sigulem
DM. (1995). Obesity and malnutrition in a shantytown population in the city
of São Paulo, Brazil. Obesity Research 3: 107-115.

Sawaya AL, Grillo LP, Verreschi I, da Silva AC, Roberts SB. (1998). Mild
stunting is associated with higher susceptibility to the effects of high fat
studies: studies in a shantytown population in São Paulo, Brazil. Journal of
Nutrition 128: S415-420.

Sawaya AL, Roberts S. (2003). Stunting and future risk of obesity: principal
physiological mechanisms. Cadernos de Saúde Pública 19: S21-28.

Sazawal S, Black RE, Ramsan M, Chwaya HM, Stoltzfus RJ, Dutta A, Dhingra U,
Kabole I, Deb S, Othman MK, Kabole FM. (2006). 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, randomized, placebo-controlled trial. Lancet 367:
133-143.

Schaible UE, Kaufmann SH. (2007). Malnutrition and infection: complex


mechanisms and global impacts. PLoS Medicine 4: e806-812.

219
Schroeder DG, Brown KH. (1994). Nutritional status as a predictor of child
survival: summarizing the association and quantifying its global impact.
Bulletin of the World Health Organization 72: 569-579.

Sclar ED, Garau P, Carolini G. (2005). The 21st century health challenge of slums
and cities. Lancet 365: 901-903.

Scott JM. (2007). Nutritional anemia: B-vitamins. In Kraemer K, Zimmermann


MB (eds): Nutritional Anemia. Sight and Life Press: Switzerland. Pages
111-132.

Scott ME, Koski KG. (2000). Zinc deficiency impairs immune responses against
parasitic nematode infections at intestinal and systemic sites. Journal of
Nutrition 130: S1412-1420.

Semba RD. (1994). Vitamin A, immunity, and infection. Clinical Infectious


Diseases 19: 489-499.

Semba RD. (2004). Impact of micronutrient deficiencies on immune function. In


Pettifror JM, Zlotkin S (eds): Micronutrient Deficiencies during the
Weaning Period and the First Years of Life. Nestle Nutrition Workshop
Series Pediatric Program. Nestle Ltd., Vevey/Basel, S. Karger AG. Vol. 54:
137-152.

Semba RD, Muhilal, Scott AL, Natadisastra G, Wirasasmita W, Mele L, Ridwan


E, West KP, Sommer A. (1992). Depressed immune response to tetanus in
children with vitamin A deficiency. Journal of Nutrition 122: 101-107.

Semba RD, Bloem MW. (2002). The anemia of vitamin A deficiency:


epidemiology and pathogenesis. European Journal of Clinical Nutrition 56:
271-281.

Semba RD, Ricks MO, Ferrucci L, Xue QL, Guralnik JM, Fried LP. (2009). Low
serum selenium is associated with anemia among older adults in the United
States. European Journal of Clinical Nutrition 63: 93-99.

Sempértegui F, Estrella B, Egas J, Carrión P, Yerovi L, Díaz S, Lascano M,


Aranha R, Ortiz W, Zabala A, Izurieta R, Griffiths JK. (1995). Risk of
diarrheal disease in Ecuadorian day-care centers. Pediatric Infectious
Diseases Journal 14: 606-612.

Senarath U, Dibley MG, Agho KE. (2007). Breastfeeding practices and associated
factors among children under 24 months of age in Timor-Leste. European
Journal of Clinical Nutrition 61: 387-397.

Sereebutra P, Solomons N, Aliyu MH, Jolly PE. (2006). Sociodemographic and


environmental predictors of childhood stunting in rural Guatemala.
Nutrition Research 26: 65-70.

220
Shankar AH, Prasad AS. (1998). Zinc and immune function: the biological basis
of altered resistance to infection. American Journal of Clinical Nutrition 68:
447-463.

Sherman AR. (2000). Immune dysfunction in iron, copper, and zinc deficiencies.
In Bogden JD, Klevay LM (eds): Clinical Nutrition of the Essential Trace
Elements and Minerals, the Guide for Health Professionals. Humana Press:
New Jersey. Pages 309-331.

Shibukawa AF, da Silva EM, Ichiki WA, Strufaldi MW, Puccini RF. (2008).
Prophylaxis for iron deficiency anemia using ferrous sulphate among infants
followed up at a primary healthcare unit in the municipality of Embu-SP
(2003/2004). São Paulo Medical Journal 126: 96-101.

Shoeps DO, Abreu LC, Valenti VE, Nascimento VG, de Oliveira AG, Gallo PR,
Wajnsztejn R, Leone C. (2011). Nutritional status of pre-school children
from low income families. Nutrition Journal 10: 43-48.

Sichieri R, Taddei JA, Everhart JE. (2000). Influence of parental height and
sociodemographic factors on adolescent height in Brazil. Journal of
Adolescent Health 26: 414-419.

Sichieri R, Chiuve SE, Pereira RA, Lopes AC, Willett WC. (2010). Dietary
recommendations: comparing dietary guidelines from Brazil and the United
States. Cadernos de Saúde Pública 26: 2050-2058.

Sigman M, McDonald A, Neumann C, Bwibo N. (1991). Prediction of cognitive


competence in Kenyan children from toddler nutrition, family
characteristics and abilities. Journal of Child Psychology 32: 307-320.

Sillanpää M, Jansson H. (1992). Status of cadmium, lead, cobalt and selenium in


soils and plants of 30 countries: FAO Soils Bulletin 65. FAO: Rome.

Silva APR, Vitolo MR, Zara LF, Castro CFS. (2006). Effects of zinc
supplementation on 1- to 5-year old children. Jornal de Pediatria 82: 227-
231.

Silva RR, da Silva CAM, de Jesus Pereira CA, de Carvalho Nicolato RL, Negrão-
Corrêa D, Lamounier JA, Carneiro M. (2009). Association between
nutritional status, environmental and socio-economic factors and Giardia
lamblia infections among children aged 6–71 months in Brazil. Transactions
of the Royal Society of Tropical Medicine and Hygiene 103: 512-519.

Simon VGN, Pacheco de Souza JM, Buongermino de Souza S. (2008).


Breastfeeding, complementary feeding, overweight and obesity in pre-
school children. Revista de Saúde Pública 43: 1-9.

221
Simondon KB, Simondon F, Simon I, Diallo A, Bénéfice E, Traissac P, Maire B.
(1998). Preschool stunting, age at menarche and adolescent height: a
longitudinal study in rural Senegal. European Journal of Clinical Nutrition
52: 412-418.

Singh GC, Nair M, Grubesic RB, Connell FA. (2009). Factors associated with
underweight and stunting among children in rural Terai of eastern Nepal.
Asia Pacific Journal of Public Health 21: 144-152.

Sinuon M, Tsuyuoka R, Socheat D, Montresor A, Palmer K. (2005). Financial


costs of deworming children in all primary schools in Cambodia.
Transactions of the Royal Society of Tropical Medicine and Hygiene 99:
664-668.

Smith AD. (2010). Folic acid nutrition: what about the little children? American
Journal of Clinical Nutrition 91: 1408-1410.

Smith JC, Butrimovitz GP, Purdy WC. (1979). Direct measurement of zinc in
plasma by atomic absorption spectroscopy. Clinical Chemistry 25: 1487-
1491.

