Professional Documents
Culture Documents
Enfermedades de Transmisión Alimentaria - WHO PDF
Enfermedades de Transmisión Alimentaria - WHO PDF
FOODBORNE DISEASE
BURDEN EPIDEMIOLOGY
REFERENCE GROUP
2007-2015
WHO ESTIMATES OF
THE GLOBAL BURDEN
OF FOODBORNE DISEASES
FOODBORNE DISEASE
BURDEN EPIDEMIOLOGY
REFERENCE GROUP
2007-2015
WHO Library Cataloguing-in-Publication Data
WHO estimates of the global burden of foodborne diseases: foodborne disease burden
epidemiology reference group 2007-2015.
I.World Health Organization.
ISBN 978 92 4 156516 5
Subject headings are available from WHO institutional repository
CONTENTS
ACKNOWLEDGEMENTS VIII
FOREWORD IX
EXECUTIVE SUMMARY X
1 INTRODUCTION 1
1.1 Motivation: the importance of food safety ............................................................................................................. 3
1.2 The value of foodborne disease burden estimates ............................................................................................. 3
1.3 Purpose and audience.......................................................................................................................................................4
1.4 Scope .........................................................................................................................................................................................4
1.5 History and structure .........................................................................................................................................................4
1.6 Objectives................................................................................................................................................................................ 5
1.7 Other relevant burden of disease estimates .......................................................................................................... 6
1.8 Timeline: FERG Meetings ................................................................................................................................................. 7
1.8.1 Overview .............................................................................................................................................................................................7
1.8.2 Extracts from reports of major meetings ............................................................................................................................7
1.8.3 Strategic revisions .........................................................................................................................................................................10
1.8.4 Key decisions ...................................................................................................................................................................................10
1.8.5 Country-level involvement ..........................................................................................................................................................11
1.9 Task Force Meetings ..........................................................................................................................................................12
1.10 Participants...........................................................................................................................................................................12
1.11 Declarations of Interest ....................................................................................................................................................12
2 COMMISSIONED WORK 13
2.1 Enteric Diseases Task Force ...........................................................................................................................................15
2.1.1 Brucella spp. ..................................................................................................................................................................................... 15
2.1.2 Diarrhoeal disease ......................................................................................................................................................................... 15
2.1.3 Mycobacterium bovis................................................................................................................................................................... 15
2.1.4 Shiga toxin-producing Escherichia coli............................................................................................................................... 15
2.1.5 Norovirus .......................................................................................................................................................................................... 16
2.1.6 Invasive non-typhoidal Salmonella enterica ..................................................................................................................... 16
2.1.7 Listeria monocytogenes ............................................................................................................................................................. 16
2.2 Parasitic Diseases Task Force .......................................................................................................................................16
2.2.1 Taenia solium.................................................................................................................................................................................... 16
2.2.2 Trematodes (includes Echinostoma spp., Fasciolopsis buski, Heterophyes spp. and Metagonimus spp.) .. 16
2.2.3 Echinococcus multilocularis ..................................................................................................................................................... 17
2.2.4 Trichinella spp. ................................................................................................................................................................................. 17
2.2.5 Toxoplasma gondii ......................................................................................................................................................................... 17
2.3 Chemicals and Toxins Task Force ...............................................................................................................................17
2.3.1 Aflatoxins .......................................................................................................................................................................................... 17
2.3.2 Arsenic ................................................................................................................................................................................................ 17
2.3.3 Cassava cyanide ............................................................................................................................................................................. 17
2.3.4 Peanut Allergens ............................................................................................................................................................................ 18
2.3.5 Dioxins ................................................................................................................................................................................................. 18
2.4 Source Attribution Task Force......................................................................................................................................18
2.5 Computational Task Force .............................................................................................................................................18
2.6 Country Studies Task Force ...........................................................................................................................................19
2.7 Other relevant publications ...........................................................................................................................................19
VI
4 HAZARD-SPECIFIC METHODOLOGY 27
4.1 Hazard Selection ............................................................................................................................................................... 30
4.2 Enteric Hazards ...................................................................................................................................................................31
4.2.1 Estimating cases, sequelae and deaths for diarrhoeal diseases............................................................................ 32
4.2.2 Estimating cases and sequelae of, and deaths due to, extra-intestinal diseases .......................................... 33
4.3 Parasitic Hazards................................................................................................................................................................35
4.4 Chemicals and toxins ...................................................................................................................................................... 38
4.4.1 Cyanide in cassava ....................................................................................................................................................................... 38
4.4.2 Peanut allergen .............................................................................................................................................................................. 39
4.4.3 Aflatoxin ........................................................................................................................................................................................... 40
4.4.4 Dioxin................................................................................................................................................................................................. 40
4.5 Outcomes and disability weights ............................................................................................................................. 42