Smith LC, Haddad L. (2000). Explaining Child Malnutrition in Developing


Countries: a Cross-Country Analysis. International Food Policy Research
Institute: Washington, D.C.

Soliman AT, Khalaf FA, Alhemaidi N, Ali MA, Zyoud MA, Yakoot K. (2008).
Linear growth in relation to the circulating concentrations of insulin-like
growth factor I, parathyroid hormone, and 25-hydroxy vitamin D in children
with nutritional rickets before and after treatment: endocrine adaptation to
vitamin D deficiency. Metabolism Clinical and Experimental 57: 95-102.

Solomons NW. (1993). Pathways to the impairment of human nutritional status by


gastrointestinal pathogens. Parasitology 107: S19-35.

Sommer A, Davidson FR. (2002). Assessment and control of vitamin A


deficiency: the Annecy Accords. Journal of Nutrition 132: S2845-2850.

Sommer A, West KP. (1997). The duration of the effect of vitamin A


supplementation. American Journal of Public Health 87: 467.

Sonati MF, Farah SB, Ramalho AS, Costa FF. (1991). High prevalence of alpha-
thalassemia in a black population of Brazil. Hemoglobin 15: 309-311.

Sousa MC, Gonçalves CA, Bairos VA, Poiares-da-Silva J. (2001). Adherence of


Giardia lamblia trophozoites to Int-407 human intestinal cells. Clinical and
Diagnostic Laboratory Immunology 8: 258-265.

222
Spivak JL, Fisher J, Isaacs MA, Hankins WD. (1992). Protein kinases and
phosphatases are involved in erythropoietin-mediated signal transduction.
Experimental Hematology 20: 500-504.

Stabler SP, Allen RH. (2004). Vitamin B12 deficiency as a worldwide problem.
Annual Review of Nutrition 24: 299-326.

Stanley SL. (2003). Amoebiasis. Lancet 361: 1025-1034.

StataCorp. (2011). Statistical software release 12. Stata Corporation: College


Station, TX USA.

Stein AD, Thompson AM, Waters A. (2005). Childhood growth and chronic
disease: evidence from countries undergoing the nutrition transition.
Maternal and Child Nutrition 1: 177-184.

Stephensen CB. (1999). Burden of infection on growth failure. Journal of


Nutrition 129: S534-538.

Stephensen CB. (2001). Vitamin A, infection, and immune function. Annual


Review of Nutrition 21: 167-192.

Stephensen CB, Gildengorin G. (2000). Serum retinol, the acute phase response,
and the apparent misclassification of vitamin A status in the third National
Health and Nutrition Examination Survey. American Journal of Clinical
Nutrition 72: 1170-1178.

Stephensen CB, Alvarez JO, Kohatsu J, Hardmeier R, Kennedy JI, Gammon RB.
(1994). Vitamin A is excreted in the urine during acute infection. American
Journal of Clinical Nutrition 60: 388-392.

Stephenson LS, Latham MC, Adams, EJ, Kinoti SN, Pertet A. (1993). Weight
gain of Kenyan school children infected with hookworm, Trichuris trichiura
and Ascaris lumbricoides is improved following once- or twice- yearly
treatment with albendazole. Journal of Nutrition 123: 656-665.

Stephenson LS, Latham MC, Ottesen EA. (2000a). Malnutrition and parasitic
helminth infections. Parasitology 121: S23-38.

Stephenson LS, Holland CV, Cooper ES. (2000b). The public health significance
of Trichuris trichiura. Parasitology 121: S73-95.

Stoltzfus RJ, Albonico M, Tielsch JM, Chwaya H, Savioli L. (1997a). School-


based deworming program yields small improvement in growth of Zanzibari
school children after one year. Journal of Nutrition 127: 2187-2193.

223
Stoltzfus RJ, Chwaya HM, Tielsch JM, Schulze KH, Albonico M, Savioli L.
(1997b). Epidemiology of iron deficiency anemia in Zanzibari
schoolchildren: the importance of hookworms. American Journal of Clinical
Nutrition 65: 153-159.

Stoltzfus RJ, Chwaya HM, Montressor A, Albonico M, Savioli L, Tielsch JM.


(2000). Malaria, hookworms and recent fever are related to anemia and iron
status indicators in 0- to 5-y old Zanzibari children and these relationships
change with age. Journal of Nutrition 130: 1724-1733.

Stoltzfus RJ, Chway HM, Montresor A, Tielsch JM, Jape JK, Albonico M, Savioli
L. (2004). Low dose daily iron supplementation improves iron status and
appetite but not anemia, whereas quarterly antihelminthic treatment
improves growth, appetite and anemia in Zanzibari preschool children.
Journal of Nutrition 134: 348-356.

Strina A, Cairncross S, Barreto ML, Larrea C, Prado MS. (2003). Childhood


diarrhea and observed hygiene behaviour in Salvador, Brazil. American
Journal of Epidemiology 157: 1032-1038.

Tamura T, Johnston KE, Freeberg LE, Perkins LL, Goldenberg RL. (1994).
Refrigeration of blood samples prior to separation is essential for the
accurate determination of plasma or serum zinc concentrations. Biological
Trace Element Research 41: 165-173.

Tan AS, Quah TC, Low PS, Chong SS. (2001). A rapid and reliable 7-deletion
multiplex polymerase chain reaction assay for alpha thalassemia. Blood 98:
250-251.

Tanumihardjo SA, Permaesih D, Muhilal. (2004). Vitamin A status and


hemoglobin concentrations are improved in Indonesian children with
vitamin A and deworming interventions. European Journal of Clinical
Nutrition 58: 1223-1230.

Taren DI, Nesheim MC, Crompton DWT, Holland CV, Barbeau I, Rivera G,
Sanjur D, Tiffany J, Tucker K. (1987). Contribution of ascariasis to poor
nutritional status in children from Chiriqui Province, Republic of Panama.
Parasitology 95: 603-613.

Tashima MT, Simões MJS, Leite CQF, Fluminhan A, Nogueira MA, Malaspina
AC. (2009). Classic and molecular study of Giardia duodenalis in children
from a daycare center in the region of Presidente Prudente, São Paulo,
Brazil. Revista do Instituto de Medicina Tropical de São Paulo 51: 19-24.

224
Taylor-Robinson DC, Jones AP, Garner P. (2011). Deworming drugs for treating
soil-transmitted intestinal worms in children: effects on growth and school
performance (review), issue 10. The Cochrane Collaboration. John Wiley &
Sons, Ltd.

Teixeira JC, Heller L, Barreto ML. (2007). Giardia duodenalis infection: risk
factors for children living in sub-standard settlements in Brazil. Cadernos de
Saúde Pública 23: 1489-1493.

Teucher B, Olivares M, Cori H. (2004). Enhancers of iron absorption: ascorbic


acid and other organic acids. International Journal of Vitamin Nutrition
Research 74: 403-419.

Thane CW, Walmsley CM, Bates CJ, Prentice A, Cole TJ. (2000). Risk factors for
poor iron status in British toddlers: further analysis of data from the
National Diet and Nutrition Survey of children aged 1.5–4.5 years. Public
Health Nutrition 3: 433-440.

Theurl I, Fritsche G, Ludwiczek S, Garimorth K, Bellmann-Weiler R, Weiss G.


(2005). The macrophage: a cellular factory at the interphase between iron
and immunity for the control of infections. BioMetals 18: 359-367.

Thibault H, Galan P, Selz F, Preziosi P, Oliveir C, Badoual J, Hercberg S. (1993).


The immune response in iron-deficient young children: effect of iron
supplementation on cell-mediated immunity. European Journal of Pediatrics
152: 120-124.

Thompson RCA. (2000). Giardiasis as a re-emerging infectious disease and its


zoonotic potential. International Journal for Parasitology 30: 1259-1267.

Thompson SC. (1994). Giardia lamblia in children and the child care setting: a
review of the literature. Journal of Paediatrics and Child Health 30: 202-
209.

Thomson CD. (2004). Assessment of requirements for selenium and adequacy of


selenium status: a review. European Journal of Clinical Nutrition 58: 391-
402.

Thomson CD, McLachlan SK, Parnell WR, Wilson N, Wohlers M, Scragg R,


Schaaf D, Fitzgerald ED. (2007). Serum selenium concentrations and
dietary selenium intake of New Zealand children aged 5-14 years. British
Journal of Nutrition 97: 357-364.

Thurlow RA, Winichagoon P, Green T, Wasantwisut E, Pongcharoen T, Bailey


KB, Gibson RS. (2005). Only a small proportion of anemia in northeast
Thai children is associated with iron deficiency. American Journal of
Clinical Nutrition 82: 380-387.

225
Thurlow RA, Winichagoon P, Pongcharoen T, Gowachirapant S, Boonpraderm A,
Manger MS, Bailey KB, Wasantwisut E, Gibson RS. (2006). Risk of zinc,
iodine and other micronutrient deficiencies among school children in North
East Thailand. European Journal of Clinical Nutrition 60: 623-632.

Thurnham DI, Northrop-Clewes CA. (2007). Infection and the etiology of anemia.
In Kraemer K, Zimmermann MB (eds): Nutritional Anemia. Sight and Life
Press: Switzerland. Pages 231-256.

Thurnham DI, Smith E, Flora PS. (1988). Concurrent liquid-chromatographic


assay of retinol, alpha-tocopherol, beta-carotene, alpha-carotene, lycopene,
and beta-cryptoxanthin in plasma with tocopherol acetate as an internal
standard. Clinical Chemistry 34: 377-381.

Thurnham DI, Northrop-Clewes CA, McCullough FSW, Das BS, Lunn PG.
(2000). Innate immunity, gut integrity, and vitamin A in Gambian and
Indian infants. Journal of Infectious Diseases 182: S23-28.

Thurnham DI, McCabe GP, Northrop-Clewes CA, Nestel P. (2003). Effects of


subclinical infection on plasma retinol concentrations and assessment of
prevalence of vitamin A deficiency: meta-analysis. Lancet 362: 2052-2058.

Thurnham DI, Mburu ASW, Mwaniki DL, De Wagt A. (2005). Micronutrients in


childhood and the influence of subclinical inflammation. Proceedings of the
Nutrition Society 64: 502-509.

Thurnham DI, McCabe LD, Haldar S, Wieringa FT, Northrop-Clewes CA,


McCabe GP. (2010). Adjusting plasma ferritin concentrations to remove the
effects of subclinical inflammation in the assessment of iron deficiency: a
meta-analysis. American Journal of Clinical Nutrition 92: 546-555.

Tomkins A. (2003). Assessing micronutrient status in the presence of


inflammation. Journal of Nutrition 133: 1649-1655.

Torun B. (2005). Energy requirements of children and adolescents. Public Health


Nutrition 8: 968-993.

Troen AM, Mitchell B, Sorensen B, Wener MH, Johnston A, Yausi Y, Oral E,


Potter JD, Ulrich CM. (2006). Unmetabolized folic acid in plasma is
associated with reduced natural killer cell cytotoxicity among
postmenopausal women. Journal of Nutrition 136: 189-194.

Tseng M, Charkraborty H, Robinson DT, Mendez M, Kohlmeier L. (1997).


Adjustment of iron intake for dietary enhancers and inhibitors in population
studies: bioavailable iron in rural and urban residing Russian women and
children. Journal of Nutrition 127: 1456-1468.

226
Tsuyuoka R, Bailey JW, Guimarães AM, Gurgel RQ, Cuevas LE. (1999). Anemia
and intestinal parasitic infections in primary school students in Aracaju,
Sergipe, Brazil. Cadernos de Saúde Pública 15: 413-421.

Ulijaszek S. (2001). Ethnic differences in patterns of human growth in stature. In


Martorell R, Haschke F (eds): Nutrition and Growth. Nestle Nutrition
Workshop Series Pediatric Program. Nestle Ltd., Vevey,
Nestec/Philadelphia, Lippincott Williams and Wilkins. Vol. 47: 1-15.

Umesono K, Giguere V, Glass CK, Rosenfeld MG, Evans RM. (1988). Retinoic
acid and thyroid hormone induce gene expression through a common
responsive element.Nature 336: 262-265.

UNICEF (United Nations Children’s Fund). (2012). Children in an urban world.


State of the World’s Children 2012. New York, New York.

USDA (US Department of Agriculture). (2005). National Nutrient Database for


Standard Reference. Release 18. Washington, DC: Nutrient Data
Laboratory, US Government Printing Office; available at
http://www.nal.usda.gov/fnic/foodcomp

USDA (US Department of Agriculture). (2007). USDA Table of Nutrient


Retention Factors, Release 6. Washington, DC: Nutrient Data Laboratory,
US Government Printing Office; available at
http://www.ars.usda.gov/nutrientdata

Utzinger J, Keiser J. (2006). Urbanization and tropical health – then and now.
Annals of Tropical Medicine and Parasitology 100: 517-533.

Van Nhien N, Khan NC, Yabutani T, Ninh NX, Chung LT, Motonaka J, Nakaya
Y. (2008). Relationship of low serum selenium to anemia among primary
school children living in rural Vietnam. Journal of Nutritional Science and
Vitaminology 54: 454-459.

van Stuijvenberg ME, Kruger M, Badenhorst CJ, Mansvelt EPG, Laubscher JA.
(1997). Response to an iron fortification programme in relation to vitamin A
status in 6-12-year-old school children. International Journal of Food
Sciences and Nutrition 48: 41-49.

Veiga GRS, Ferreira HS, Sawaya AL, Calado J, Florêncio TM. (2010).
Dyslipidaemia and undernutrition in children from impoverished areas of
Maceió, State of Alagoas, Brazil. International Journal of Environmental
Research and Public Health 7: 4139-4151.

Velásquez-Melendez G, Schlüssel MM, Brito AS, Silva AA, Lopes-filho JD, Kac
G. (2011). Mild but not light or severe food insecurity is associated with
obesity among Brazilian women. Journal of Nutrition 141: 898-902.

227
Vereecken CA, Keukelier E, Maes L. (2004). Influence of mother’s educational
level on food parenting practices and food habits of young children.
Appetite 43: 93-103.

Victora CG, Vaughan JP, Barros FC, Silva AC, Tomasi E. (2000). Explaining
trends in inequities: evidence from Brazilian child health studies. Lancet
356: 1093-1098.