4.6 Attribution ............................................................................................................................................................................ 45
4.6.1 Identification of experts ............................................................................................................................................................45
4.6.2 Selection of experts .....................................................................................................................................................................46
4.6.3 Expert panels .................................................................................................................................................................................. 47
4.6.4 Analytical method ........................................................................................................................................................................ 47
4.6.5 Seed questions ............................................................................................................................................................................... 47
4.6.6 Target questions ............................................................................................................................................................................49
4.6.7 Data analysis ...................................................................................................................................................................................49
4.7 Computation .........................................................................................................................................................................51
4.7.1 Disease models and epidemiological data....................................................................................................................... 52
4.7.2 CTF database template ............................................................................................................................................................. 56
4.7.3 Imputation ........................................................................................................................................................................................ 56
4.7.4 Probabilistic burden assessment .......................................................................................................................................... 58
5 RESULTS 61
5.1 Attribution ............................................................................................................................................................................ 63
5.1.1 Expert performance........................................................................................................................................................................64
5.1.2 Pathway attribution results ........................................................................................................................................................ 65
5.2 DALY Estimates: Overview ............................................................................................................................................72
5.3 DALY Estimates: Enteric diseases ............................................................................................................................ 84
5.4 DALY Estimates: Parasites ............................................................................................................................................ 86
5.5 DALY Estimates: Chemicals ......................................................................................................................................... 89
6 DISCUSSION 93
6.1 Attribution ............................................................................................................................................................................ 98
6.2 Enteric Diseases ...............................................................................................................................................................102
6.3 Parasites .............................................................................................................................................................................. 104
6.4 Chemicals .............................................................................................................................................................................110
VII
8 CONCLUSION 123
8.1 Reflections on the WHO Initiative to Estimate the Global Burden of Foodborne Diseases....... 125
APPENDICES 129
Appendix 1. Formal Description of the Project and Participants ...................................................................... 131
Appendix 2. Subregions .......................................................................................................................................................143
Appendix 3. Preliminary hazards considered by each task force ....................................................................144
Appendix 4. Hazard-specific input parameter sources and methods ...........................................................145
Appendix 5. Disease models ............................................................................................................................................. 190
Appendix 6. Derivation of Disability Weights ............................................................................................................195
Appendix 7: Attribution – Expert Elicitation Results..............................................................................................199
Appendix 8. Data tables for individual hazard classes: enteric, parasitic, chemical ................................ 211
REFERENCES 223
GLOSSARY 251
ABBREVIATIONS 252
VIII
ACKNOWLEDGEMENTS
WHO gratefully acknowledges the financial and in-kind contributions from Canada,
Denmark, Germany, Ireland, Japan, Republic of Korea, Netherlands, United Kingdom of
Great Britain and Northern Ireland, United States of America and other donors. WHO
also acknowledges the invaluable contribution of the members of the Foodborne
Disease Burden Epidemiology Reference Group, the resource advisors, and their
affiliated organizations who supported this initiative. They are named in Appendix 1.
This report is a compilation of the main reports of the task forces of the Reference
Group. Numerous persons contributed to or read the manuscript and provided
comments, including several units in WHO headquarters and Regional Offices.
The text was then edited to introduce conformity in language and style.
IX
FOREWORD
Foodborne diseases have been an issue The objective of the initiative was not
for all societies since the beginning limited to providing estimates on the
of humanity. The types, severity global burden of foodborne diseases
and impacts of these illnesses have for a defined list of causative agents of
changed through the ages and are microbial, parasitic and chemical origin.
still diverse across regions, countries The initiative also aimed at strengthening
and communities. the capacity of countries to conduct
assessments of the burden of foodborne
Yet there are some challenges common
disease, and encouraging them to use
to all countries. Only a fraction of the
burden of foodborne disease estimates
people who become sick from food they
for cost-effectiveness analyses of
have eaten seek medical care. Only a
prevention, intervention and control
fraction of those cases are recognized
measures including implementation of
as having been caused by a hazard in
food safety standards in an effort to
food, treated accordingly, reported to
improve national food safety systems.
public health authorities and recorded in
official disease statistics. Certain chronic Six taskforces were established under
diseases, such as cancer, kidney or liver FERG, focusing on groups of hazards
failure, that result from contaminated or aspects of the methodology. These
food appear long after the ingestion of taskforces commissioned systematic
food and the causal link is never made reviews and other studies to provide
for each case. This points to some of the data from which to calculate the
the challenges inherent in measuring the burden estimates.
burden of foodborne diseases and the
This report is an outcome of a decade
toll they take on lives and economies.
of work by WHO, key partners and
Up to now, the global burden of illness a number of dedicated individuals.
and deaths caused by foodborne Some additional findings, which cannot
disease has never been quantified. In be integrated into this report, will be
order to fill this data vacuum, the World published and user-friendly online tools
Health Organization (WHO), together made available separately.
with its partners, launched in 2006 the
This report and related tools should
Initiative to Estimate the Global Burden
enable governments and other
of Foodborne Diseases. After an initial
stakeholders to draw public attention to
consultation, WHO in 2007 established a
this often under-estimated problem and
Foodborne Disease Burden Epidemiology
mobilize political will and resources to
Reference Group (FERG) to lead the
combat foodborne diseases.
initiative.
Kazuaki Miyagishima
Director
Department of Food Safety
and Zoonoses
WHO Estimates of the global burden of foodborne diseases
X
EXECUTIVE SUMMARY
significantly to the foodborne disease burden of disease, and pilot studies were
burden. The global burden of foodborne conducted in four countries (Albania,
diseases is considerable, with marked Japan, Thailand and Uganda). Data gaps
regional variations. The burden of were the major hurdle in estimating
foodborne diseases is borne by the foodborne disease burden in these
individuals of all ages, but particularly national studies, and the global and
by children under 5 years of age, and by regional estimates provided by FERG
persons living in low-income subregions offer an interim solution, until improved
of the world. surveillance and laboratory capacity is
developed.