Victora CG, de Onis M, Hallal PC, Blössner M, Shrimpton R. (2010). Worldwide


timing of growth faltering: revisiting implications for interventions.
Pediatrics 125: 473-480.

Victora CG, Aquino EML, Leal M, Monteiro CA, Barros FC, Szwarcwald CL.
(2011). Maternal and child health in Brazil: progress and challenges. Lancet
377: 1863-1876.

Vidmar S, Carlin J, Hesketh KH, Cole T. (2004). Standardizing anthropometric


measures in children and adolescents with new functions for egen. The
STATA Journal 4: 50-55.

Vieira ACF, Diniz AS, Cabral PC, Oliveira RS, Lóla MMF, Silva SMM,
Kolsteren P. (2007). Nutritional assessment of iron status and anemia in
children under 5 years old at public daycare centers. Jornal de Pediatria 83:
370-376.

Vieira MM, Paik J, Blaner WS, Soares AM, Mota RMS, Guerrant RL, Lima
AAM. (2008). Carotenoids, retinol, and intestinal barrier function in
children from northeastern Brazil. Journal of Pediatric Gastroenterology and
Nutrition 47: 652-659.

Villalpando S, García-Guerra A, Ramírez-Silva CI, Mejía-Rodríguez, Matute G,


Shama-Levy T, Rivera JA. (2003a). Vitamins A, and C and folate status in
Mexican children under 12 years and women 12-49 years of age: a
probabilistic national survey. Salud Pública de México 45: S508-519.

Villalpando S, García-Guerra A, Ramírez-Silva CI, Mejía-Rodríguez, Matute G,


Shama-Levy T, Rivera JA. (2003b). Iron, zinc, and iodide status in Mexican
children under 12 years and women 12-49 years of age: a probabilistic
national survey. Salud Pública de México 45: S520-529.

Villamor E, Fawzi WW. (2005). Effects of vitamin A supplementation on immune


responses and correlation with clinical outcomes. Clinical Microbiology
Reviews 18: 446-464.

Viteri FE, Gonzalez H. (2002). Adverse outcomes of poor micronutrient status in


childhood and adolescence. Nutrition Reviews 60: S77-83.

228
Votruba-Drzal E, Coley RL, Chase-Lansdale P. (2004). Child care and low-
income children’s development: direct and moderated effects. Child
Development 75: 296-312.

Wallock LM, King JC, Hambidge KM, English-Westcott JE, Pritts J. (1993).
Meal-induced changes in plasma, erythrocyte, and urinary zinc
concentrations in adult women. American Journal of Clinical Nutrition 58:
695-701.

Walser BL, Lima AAM, Guerrant RL. (1996). Effect of high-dose oral vitamin A
on diarrheal episodes among children with persistent diarrhea in a Northeast
Brazilian community. American Journal of Tropical Medicine and Hygiene
54: 582-585.

Walter PB, Fung EB, Killilea DW, Jiang Q, Hudes M, Madden J, Porter J, Evans
P, Vichinsky E, Harmatz P. (2006). Oxidative stress and inflammation in
iron-overloaded patients with ß-thalassemia or sickle cell disease. British
Journal of Haematology 135: 254-263.

Wang Y, Popkin B, Zhai F. (1998). The nutritional status and dietary pattern of
Chinese adolescents, 1991 and 1993. European Journal of Clinical Nutrition
52: 908-916.

Weiss G. (2002). Iron and immunity: a double-edged sword. European Journal of


Clinical Investigation 32: 70-78.

West CE, Eilander A, van Lieshout M. (2002). Consequences of revised estimates


of carotenoid bioefficacy for dietary control of vitamin A deficiency in
developing countries. Journal of Nutrition 132: S2920-2926.

West KP, LeClerq SC, Shrestha SR, Wu LS, Pradhan EK, Khatry SK, Katz J,
Adhikari R, Sommer A. (1997). Effects of vitamin A on growth of vitamin
A-deficient children: field studies in Nepal. Journal of Nutrition 127: 1957-
1965.

West KP, Gernand AD, Sommer A. (2007). Vitamin A in nutritional anemia. In


Kraemer K, Zimmermann MB (eds): Nutritional Anemia. Sight and Life
Press: Switzerland. Pages 133-153.

Whittaker P. (1998). Iron and zinc interactions in humans. American Journal of


Clinical Nutrition 68: 442-446.

WHO (World Health Organization). (1982). Control of vitamin A deficiency and


xerophthalmia. WHO technical report series 672. WHO: Geneva.

229
WHO (World Health Organization). (1996). Indicators for assessing vitamin A
deficiency and their application in monitoring and evaluating intervention
programmes. WHO/NUT/96.10. WHO: Geneva.

WHO (World Health Organization). (1998). Preparation and use of food-based


dietary guidelines, WHO technical report series 880. WHO: Geneva.

WHO (World Health Organization). (2000). Obesity: preventing and managing


the global epidemic. WHO Technical Report Series 894. WHO: Geneva.

WHO (World Health Organization). (2001). Iron deficiency anaemia: assessment,


prevention and control – a guide for programme managers.
WHO/NHD/01.3. WHO: Geneva.

WHO (World Health Organization). (2002). Prevention and control of


schistosomiasis and soil-transmitted helminthiasis. WHO Technical Report
Series 912. WHO: Geneva.

WHO (World Health Organization). (2003). Diet, nutrition and the prevention of
chronic diseases. WHO technical report series 916. WHO: Geneva.

WHO (World Health Organization). (2004a). Anthropometry training video: the


WHO multicentre growth reference study (DVD version). WHO: Geneva.

WHO (World Health Organization). (2004b). Training manual on diagnosis of


intestinal parasites, tutor’s guide. WNO/CTD/SIP/98.2 CD-Rom. WHO:
Geneva.

WHO (World Health Organization). (2008). WHO global database on child


growth and malnutrition: Brazil. Publication October 15. WHO: Geneva.

WHO (World Health Organization). (2010). Indicators for assessing infant and
young child feeding practices: Part 2 Measurement and Part 3 Country
Profiles. WHO: Geneva.

WHO (World Health Organization) Expert Consultation. (2004). Appropriate


body-mass index for Asian populations and its implications for policy and
intervention strategies. Lancet 363: 157-163.

WHO (World Health Organization) Multicentre Growth Reference Study Group.


(2006). Complementary feeding in the WHO multicentre growth reference
study. Acta Pædiatrica 450: 27-37.

WHO/CDC (World Health Organization/Centers for Disease Control and


Prevention). (2007). Assessing the iron status of populations, 2nd edition.
WHO: Geneva.

230
WHO/FAO (World Health Organization/Food and Agriculture Organization of
the United Nations). (2004). Vitamin and mineral requirements in human
nutrition, 2nd edition. WHO: Geneva.

WHO/FAO (World Health Organization/Food and Agriculture Organization of


the United Nations). (2006). Guidelines on food fortification with
micronutrients, edited by Allen L, de Benoist B, Dary O, Hurrell R. WHO:
Geneva.

WHO/FAO/IAEA (World Health Organization/Food and Agriculture


Organization/ International Atomic Energy Association). (1996). Trace
elements in human health and nutrition, 2nd edition. WHO: Geneva.

Wieringa FT, Dijkhuizen MA, West CE, Northrop-Clewes CA, Muhilal. (2002).
Estimation of the effect of the acute phase response on indicators of
micronutrient status in Indonesian infants. Journal of Nutrition 132: 3061-
3066.

Wintergerst ES, Maggini S, Hornig DH. (2007), Contribution of selected vitamins


and trace elements to immune function. Annals of Nutrition and Metabolism
51: 301-323.

Wirth JJ, Fraker PJ, Kierszenbaum F. (1989). Zinc requirement for macrophage
function: effect of zinc deficiency on uptake and killing of a protozoan
parasite. Immunology 68: 114-119.

Wong CT, Saha N. (1987). Haemoglobin, serum iron transferrin, ferritin


concentrations and total iron-binding capacity in erythrocyte glucose-6-
dehydrogenase deficiency. Tropical and Geographical Medicine 39: 350-
353.

Worwood M. (2002). Serum transferrin receptor assays and their application.


Annals of Clinical Biochemsitry 39: 221-230.

Yamaguchi M, Fukagawa M. (2005). Role of zinc in regulation of protein tyrosine


phosphatase activity in osteoblastic MC3T3_E1 cells: zinc modulation of
insulin-like growth factor-I’s effect. Calcified Tissue International 76: 32-
38.

Yanoff LB, Menzie CM, Denkinger B, Sebring NG, McHugh T, Remaley AT,
Yanovski JA. (2007). Inflammation and iron deficiency in the hypoferremia
of obesity. International Journal of Obesity 31: 1412-1419.

Yehuda S, Rabinovitz S, Mostofsky DI. (2006). Nutritional deficiencies in


learning and cognition. Journal of Pediatric Gastroenterology and Nutrition
43: S22-25.

231
Yip R, Ramakrishnan U. (2002). Experiences and challenges in developing
countries. Journal of Nutrition 132: S827-830.

Young MF, Glahn RP, Ariza-Nieto M, Inglis J, Olbina G, Westerman M, O’Brien


KO. (2009). Serum hepcidin is significantly associated with iron absorption
from food and supplemental sources in healthy young women. American
Journal of Clinical Nutrition 89: 533-538.

Zemel BS, Kawchak DA, Fung EB, Ohene-Frempong K, Stallings VA. (2002).
Effect of zinc supplementation on growth and body composition in children
with sickle cell disease. American Journal of Clinical Nutrition 75: 300-307.

Zijp IM, Korver O, Tijburg LBM. (2000). Effect of tea and other dietary factors
on iron absorption. Critical Reviews in Food Science and Nutrition 40: 371-
398.

Zimmermann MB, Biebinger R, Rohner F, Dib A, Zeder C, Hurrell RF, Chaouki


N. (2006). Vitamin A supplementation in children with poor vitamin A and
iron status increases erythropoietin and haemoglobin concentrations without
changing total body iron. American Journal of Clinical Nutrition 84: 580-
586.

Zimmermann MB. (2007). Interactions between iron and vitamin A, riboflavin,


copper, and zinc in the etiology of anemia. In Kraemer K, Zimmermann MB
(eds): Nutritional Anemia. Sight and Life Press: Switzerland. Pages 199-
213.

Zimmermann MB, Fucharoen S, Winichagoon P, Sirankapracha P, Zeder C,


Gowachirapant S, Judprasong K, Tanno T, Miller JL, Hurrell RF. (2008a).
Iron metabolism in heterozygotes for hemogbobin E (HbE), α-thalassemia
1, or ß-thalassemia and in compound heterozygotes for HbE/ ß-thalassemia.
American Journal of Clinical Nutrition 88: 1026-1031.

Zimmermann MB, Zeder C, Muthayya S, Winichagoon P, Chaouki N, Aeberli I,


Hurrell RF. (2008b). Adiposity in women and children from transition
countries predicts decreased iron absorption, iron deficiency and a reduced
response to iron fortification. International Journal of Obesity 32: 1098-
1104.

232
Appendices

233
234
Appendix A Socio-demographic questionnaire
ESTUDO DE MICRONUTRIENTES EM SALVADOR
Recrutamento e Demografia (Primeira coleta de data)
SALVADOR MICRONUTRIENT STUDY
Recruitment and Demographics (Baseline only)

Entrevistador
Interviewer
Data da entrevista (dia/mês/ano)
Date of Interview (dd/mm/yyyy)

Nome da crèche
Day care centre name

Mãe ou Guradião / Mother or Guradian


Mãe/Mother ou Guardião/Guardian? Especificar/Specify________________
Nome da mãe
Mother’s name
Endereço da mãe
Mother’s address
Data de nascimento da mãe (dia/mês/ano)
Mother’s date of birth (dd/mm/yyyy)
Idade da mãe (ano)
Mother’s age (y)

Criança do estudo / Study child


Nome da criança
Child’s name
Número da criança
Child's number

Data de nascimento da criança (certificado de nascimento ou dados da creche)


(dia/mês/ano) Child’s date of birth (from certificate or from Center) (dd/mm/yyyy)

Idade da criança (mês)


Child’s age (months)

Sexo da criança (M/F)


Child’s sex (M/F)

A ordem de nascimento da criança / Child's birth order


0 = primeiro ou segunda criança / first or second child
1 = terceiro ou subseqüente / third or subsequent

235
Irmãos / Siblings / Inclua qualquer outra criança que more em tempo integral na casa,
até mesmo que elas não sejam parentes. Se houver outras crianças menores que 3
anos, preencha os detalhes abaixo. / Include any other children living full-time in the
household, even if they are the mother’s sister’s children. If there are other children
under 3 years, fill in details below.

nº1- Crianças abaixo de 3 anos / Sibling #1 under 3 years


Nome da criança
Child’s name

Data de nascimento da criança (dia/mês/ano)


Child’s date of birth (dd/mm/yyyy)

Idada da criança (mês)


Child’s age (months)

Sexo da criança (M/F)


Child’s sex (M/F)

nº 2 - Crianças abaixo de 3 anos / Sibling #2 under 3 years


Nome da criança
Child’s name
Data de nascimento da criança (dia/mês/ano)
Child’s date of birth (dd/mm/yyyy)

Idade da criança (mês)


Child’s age (months)
Sexo da criança (M/F)
Child’s sex (M/F)
Mãe / Mother
Ocupação de mãe / Mother’s occupation
1 = Assalariada regular (todo mês recebe salário) / Regular wage earner
2 = Assalariada ocasional (não recebe todo mês, faz bicos) / Casual wage earner
3 = Dona de comércio / Business or trade owner
4 = Trabalhadora doméstica (fora de casa) / Domestic worker (works outside of home)
5 = Desempregada / Unemployed
6 = Dona de casa / Housewife
7 = Outro/Other (especificar/ Specify) _______________________

Você trabalha fora de casa?