These estimates are conservative; further
studies are needed to address the data Despite the data gaps and limitations of
gaps and limitations of this study. these initial estimates, it is apparent that
the global burden of foodborne disease
In addition to providing global and
is considerable, and affects individuals
regional estimates, the Initiative sought
of all ages, but particularly children
to promote actions at a national level.
under 5 years of age and persons living
This involved capacity building through
in low-income subregions of the world.
national foodborne disease burden
All stakeholders can contribute to
studies, and encouraging the use of
improvements in food safety throughout
burden information in setting evidence-
the food chain by incorporating these
informed policies. A suite of tools and
estimates into policy development at
resources were created to facilitate
national, regional and international levels.
national studies of the foodborne
WHO Estimates of the global burden of foodborne diseases
1
INTRODUCTION
1
WHO Estimates of the global burden of foodborne diseases
3
INTRODUCTION
INTRODUCTION
economic development, food safety 1.4 Scope
remains marginalized. A major obstacle
This report covers:
to adequately addressing food safety
concerns is the lack of accurate data on f history of the project;
the full extent and cost of foodborne f participants;
diseases, which would enable policy- f scientific work commissioned by
makers to set public health priorities and the project;
allocate resources. Epidemiological data f overview of approach to estimating
on foodborne diseases remain scarce, burden of foodborne disease;
particularly in the developing world. Even f methods, results, discussion, using a
the most visible foodborne outbreaks hazards-based approach;
often go unrecognized, unreported or f outputs, implications and context of
uninvestigated, and may only be visible results; and
if connected to major public health or f future plans.
economic impact. Precise information
on the burden of FBD is needed to 1.5 History and structure
adequately inform policy-makers and
In September 2006, FOS launched the
allocate appropriate resources for food
Initiative to Estimate the Global Burden
safety control and intervention efforts.
of Foodborne Diseases at an international
In order to fill this data vacuum, the consultation attended by over 50
World Health Organization (WHO) international experts. This consultation
Department of Food Safety, Zoonoses provided the strategic framework
and Foodborne Diseases (FOS) together for the assessment of FBD burden,
with its partners launched the Initiative and mandated WHO to establish a
to Estimate the Global Burden of Foodborne Disease Burden Epidemiology
Foodborne Diseases. The primary goal of Reference Group (FERG) to engage in:
the Initiative is: f assembling, appraising and reporting on
To enable policy-makers and other currently existing burden of foodborne
stakeholders to set appropriate, disease estimates;
evidence-based priorities in the area of f conducting epidemiological reviews for
food safety. mortality, morbidity and disability in
each of the major FBDs;
f providing models for the estimation of
1.3 Purpose and audience
FBD burden where data are lacking;
This report is a supplement to the f developing cause and source attribution
scientific papers published in journals models to estimate the proportion of
from the Public Library of Science diseases that are foodborne, and
(PLOS), which cover the estimates f developing user-friendly tools for
generated by the WHO Initiative to burden of FBD studies at country level.
Estimate the Global Burden of Foodborne
Following a public call for advisers in
Diseases. In addition to collating the
the scientific press the WHO Director-
results, this report is intended to provide
General appointed the FERG members
background and context on the project
who met for the first time in November
itself, as well as examining particular
2007. This multi-disciplinary meeting
scientific issues in more detail. As such,
commenced with a stakeholder
it provides a comprehensive source of
consultation that informed the technical
information on the Initiative.
discussions of FERG. The meeting
WHO Estimates of the global burden of foodborne diseases
5
saw the establishment of the FERG and external resource and technical
Core or Steering Group (coordinating advisers who are invited on an ad hoc
and overseeing the burden work) as basis to provide specific expertise.
well as several thematic Task Forces
(TFs) to advance the work in specific
1.6 Objectives
areas, including:
The first report from the Initiative,
f Enteric Diseases Task Force (EDTF);
published in 2008, described the
f Parasitic Diseases Task Force (PDTF);
following objectives1:
and,
f Chemicals and Toxins Task f To strengthen the capacity of countries
Force (CTTF). in conducting burden of foodborne
disease assessments and to increase
Subsequently, three additional Task
the number of countries who have
Forces were established to address the
undertaken a burden of foodborne
following topics:
disease study.
f Source Attribution Task Force (SATF) f To provide estimates on the global
(established 2008); burden of foodborne diseases
f Country Studies Task Force (CSTF) according to age, sex and regions for
(established 2009), with a sub-group, a defined list of causative agents of
the Knowledge Translation and Policy microbial, parasitic and chemical origin.
Group (KTPG) (established 2010); f To increase awareness and
f Computational Task Force (CTF) commitment among Member States
(established 2012). for the implementation of food
As shown in Figure 1, FERG consists of a safety standards.
f To encourage countries to use burden
Core (or Steering) Group to coordinate
and oversee the scientific work, Thematic of foodborne disease estimates for
TFs advancing the work in specific areas; cost-effective analyses of prevention,
intervention and control measures.
1
http://www.who.int/foodsafety/foodborne_
disease/Summary_Doc.pdf?ua=1 Accessed 9
July 2014
Introduction
6
INTRODUCTION
Figure 1. Structure of the initiative to estimate the global burden of foodborne diseases
To meet these goals and objectives, the f Global Burden of Disease 2010
Initiative took two approaches. (GBD2010) study, undertaken by
the Institute of Health, Metrics and
f A Foodborne Disease Burden
Evaluation (IHME)2
Epidemiology Reference Group (FERG)
f Mortality and Burden of Disease Unit
was established to assemble, appraise
of WHO3
and report on burden of foodborne
f Estimated Cancer Incidence, Mortality
disease estimates.
f In-depth country studies to supplement
and Prevalence 2012, published by the
International Agency for Research on
the work of FERG and enable countries
Cancer (GLOBOCAN)4
to conduct their own burden of
disease studies. Throughout the course of the burden
of foodborne disease project, FERG
1.7 Other relevant burden of di- communicated with these groups, to
sease estimates share data and promote consistency of
the estimates.