Do you work outside the home?
0 = não / no 1 = sim / yes
Nível de educação / Educational Level
0 = Nenhuma instrução / No schooling
1 = Primário ou Ensino Fundamental / Primary (1-7 Grades)
2 = Ensino médio / Secondary (8 -10 Grades)
3 = Ensino técnico ou afins / Tertiary colleges
4 = Universidade / University

236
Estado civil / Marital Status
1 = Casada / Married
2 = Viúva / Widow
3 = Divorciada / Divorced
4 = Solteira/ Single
5 = "Mora junto" / living together

Pai / Father
Ocupação do pai / Father’s occupation
1 = Assalariado regular (recebe todo mês salário) / Regular wage earner
2 = Assalariado ocasional (não recebe todo mês, faz bicos) / Casual wage earner
3 = Dono de comércio / Business or trade owner
4 = Trabalhador doméstico / Domestic worker
5 = Desempregado / Unemployed
6 = Outro (especifique) / Other (specify) ______________________

Nível de educação / Educational Level


0 = Nenhuma instrução / No schooling
1 = Primário ou Ensino Fundamental / Primary (1-7 Grades)
2 = Ensino médio / Secondary (8 -10 Grades)
3 = Ensino técnico ou afins / Tertiary colleges
4 = Universidade / University

Agregado familiar (Pessoa que mora na casa e não faz ou não parte da
família)/Household
Número total de adultos (mais18 anos) na casa
Total number of adults (over 18 years) in the household
Número de crianças menores de 5 anos na casa
Number of children < 5 years in the household
Número de crianças de 6 a 12 anos na casa
Number of children 6-12 years in the household
Número de crianças de 13 a 17 anos na casa
Number of children 13-17 years in the household
Número total de pessoas na casa
Total number of persons in the household
Número de quartos na casa
Number of rooms in the household
Renda familiar total / Total family income
0 = um ou menos de um salário mínimo mensal / one or less min. monthly salary
1 = mais de um salário mínimo mensal / more than one minimum monthly salary
Você possui casa própria? / Do you own your own house?
0 = não/no; 1 = sim/yes
Que tipo de construção é sua casa? / What kind of construction is your house?
0 = construção permanente (com tijolos) / permanent construction
1 = construção móvel / movable construction

237
O que é o material do assoalho de sua casa?
What is the material of the floor of your house?
1 = terra ou chão batido / earth or mud
2 = blocos de madeira / wood planks or blocks
3 = concreto ou cimento / concrete or cement
4 = azuleijo ou telhas de cerâmica / terrazo or ceramic tiles
5 = Outro (especifique) / Other (specify) ______________________

Tem camas em sua casa? / Does the family have beds in your house?
0 = não/no; 1 = sim/yes
Tem energia elétrica em casa? / Does the family have electricity in your house?
0 = não/no; 1 = sim/yes

Tem rádio em casa? / Does the family own a radio?


0 = não/no; 1 = sim/yes
Tem aparelho de som em casa ?/ Does the family own a stereo or CD player?
0 = não/no; 1 = sim/yes
Tem televisão em casa? / Does the family own a television?
0 = não/no; 1 = sim/yes
Tem telefone ou celular? / Does the family own a telephone or mobile?
0 = não/no; 1 = sim/yes
Tem geladeira em casa? / Does the family own a refrigerator?
0 = não/no; 1 = sim/yes
Tem bicicleta própria? / Does the family own a bicycle?
0 = não/no; 1 = sim/yes
Tem moto própria ? / Does the family own a motorcycle?
0 = não/no; 1 = sim/yes
Tem carro ou caminhão próprio ? / Does the family own a car or truck?
0 = não/no; 1 = sim/yes
Número de refeições que você normalmente faz por dia (não conte lanches)
Number of meals per day you usually eat (exclude snacks)
Você recebeu cesta básica ou alguma doeção de alimentos nos últimos 06 meses? /
Have you received any food assistance in the last 6 months?
0 = não/no; 1 = sim/yes
Sua casa tem jardim? / Does your house have a garden?
0 = não/no; 1 = sim/yes
Você planta algum alimento para consumo (exceto temperos)? / Is food for the family
grown in the garden?
0 = não/no; 1 = sim/yes
Alguém fuma cigarros na casa? / Does anyone regularly smoke cigarettes in the house?
0 = não/no; 1 = sim/yes

A mãe fuma cigarro? Does the mother regularly smoke cigarettes in the house?
0 = não/no; 1 = sim/yes
História de asma entre os irmãos e a mãe da criança? Does the mother or child's
siblings have a history of asthma?
0 = não/no; 1 = sim/yes

238
Higiene e saneamento (Primeira coleta de dados)
Lembre-se de fazer estas perguntas de uma maneira delicada
Remember to ask these questions in a non-judgemental manner
Que tipo de calçamento estão em volta da sua casa? / What kind of roads are by your
house? 0 = Pavimentado / Paved: 1 = Não pavimentado / Non-paved
Sua casa inunda quando chove muito forte? / Does your house get flooded when it rains
very hard? 0 = Não/No 1 = Sim/Yes
Que tipo de banheiro sua família usa? / What type of toilet facilities does your family
use?
1 = Banheiro em recinto fechado e privado / Private indoor toilet
2 = Banheiro compartilhado entre várias famílias / WC shared between several families
3 = Banheiro do lado de fora da casa (tipo fossa) / Pit latrine (private or communal)
4 = Não há banheiro / No household facility
Há esgoto aberto próximo a sua casa? / Is there open sewage near your house?
0 = Não/No 1 = Sim/Yes
Qual é a principal fonte de água usada por sua família para beber?
What is the main source of water used by your family for drinking?
1 = Água encanada / Private indoor piped water
2 = Estação de água da comunidade / Community water station
3 = Rio, córrego, lago ou afins / River/stream/spring
4 = Poço coberto / Protected Well
5 = Poço não coberto / Unprotected Well
6 = Mineral ou filtrada / Mineral or filtred water

Sua família geralmente ferve a água que bebe? / Does your family generally boil their
drinking water? 0 = Não/No 1 = Sim/Yes
Você ferve a água que dá para sua criança beber? / Do you boil the drinking water you
give to your child to drink? 0 = Não/No 1 = Sim/Yes
Quem prepara a comida da sua criança? / Who usually cooks your child’s food?
1 = Mãe / Mother
2 = Quem cuida da criança / Care-giver
3 = Irmãos / Sibling
4 = Outros / Other
Quem normalmente dá a comida para sua criança? / Who usually feeds your child?
1 = Mãe / Mother
2 = Quem cuida da criança / Care-giver
3 = Irmãos / Sibling
4 = Outros / Other
Você sempre cobre a comida até a hora em que se vai comer? / Do you always cover
food until it is eaten? 0 = Não/No 1 = Yes/Sim
Você sempre lava os alimentos crus antes de dar para sua criança? Do you always
wash raw food before giving it to your child?
0 = Não/No 1 = Sim/Yes
Você sempre lava suas mãos (com ou sem sabão) antes de alimentar sua criança? / Do
you always wash your hands (with or without soap) before feeding your child?
0 = Não/No 1 = Sim/Yes
Você usa outras formas de higiene (por exemplo lavar as mãos de sua criança; lavar o
rosto de sua criança)? / Do you use other hygiene strategies (e.g. washing your child's
hands; washing your child's face)? 0 = Não/No 1 = Sim/Yes
Por favor descreva / Please describe___________________________

239
Quando bebê (amamentação) e quando criança
Breast feeding, infant and child nutrition
As respostas a todas as perguntas são baseadas nas respostas da mãe ou quem cuida
da criança. Cada pergunta deve ser feita como está escrito no questionário: isto é muito
importante. Se a mãe ou quem cuida da criança não compreenderem a pergunta, você
pode precisar de outras perguntas para esclarecer a dúvida.
Responses to all the questions are based on mother's (or other caregivers) recall. Each
question must be asked as it is written on the questionnaire: this is very important. If the
mother or caregiver does not understand the question, you may need to use extra
probing questions.

Além do que as perguntas, há as palavras que aparecem em maiúsculo. Estas são


instruções do entrevistador. Estes não devem ser lidos alto à mãe ou ao cuidador. A
maioria de perguntas devem ser respondidas com os códigos. É importante que você
não lê estas respostas possíveis alto. Quando você faz uma pergunta, você deve
escutar a resposta da mãe ou do cuidador. Verifique então as opções ao lado da
categoria para escolher a melhor resposta.
In addition to the questions, there are words that appear in capital letters. These are
interviewer instructions. These should NOT be read aloud to the mother or caregiver.
Most questions have precoded responses. It is important that you do not read these
possible responses aloud. When you ask a question, you should listen to the answer of
the mother or caregiver. Then check the box next to the category that best matches the
answer.

Quem está respondendo?/ Who is responding?: 1 = mãe/ mother, 2 = outro/other

Você amamentou a criança (NOME)? / Did you ever breastfeed (NAME)?


0 = Não/No 1 = Sim/Yes
Quanto tempo depois do nascimento você amamentou a criança (NOME) pela primeira
vez? / How long after birth did you first put (NAME) to the breast?
0 = Imediatamente ou dentro da primeira hora depois de nascimento /
Immediately/within first hour after birth
1 = Depois da primeira hora / After the first hour
2 = Não se lembra ou não sabe / Don't remember/don't know
Durante os primeiros três dias depois do nascimento, você amamentou (NOME) com o
leite do seu peito?
During the first three days after delivery, did you give (NAME) the liquid that came from
your breasts?
0 = Não/No 1 = Sim/Yes 2 = Não sabe/Don't know
Durante os primeiros três dias depois do nascimento, você (NOME) deu outro alimento
ou outra bebida para criança antes de amamenta-la?
During the first three days after delivery, did you give (NAME) anything else to eat or
drink before feeding him/her breastmilk?
0 = Não/No 1 = Sim/Yes 2 = Não sabe/Don't know

240
Se Sim, o que você (NOME) deu? / If Yes, what did you give (NAME)?
(NÃO LEIA A LISTA - REGISTRAR TUDO QUE MENCIONAREM. / DO NOT READ
THE LIST - RECORD ALL MENTIONED)
1 = Leite (diferente do leite do seu peito) / Milk (other than breast milk )
2 = Manteiga / Butter
3 = Água corrente / Plain water
4 = Água com açúcar e (ou) sal / Water with sugar and/or salt
5 = Suco de fruta / Fruit juice
6 = Chá ou café / Tea or coffee
7 = Leite enlatado ou em pó / Infant formula
8 = Outros / Other
9 = Dois ou mais dos anteriores / Two or more of the above

Por quanto tempo você amamentou (NOME) em meses? SE MENOS DE UM MÊS,


COLOQUE 00 MESES
For how long did you breastfeed (NAME) in months? IF LESS THAN ONE MONTH,
RECORD 00 MONTHS

Por que você deixou de amamentar? / Why did you stop breastfeeding?
1 = Mãe grávida / Mother pregnant
2 = Mãe doente / Mother sick
3 = Mãe cansou de amamentar no peito / Mother tired of breast feeding
4 = Introduziu sólidos na alimentação / Introduced solids
5 = Amamentação fez a criança ficar doente / Breast milk making child sick
6 = Não tinha leite suficiente / Not enough milk
7 = A criança não quis/ Child refused breastmilk
8 = Outros_________(Por favor especifique) / Other____(Please specify)

Eu gostaria de perguntar os tipos de líquidos (NOME) que a criança bebeu ontem


durante o dia e a noite. Ela bebeu ontem durante o dia ou noite quaisquer dos líquidos
seguintes? LEIA A LISTA DE LÍQUIDOS. ESCREVA O NÚMERO CORRESPONDENTE
AO LÍQUIDO INGERIDO PELA CRIANÇA E SE FOR MAIS DE UM, ANOTE TODOS E
COLOQUE TAMBÉM O NÚMERO 8.
Now I would like to ask about the type of liquids (NAME) drank yesterday during the day and night.
Did (NAME) drink any of the following liquids yesterday during the day or night? READ THE LIST
OF LIQUIDS. WRITE ALL THE NUMBERS IN THE BOX IF MORE THAN ONE LIQUID. PLEASE
ALSO WRITE ‘8’ BELOW THE NUMBERS IN THE BOX IF THERE ARE TWO OR MORE
NUMBERS LISTED.

Outros Líquidos / Other Fluids


1 = Água corrente? / Plain water?
2 = Qualquer outro leite como em caixa, em pó ou fresco diretamente da vaca? / Any
other milk such as tinned, powdered, or fresh animal milk?
3 = Suco de fruta? / Fruit juice?
4 = Café ou chá preto? / Coffee or black tea?
5 = Qualquer outro líquido como água com açúcar, refrigerantes, ou caldo da sopa? /
Any other liquids such as sugar water, carbonated drinks, or soup broth?
6 = Remédio feito em casa? / Liquid or semi liquid traditional medicine?
7 = Outros_________(Por favor especifique) / Other ___(Please specify)
8 = Dois ou mais dos anteriores / Two or more of the above

241
O que fez você começar a dar alimentos sólidos para sua criança?
What made you decide to start giving solid foods to your child?
1 = não tinha leite suficiente / not enough breast milk
2 = o leite era fraco / weak (low quality) breast milk
3 = não tinha tempo para amamentar / not enough time to breastfeed
4 = bebê sempre chorava / baby always crying
5 = bebê procurando por comida / baby reaching for food
6 = bebê que é muito ativo / baby active
7 = tradição / tradition
8 = outros falaram para começar a dar comida / told to
9 = não sabe / don't know
10 = Necessidade de trabalhar / Necessity to work
11 = Outros/Others
(specificar/specify)_________________________________________

Depois de quanto tempo você começou a dar (NOME) comidas que não o leite do seu
peito?
How old was (NAME) when you started giving (NAME ) foods other than breast milk?
1 = um a dois meses / one to two months
2 = três a quatro meses / three to four months
3 = cinco meses / five months
4 = seis meses ou mais / six months or more
5 = não sabe / don't know
Você dá comidas especiais a sua criança quando (NOME) está doente? / Do you give
special foods to your child when (NAME) is ill?
0 = Não/No 1 = Sim/Yes (Especifique / Specify _________)
Você para de dar algum tipo de comida quando a criança (NOME) está doente? / Do
you withhold certain food from your child when (NAME) is ill?
0 = Não/No 1 = Sim/Yes (Especifique/Specify _________)
Você dá comidas especiais a sua criança depois que ela sarou? / Do you give special
foods to your child after illness?
0 = Não/No 1 = Sim/Yes (Especifique/Specify _________)
Vocë deu vermífugo para seu filho nos últimos 6 meses? Has your child been given
worm medicine in the past 6 months?
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know