Estimates for the burden of diseases
considered to be at least partially
foodborne have been published by a
2
number of research groups. The most http://www.healthdata.org/gbd Accessed 24
September 2014
comprehensive estimates are those 3
http://www.who.int/topics/global_burden_of_
published by the following: disease/en/ Accessed 24 September 2014
4
http://globocan.iarc.fr/Default.aspx Accessed 24
September 2014
WHO Estimates of the global burden of foodborne diseases
7
INTRODUCTION
Following a public call for advisers f agreement on the terms of reference
in the scientific press, the Director- of the new FERG Country Studies Task
General of WHO appointed the FERG Force (CSTF) in 2009;
members, who met for the first time in f formal evaluation of the activities,
November 2007. This multi-disciplinary processes and outputs of the first year
meeting commenced with a stakeholder of FERG activities; and
consultation that informed the technical f A major, multisectoral stakeholder
discussions of FERG. The meeting saw meeting, which provided valuable
the establishment of the FERG Core input and recommendations to WHO
or Steering Group (coordinating and in the areas of technical reviews,
overseeing the burden of the work), as communication and policy.
well as several thematic Task Forces
26–30 October 2009 – FERG 3, Geneva
(TFs) to advance the work in specific
(plus Stakeholder Meeting)
areas, including:
The Third Foodborne Diseases
f parasitic diseases;
Stakeholder Meeting brought together
f chemicals and toxins; and
international representatives from the
f enteric diseases.
various constituencies and sectors with
In their respective areas, the TFs an interest in ensuring food safety, be
provided: (1) priority lists of causative it through decision-making, research,
agents for which burden assessments production, consumption or advocacy.
should be conducted; (2) developed They included: WHO Member States;
concrete and very detailed work plans bilateral and multilateral donors; non-
to commission the individual burden governmental organizations (NGOs);
work; and (3) agreed on the logistic and consumer groups; industry; and public
technical steps to be taken by FERG over and scientific media. The purpose of the
the next year. meeting was to enable stakeholders to:
17–21 November 2008 – FERG 2, Geneva f actively engage with the Foodborne
(plus Stakeholder Meeting) Disease Burden Epidemiology
Reference Group (FERG) and
The second formal meeting of FERG in
its research;
November 2008 (FERG 2) highlighted
f open new channels for multisectoral
the progress made during the Initiative’s
cooperation; and
first year, which included:
f provide direct input into discussions
f an appraisal of the methods, and about how to bridge the gap between
preliminary results of ten systematic evidence and policy.
reviews commissioned in the areas
8–12 November 2010 – FERG 4, Geneva
of enteric, parasitic and chemical
causes of foodborne diseases, as well Continuing on the path taken during the
as mortality; previous FERG meeting, a large number
f the development of detailed new of new foodborne disease morbidity,
work plans for all FERG TFs for 2009, mortality and burden estimates were
including new burden work to be presented and discussed at FERG 4:
commissioned; f the global burden of diarrhoeal
f establishment of the FERG
diseases;
Source Attribution Task Force f the global burden of foodborne
(SATF) and execution of its
trematodiasis;
technical recommendations;
WHO Estimates of the global burden of foodborne diseases
9
INTRODUCTION
The advocacy efforts of the coordinator that high level senior management at
of the Initiative were praised and WHO reiterate their support for the
considered to be very effective. Initiative through providing the necessary
resources to ensure the success of the
The main challenge to the Initiative
Initiative and the considerable investment
was how to deal with the expansion to
that had been made.
the scope of the Initiative. The need to
plan to collect primary data, overcome 7–10 November 2011– Strategy Meeting
methodological challenges, integrate and Commencement of Country Studies,
knowledge translation and respond Durrës, Albania
appropriately to the 63rd World Health
Assembly (WHA) resolution on food 1.8.3 Strategic revisions
safety– all these were part of the In view of the increased complexity
expanded scope of the Initiative. FERG of the WHO FERG Initiative, as well
experts were in agreement that the as the changed environment in which
expansion of the scope was necessary the Initiative was operating, the WHO
and appropriate. Because quality must be Secretariat convened a meeting with the
maintained, adjustments must be made objectives of:
to timelines and resources. Timelines
f updating the Initiative’s strategic
can be reviewed, but FERG experts
framework, its milestones and timelines;
and stakeholders stated that there was
f redefining the technical scope of the
a limitation on timeline extension due
Initiative, including the selection of
to the risk of loss of momentum, and
priority areas for foodborne disease
there was also the need to fulfil Member
burden estimation;
State and donor expectations for initial
f identifying key activities and resource
estimation of the global burden of
needs for implementation; and
foodborne diseases. Therefore, increasing
f updating FERG processes, roles
the Initiative’s human and financial
and responsibilities.
resources was the most appropriate
change that could be made.
1.8.4 Key decisions
FERG experts were concerned about f Scope of technical work: The thematic
a major threat to the Initiative, namely TF chairs, in consultation with their
the dependence of the Initiative and TFs, established a shortened list of
its success on such a small number of pathogens and hazards for which
key personnel in the WHO Secretariat. they intended to deliver incidence and
These few key people were considered mortality estimates by the end of 2012.
excellent in terms of technical expertise, f Methodological decisions: A
enthusiasm, energy, dedication and range of important technical and
motivation, and much of the success methodological issues linked to the
so far was ascribed to these qualities. estimation of foodborne disease
FERG experts were concerned that if burden were discussed at the meeting
there were any changes to personnel, the in Albania, and actions agreed upon
Initiative would be very vulnerable and in order to ensure accuracy, utility and
could fail. They were concerned about compatibility with other existing health
sustainability and lack of a ‘safety net’, metric indicatives.
and therefore requested an expanded f New FERG Computational Task
team at the Secretariat, with more of the Force: Continuing in this vein, a
existing skills. FERG experts requested new Computational Task Force
WHO Estimates of the global burden of foodborne diseases
11
(CTF) to work on the mathematical the pilot studies and finishing the country
modelling to calculate DALYs would tools. Each TF had outlined its priority
be established. The TF was currently activities for the coming year and WHO
set to be operational by the end of would use these to solicit the funding
February 2012. required to complete the FERG project.
f Source attribution expert elicitation:
The hazard TFs– EDTF, PDTF and CTTF–
An expert elicitation would be
completed the technical review of the
conducted in 2012 to determine what
systematic reviews; reviewed and revised
proportion of the burden of each
the final outcome trees; and made plans
hazard was foodborne, and which
for completion for each hazard.
were the major foods associated
with transmission. A list of hazards SATF finalized the expert elicitation
that would be included in the expert protocol for: chemicals and toxins
elicitation was established. (inorganic arsenic, lead, cadmium
and dioxins); for parasitic diseases
1.8.5 Country-level involvement
(Entamoeba histolytica, Cryptosporidium
The ‘kick-off’ meeting of the FERG spp., Giardia spp., Echinococcus
pilot country studies marked a major granulosus, Toxoplasma gondii,
milestone for the work of the Initiative Echinococcus multilocularis and Ascaris
in fostering national studies of the spp.); and for enteric diseases (diarrhoeal
burden of foodborne disease. For the diseases [non-typhoidal Salmonella
first time, representatives of the FERG spp., Campylobacter spp., Shiga-toxin
pilot countries met to present progress producing, enteropathogenic and
on the implementation of a national enterotoxigenic E. coli, norovirus, Shigella
foodborne disease burden study. They spp., Vibrio cholerae], typhoid, brucellosis
also learnt about study tools that FERG and hepatitis A).
had developed, as well as the future
technical support that would be provided The methodology and elicitation
by FERG. instrument were agreed with each of the
hazard TFs. This expert elicitation would
The pilot countries during the kick- be the first time that the methodology
off meeting: had been applied at a global level for
f drafted pilot country study work plans food safety and would involve disease
outlining the way forward; experts from all six WHO regions. The
f provided recommendations and input logistics of such an enormous task were
to align FERG procedures and tools also mapped out and agreed during
for national foodborne disease burden the meeting.
estimation and food safety policy CTF (established October 2012) was
situation analyses specific to country able to agree on the disease models
requirements; and for the majority of the pathogens, as
f delivered feedback and agreed on
well as meeting individually with each
processes to communicate between TF to advance the DALY calculations.
participating countries, and between The database was revised, methods
the countries and FERG Secretariat. for imputation of missing data were
8–12 April 2013 – FERG 5, Geneva advanced, and disability weights (DWs)
were mapped to all outcomes.
There was a very clear path towards the
end goal of publishing the estimates of CSTF and KTPG agreed the aims,
burden of foodborne disease, completing objectives and outline for the joint
Introduction
12
INTRODUCTION
country study workshop, initiated the CTF
development of the communications 2 - 4 October 2012 – Establishment
strategy for the global and regional FERG meeting, Antwerp, Belgium*
results, and reviewed the situational April 2013 – Sunday pre-meeting at
analysis document and the outcome of FERG 5 Geneva, Switzerland
the commissioned work. 2 August 2013 – Data imputation
23–25 June 2014– Strategy Meeting, meeting. RIVM, Bilthoven,
Copenhagen The Netherlands
31 January 2014 – DALY calculation &
This working meeting involved detailed Disability Weights meeting, Brussels
discussions of burden estimation across February 2014 – Data imputation
all the TFs. Progress was enhanced meeting, Antwerp, Belgium
greatly by the attendance by Dr Colin
Mathers, the Coordinator of the Mortality CSTF
and Burden of Disease Unit in the 10 - 12 June 2009 – Rome, Italy*
Health System and Innovation Cluster at 18 - 20 March 2010 – Atlanta, USA*
WHO, Geneva. This enabled the FERG 7 - 10 November 2011 – Kick off
estimation approaches to be harmonized meeting (Albania, Japan, Thailand),
with those used by the WHO unit. Durrës, Albania
4 - 6 March 2012 – Kick off meeting
(Uganda), Kampala, Uganda
1.9 Task Force Meetings
Only face-to-face meetings are listed. In 1.10 Participants
addition to these meetings, numerous
See Appendix 1.
teleconferences were held by each TF.
For the meetings marked*, finalized or
draft meeting reports are available. 1.11 Declarations of Interest
EDTF All experts and resource advisers
7 - 9 June 2009 – Rome* invited to participate in FERG meetings
14 - 18 July 2010 – Tunis, Tunisia completed beforehand the WHO
standard form for Declaration of
CTTF Interests. At the start of each meeting, all
14 - 16 July 2009 – Geneva* participants were asked to confirm their
14 - 18 July 2010 – Tunis, Tunisia interests, and to provide any additional
PDTF information relevant to the subject
matter of the meeting. All declared
7 - 9 June 2009 – Rome*
interests were assessed by the WHO
14 - 18 July 2010 – Tunis, Tunisia
Secretariat to ensure the neutrality and
SATF unbiasedness of the work.
28 - 30 April 2008 – Kuala
Lumpur, Malaysia*
20 - 22 April 2010 – Atlanta, USA*
14 - 18 July 2010 – Tunis, Tunisia
WHO Estimates of the global burden of foodborne diseases
13
2
COMMISSIONED
WORK
WHO Estimates of the global burden of foodborne diseases
15
COMMISSIONED WORK
Each TF commissioned specific pieces f Fischer Walker, C.L., & Black, R.E. 2010.
of work to provide scientific evidence Diarrhoea morbidity and mortality
on which to base estimates. Most of in older children, adolescents, and
these were systematic reviews, either of adults. Epidemiology and Infection,
available data on diseases, or reviews 138(9): 1215–1226. Available at http://
of methodology. The majority of journals.cambridge.org/action/
commissioned work resulted in published displayAbstract?fromPage=online&aid=
papers, as listed below. Some of these 7849267&fileId=S0950268810000592
publications were part funded by FERG, Accessed 2015-10-16.