Sua criança recebeu um suplemento de vitamina A, iodo, ferro ou qualquer outro


suplemento dietético (por exemplo, vitamina C)? / Has your child received a supplement
of vit. A, iodine, iron or any other dietary supplement (e.g., vitamin C)?
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Sua criança recebeu alguma vez suplemento de Vitamina A?/Has your child ever
received vitamin A supplements?
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Iodo (Mostrar iodo, sal ou cápsula de iodo) / Iodine (Show salt or iodine capsule)
0 = Não/No 1 = Sim/Yes 2 = Não sabe/Don’t know
Sua criança recebeu xarope de ferro nos últimos 06 meses?/Has your child received
Iron syrup in the past six months?
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Sua criança foi vacinada? / Has your child been immunized?
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Vacinação para tuberculose (BCG) / tuberculosis vacination (BCG)
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know

242
Vacinação para pólio / Polio vaccination
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Vacinação para Difiteria, coqueluxe e tétano (DPT) / Diptheria, Pertussis and Tetanus
(DPT) vacinnation
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Vacinação para Sarampo / Measles vaccination
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Vacinação para Rotavírus / Rotavirus vaccination
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Vacinação para Pneumococco / Pneumococco vaccination
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Vacinação para Influenza A (H1N1) / Influenza A (H1N1) vaccination
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Você sabe se sua criança tem anemia falciforme, traço falcêmico ou talassemia? / Do
you know if your child has sickle cell anemia, "traço falcêmico", or thalassemia?
0 = Não/No 1 = Sim/Yes 2 = Não sabe / Don’t know
Sua criança já teve dengue? / Has your child ever had dengue fever?
0 = Não/Yes 1 = Sim/Yes 2 = Não sabe / Don’t know
Sua criança já teve meningite? / Has your child ever had menigitis?
0 = Não/Yes 1 = Sim/Yes 2 = Não sabe / Don’t know
Sua criança alguma vez foi hospitalizada? / Has your child ever been hospitalized?
0 = Não/Yes 1 = Sim/Yes 2 = Não sabe / Don’t know
Se sim, quantas vezes sua criança foi hospitalizada? / If yes, how many times has your
child been hospitalized?
0 = uma ou duas vezes / one to two times
1 = três ou mais vezes / three or more times
2 = não sabe / don't know
Se sim, o qual a razão principal sua criança foi hospitalizado? / If yes, what was the
main reason your child was hospitalized?
0 = por causa de diarréia / for diarrhea
1 = por causa de uma infecção respiratória como gripe ou resfriado / for an upper
respiratory infection
2 = for another reason (2 = por outras razões)
3 = don't know (3 = não sabe)
Medidas de antropométricas da mãe
Anthropometric measurements of mother
Faça sempre duas medidas. Faça uma terceira medida se a diferença entre a primeira
medição e a segunda medição exceder a "diferença permitida"
Always make two measurements. Make a third measurement if the difference between the first observation
and the second observation exceeds the "allowable difference"
Diferença de peso (kg) permitida = 0.1kg
Weight (kg) Allowable difference = 0.1kg
Primeira medição / First observation
Segunda medição / Second observation
Terceira medição / Third Observation
Diferença de altura (cm) permitida = 0.7cm
Height (cm). Allowable difference = 0.7 cm
Primeira medição / First observation
Segunda medição / Second observation
Terceira medição / Third Observation

243
Appendix B Socio-economic scoring for household
characteristics or possessions

SES household characteristics or possessions Score

Number of people living and sleeping in the house


1-4 people 4
5-8 people 3
9-12 people 2
13 or more people 0
Mother's marital Status
Married or living with partner 4
Divorced 2
Single or widowed 0
Mother's educational Level
University 4
Tertiary 3
High school 2
Primary school 1
Illiterate 0
Mother's occupation
Small business owner 4
Regular employee 3
Domestic worker 2
Casual worker 1
Unemployed 0
House Ownership
Own 4
Rent 1
Housefloor type
Terrazo or ceramic 4
Concrete or cement 3
Wood planks or blocks 2
Earth or mud 1
Bedrooms in house
4-5 bedrooms 4
3 bedrooms 3
2 bedrooms 2
1 bedroom 1
No bedrooms 0
Boiled drinking water
Boiled 4
Not boiled 1

244
Sewage disposal
Closed 4
Open 1
Toilet facilities
Indoor 4
WC shared between several families 3
Pit latrine (private or communal) 2
No household facility 1
Electricity in the house
Available 4
Not available 0
House flooded in heavy rain
Not flooded 4
Flooded 1
Home goods
Motorcyle and / or automobile 8
Bicycle 4
Refrigerator 2
TV 1
Sum of home goods
15 points 4
10-14 points 3
4-9 points 2
1-3 points 1
0 points 0

245
Appendix C Differences by daycare group for specific feeding and health
characteristics of the participants (n=376)

City centre Peri-urban favela Peri-urban gated


p
N=79 N=202 N=95
n (%) n (%) n (%)

Age group (≥ 4 years) 49 (62.0) 126 (62.4) 58 (61.1) 0.989

Sex (male) 39 (49.4) 104 (51.5) 53 (55.8) 0.678

Breastfed ever 74 (93.7) 192 (95.1) 91 (95.8) 0.858


246

Introduction of solids < 6 mos 36 (45.6) 109 (54.0) 69 (72.6) 0.001

Deworming treatment within 6 mos 33 (41.8) 97 (48.0) 62 (65.3) 0.004

Iron syrup supplementation within 6 mos 16 (20.3) 37 (18.3) 17 (17.9) 0.921

Vitamin A supplementation ever 41 (51.9) 130 (64.4) 29 (30.5) < 0.001

Childhood vaccinations 66 (83.5) 182 (90.1) 67 (70.5) < 0.001


Appendix D Instruction for stool collection

Faeces collection for analysis of worm infestation

Please collect the faeces the night or morning before bringing the sample to the
daycare.
Instructions:
1. Place newspaper or plastic wrap under the seat at the back of the toilet to
catch the faeces only and to avoid mixing with urine.
2. Lower the toilet seat and allow child to pass faeces as usual.
3. Using the scoop attached to the lid of the container, place one scoop into
the container.
4. Flush the remaining stool and put soiled newspaper in a plastic bag to
throw away.
5. Do not place the faeces container in the refrigerator.
6. Bring immediately to the daycare.
PLEASE REMEMBER: NO URINE, NO WATER, ONLY FAECES IN THE
CONTAINER!

Coleta de fezes para análise de verminoses

Por favor, colete as fezes em casa à noite ou pela manhã e leve a amostra para a
creche.
Passo a passo de como fazer:
1. Coloque jornal ou plástico debaixo do assento do vaso sanitário, para
auxiliar na coleta das fezes.
2. Abaixe o assento do vaso sanitário e deixe a criança defecar como de
costume. Lembre-se de evitar misturar as fezes com urina. Se preferir,
coloque o jornal ou plástico no chão e deixe a criança defecar.
3. Com auxílio da paleta (que está fixada à tampa do recipiente), coloque um
pouco de fezes no pote coletor.
4. Jogue fora o jornal ou plástico com o restante das fezes.

247
5. Não coloque na geladeira o recipiente com as fezes.
6. Se as fezes forem coletadas à noite, leve para a creche no outro dia de
manhã. Se for coletada pela manhã, leve imediatamente para a creche.
LEMBRE-SE: COLOQUE APENAS AS FEZES NO COLETOR. NÃO
MISTURE NEM URINA E NEM ÁGUA!

248