while others were generated as “in kind” f Pires, S.M., Fischer Walker, C.L., Lanata,
contributions by the authors. C.F., Devleesschauwer, B., Hall, A.,
Kirk, M.D., Duarte, A.S.R., Black, R.E.,
& Angulo, F.J. Aetiology-specific
2.1 Enteric Diseases Task Force
estimates of the global and regional
The EDTF commissioned the following incidence and mortality of diarrhoeal
systematic reviews: diseases commonly transmitted
through food. PLOS ONE, vol 10, iss 12,
2.1.1 Brucella spp. DOI: 10.1371/journal.pone.0142927
f Dean, A.S., Crump, L., Greter, H.,
2.1.3 Mycobacterium bovis
Hattendorf, J., Schelling, E. & Zinsstag,
f Muller, B., Durr, S., Alonso, S.,
J. 2012. Clinical manifestations of human
brucellosis: a systematic review and Hattendorf, J., Laisse, C.J., Parsons, S.D.,
meta-analysis. PLOS Neglected Tropical van Helden, P.D. & Zinsstag, J. 2013.
Diseases, 6(12): Art. e1929. Available Zoonotic Mycobacterium bovis-induced
at http://www.plosntds.org/article/ tuberculosis in humans. Emerging
info%3Adoi%2F10.1371%2Fjournal. Infectious Diseases, 19(6): 899–908.
pntd.0001929 Accessed 2015-10-16. Available at: http://wwwnc.cdc.gov/eid/
f Dean, A.S., Crump, L., Greter, H.,
article/19/6/12-0543_article
Schelling, E. & Zinsstag, J. 2012. f Durr, S., Muller, B., Alonso, S.,
Global burden of human brucellosis: Hattendorf, J., Laisse, C.J., van Helden,
a systematic review of disease P.D. & Zinsstag, J. 2013. Differences
frequency. PLOS Neglected Tropical in primary sites of infection between
Diseases, 6(10): Art e1865. Available zoonotic and human tuberculosis:
at http://www.plosntds.org/article/ results from a worldwide systematic
info%3Adoi%2F10.1371%2Fjournal. review. PLOS Neglected Tropical
pntd.0001865 Accessed 2015-10-16. Diseases, 7(8): Art e2399. Available
2.1.2 Diarrhoeal disease at http://www.plosntds.org/article/
info%3Adoi%2F10.1371%2Fjournal.
f Fischer Walker, C.L., Sack, D. & Black,
pntd.0002399 Accessed 2015-10-16.
R.E. 2010. aetiology of diarrhoea
in older children, adolescents and 2.1.4 Shiga toxin-producing
adults: a systematic review. PLOS Escherichia coli
Neglected Tropical Diseases, f Majowicz, S.E., Scallan, E., Jones-Bitton,
4(8): Art e768. Available at A., Sargeant, J.M., Stapleton, J., Angulo,
http://www.plosntds.org/article/ F.J., Yeung, D.H. & Kirk, M.D. 2014
info%3Adoi%2F10.1371%2Fjournal. Global incidence of human Shiga toxin-
pntd.0000768 Accessed 2015-10-16. producing Escherichia coli infections
and deaths: a systematic review and
Commissioned work
16
COMMISSIONED
Accessed 2015-10-17. article/pii/S1473309914708709
WORK
Accessed 2015-10-17.
2.1.5 Norovirus
f Ahmed, S.M., Hall, A.J., Robinson, A.E., 2.2 Parasitic Diseases Task Force
Verhoef, L., Premkumar, P., Parashar, PDTF commissioned the following
U.D., Koopmans, M. & Lopman, B.A. systematic reviews:
2014. Global prevalence of norovirus in
cases of gastroenteritis: a systematic 2.2.1 Taenia solium
review and meta-analysis. Lancet f Carabin, H., Ndimubanzi, P.C., Budke,
Infectious Diseases, 14(8): 725–
C.M., Nguyen, H., Qian, Y., Cowan, L.D.,
730. Available at http://www.
Stoner, J.A., Rainwater, E. & Dickey, M.
sciencedirect.com/science/article/pii/
2011. Clinical manifestations associated
S1473309914707674 Accessed
with neurocysticercosis: a systematic
2015-10-17.
review. PLOS Neglected Tropical
f Verhoef, L., Hewitt, J., Barclay, L.,
Diseases, 5(5): Art e1152. Available
Ahmed, S.M., Lake, R., Hall, A.J.,
at http://www.plosntds.org/article/
Lopman, B., Kroneman, A., Vennema, H.,
info%3Adoi%2F10.1371%2Fjournal.
Vinje, J. & Koopmans, M. 2015. Norovirus
pntd.0001152 Accessed 2015-10-17.
genotype profiles associated with f Ndimubanzi, P.C., Carabin, H., Budke,
foodborne transmission, 1999– 2012.
C.M., Nguyen, H., Qian, Y.J., Rainwater,
Emerging Infectious Diseases,
E., Dickey, M., Reynolds, S. & Stoner,
45: 95– 99. Available at: http://wwwnc.
J.A. 2010. A systematic review of the
cdc.gov/eid/article/21/4/14-1073_article
frequency of neurocysticercosis with
2.1.6 Invasive non-typhoidal a focus on people with epilepsy. PLOS
Salmonella enterica Neglected Tropical Diseases, 4(11): Art
e870. Available at http://www.plosntds.
f Ao, T.T., Feasey, N.A., Gordon, M.A.,
org/article/info:doi/10.1371/journal.
Keddy, K.H., Angulo, F.J. & Crump, pntd.0000870 Accessed 2015-10-17.
J.A. 2015. Global burden of invasive
non-typhoidal Salmonella disease, 2.2.2 Trematodes (includes
2010. Emerging Infectious Diseases, Echinostoma spp., Fasciolopsis
21(6): 941– 949. buski, Heterophyes spp. and
f Crump, J.A. & Kirk, M.D. Estimating
Metagonimus spp.)
the burden of febrile illnesses. PLOS f Furst, T., Keiser, J. & Utzinger, J. 2012.
Neglected Tropical Diseases, (in press).
Global burden of human food-borne
2.1.7 Listeria monocytogenes trematodiasis: a systematic review
and meta-analysis. Lancet Infectious
f Maertens de Noordhout, C.,
Diseases, 12(3): 210– 221. Available at
Devleesschauwer, B., Angulo, F.J., http://www.thelancet.com/journals/
Verbeke, G., Haagsma, J., Kirk, M., laninf/article/PIIS1473– 3099(11)70294–
Havelaar, A. & Speybroeck, N. 2014. The 8/fulltext Accessed 2015-10-17.
global burden of listeriosis: a systematic
WHO Estimates of the global burden of foodborne diseases
17
COMMISSIONED
Systematic literature review. European f Aspinall, WP., Cooke, RM., Havelaar,
WORK
Journal of Food Research and Review, AH., Hoffman, S., Hald, T. 2015. Science-
2(2): 46– 48. Available at http:// based global attribution of foodborne
www.rivm.nl/en/Library/Scientific/ diseases: Findings of WHO expert
Reports/2012/april/Parameters_ elcitiation. PLOS ONE. (in press).
needed_to_estimate_the_global_
burden_of_peanut_allergy_Systematic_ 2.5 Computational Task Force
literature_review Accessed 2015-10-17.
f McDonald, S.A., Devleesschauwer, B.,
2.3.5 Dioxins Speybroeck, N., Hens, N., Praet, N.,
f Zeilmaker, M.J., Devleesschauwer, B., Torgerson, P.R., Havelaar, A.H., Wu, F.,
Mengelers, M.J.B., Hoekstra, J., Brandon, Tremblay, M., Amene, E.W. & Döpfer, D.
E.F.A. & Bokkers, B.G.H. The disease 2015. Data-driven methods for imputing
burden of dioxins: A global perspective. national-level incidence in global
RIVM Report National Institute for burden of disease studies. Bulletin
Public Health and the Environment of the World Health Organization,
(RIVM), Netherlands. 93(4): 228– 236 doi: http://dx.doi.
org/10.2471/BLT.14.139972
2.4 Source Attribution Task Force The following two papers were not
SATF commissioned the commissioned by the Computational Task
following papers: Force, but several of the authors were TF
members, and the papers are relevant to
f Pires, S.M. 2013. Assessing the
the Initiative estimates.
applicability of currently available
methods for attributing foodborne f Devleesschauwer, B., Havelaar, A.H.,
disease to sources, including food Maertens de Noordhout, C., Haagsma,
and food commodities. Foodborne J.A., Praet, N., Dorny, P., Duchateau,
Pathogens and Disease, 10(3): 206– 213. L., Torgerson, P.R., Van Oyen, H. &
f Pires, S.M., Evers, E.G., van Pelt, W., Speybroeck, N. 2014. Calculating
Ayers, T., Scallan, E., Angulo, F.J., disability-adjusted life years to quantify
Havelaar, A. & Hald, T. and the Med- burden of disease. International Journal
Vet-Net Workpackage 28 team. 2009. of Public Health, 59(3): 565– 569.
Attributing the human disease burden f Devleesschauwer, B., Havelaar, A.H.,
of foodborne infections to specific Maertens de Noordhout, C., Haagsma,
sources. Foodborne Pathogens and J.A., Praet, N., Dorny, P., Duchateau,
Disease, 6(4): 417– 424. L., Torgerson, P.R., Van Oyen, H. &
f Hoffman, S., Aspinall, W., Cooke, R., Speybroeck, N. 2014. DALY calculation
Cawthorne, A., Corrigan, T., Havelaar, in practice: a stepwise approach.
A., Gibb, H., Torgerson, P., Kirk, M., International Journal of Public Health,
Angulo, F, Lake R, Speybroeck N, 59(3): 571– 574.
Devleesschauwer B, Hald T. 2015
Perspective: Research synthesis
WHO Estimates of the global burden of foodborne diseases
19
COMMISSIONED
WORK
WHO Estimates of the global burden of foodborne diseases
21
3
OVERVIEW
OF THE
SCIENTIFIC
APPROACH
WHO Estimates of the global burden of foodborne diseases
23
OF SCIENTIFIC
Reference year
APPROACH
OVERVIEW
the point of infection (or primary health
effect from exposure) as starting point The reference year for the calculation of
for the calculations; and (3) is consistent absolute numbers was 2010.
with the estimation of YLLs, which by Attribution
definition follows an incidence-based
The choice of a method to attribute
approach, as mortality can be seen as
a proportion of disease incidence
the incidence of death [3]. Nevertheless,
to foodborne transmission was a
the prevalence- and incidence-based
major decision for the project. The
approaches yield similar overall results if
rationale for choosing a global expert
the epidemiology of disabilities and the
elicitation process was developed
population age-structure are constant
after consideration of alternatives, as
over time [2]. However, burden estimates
described below.
for specific age groups will always differ
between the prevalence- and incidence- Estimating the burden of FBD is
based approaches, because the former complicated because most of the hazards
assigns the burden to the age at which causing foodborne disease are not
the burden is experienced, while the transmitted solely by food. The relative
latter assigns the burden to the age of impact of each route differs depending
disease onset [4]. on the epidemiology of the disease
causing microorganism (bacteria, virus
Using the incidence for the burden
or parasite) or chemical hazards. Other
estimations is important for diseases
factors such as the geographical region,
having a long period between exposure
season and food consumption patterns
and appearance of clinical signs. An
also influence the role of different
incidence-based approach for the
exposures routes [7, 8]. The estimation
burden estimations fits better with
of the burden of FBD, therefore, requires
a hazard-based approach. However,
incidence figures are not always 1
http://www.who.int/foodsafety/chem/gems/en/
available. For example, in the case of index1.html Accessed 23 July 2014
peanut [Arachis hypogaea] allergy, only 2
The subregions are defined on the on the basis of
child and adult mortality, as described by Ezzati
prevalence figures are available. When et al. [5] Stratum A = very low child and adult
only prevalence figures are available, mortality; Stratum B = low child mortality and very
incidence can be estimated based on the low adult mortality; Stratum C = low child mortality
and high adult mortality; Stratum D = high child and
prevalence figures and on the duration of
adult mortality; and Stratum E = high child mortality
the disease. and very high adult mortality. The use of the term
‘subregion’ here and throughout the text does
not identify an official grouping of WHO Member
States, and the “subregions” are not related to the
six official WHO regions.
WHO Estimates of the global burden of foodborne diseases
25
OF SCIENTIFIC
APPROACH
OVERVIEW
WHO Estimates of the global burden of foodborne diseases
27
4
HAZARD-
SPECIFIC
METHODOLOGY
WHO Estimates of the global burden of foodborne diseases
29
HAZARD-SPECIFIC METHODOLOGY
PLOS ONE, vol 10, iss 12 DOI: 10.1371/ f Likely magnitude of foodborne
journal.pone.0142498 component of burden of disease.
Lake, R., Devleesschauwer, B., Nasinyama, Each of the three papers from the
G., Havelaar, A.H., Kuchenmüller, T., hazard-based TFs (EDTF, PDTF and
Haagsma, J.A., Jensen, H., Jessani, N., CTTF) includes supplementary material
Maertens de Noordhout, C., Angulo, F.J., discussing the sources and methodology
Ehiri, J., Molla, L., Agaba, F., Aungkulanon, for the parameters used to estimate:
S., Kumagai, Y. & Speybroeck, N. 2015. incidence, clinical outcomes, duration,
National studies as a component of the DW, mortality, age and sex distribution.
World Health Organization initiative to The full details have been combined
estimate the global and regional burden in Appendix 4. The material below
of foodborne disease. PLOS ONE, vol 10, explains the rationale for the sources
iss 12, DOI: 10.1371/journal.pone.0140319 and methods.
Burden of foodborne disease estimates
4.1 Hazard Selection were prepared for the 40 foodborne
hazards causing 41 diseases shown in
At the first meeting after the
HAZARD SPECIFIC
METHODOLOGY
Figure 2.
establishment of FERG, each hazard-
based TF compiled a comprehensive The following methodology section
universal list of foodborne hazards that describes the inputs and processes used
could be addressed (see Appendix 3). to generate DALY estimates. This material
Pragmatic decisions were then made is broadly structured as follows:
about specific hazards for further work, f estimation of incidence (or population
based on the knowledge of TF members attributable fraction);
and applying the following criteria: f health states and disability weights;
f Availability of data to estimate f attribution of foodborne
Figure 2. Hazards for which burden of foodborne disease estimates were prepared by FERG,
grouped according to TF. Hazards in grey boxes were addressed by individual TFs but were not
included in the global overview. Hazards in blue boxes are pending.
Note that salmonellosis and invasive salmonellosis are counted as a single hazard causing two diseases.
Notes: (1) 61 EUR and other subregion A (low mortality) countries only. (2) Includes Guillain-Barré Syndrome cases and deaths.
(3) 61 EUR and other subregion A (low mortality) countries only, excluding WPR countries. (4) Includes selected species of the
families Echinostomatidae, Fasciolidae, Gymnophallidae, Heterophyidae, Nanophyetidae, Neodiplostomidae and Plagiorchiidae
(depending on data availability). (5) Only the burden for AMR A, EUR A and WPR A was assessed. (6) Separate estimates for
congenital and acquired toxoplasmosis.
infection: invasive infections due to For this approach, the WHO Child
S. enterica serotype Typhi Health Epidemiology Reference Group
(S. Typhi); invasive infections due to (CHERG) method was modified to
S. serotype Paratyphi A (S. Paratyphi A); estimate diarrhoeal incidence and
invasive infections due to non-typhoidal mortality for all age groups [50]. First,
S. enterica (iNTS); and diarrhoeal disease the overall incidence of diarrhoea from all
due to non-typhoidal S. enterica. causes (i.e. the “envelope” of diarrhoeal
incidence) was estimated for 2010
Diarrhoea is a dominant feature for
by combining estimates of diarrhoeal
14 of these diseases – ten caused by
incidence for children <5 years of age
bacteria, three by protozoa, and one
and persons *5 years of age [51, 52].
by a virus. One or more extra-intestinal
The overall diarrhoeal mortality (i.e. the
manifestations, including bacteraemia,
envelope for diarrhoeal deaths) derived
hepatitis and meningitis, are the
by WHO for 2010 was used.3
dominant feature for the other eight
An aetiological proportion for each
diseases – seven caused by bacteria and
disease by region was derived from
one caused by a virus.
systematic reviews of stool sample
HAZARD SPECIFIC
METHODOLOGY
isolation or detection proportions from
4.2.1 Estimating cases, sequelae and
inpatient, outpatient and community-
deaths for diarrhoeal diseases
based studies of persons with diarrhoea.
For diarrhoeal diseases caused by Following the CHERG standard approach,
Campylobacter spp., Cryptosporidium developed because there is limited
spp., Entamoeba histolytica, ETEC, information on pathogens among
EPEC, Giardia spp., norovirus, non- people who have died, it was assumed
typhoidal Salmonella spp. and Shigella that the distribution of pathogens
spp., because national estimates of observed among inpatients hospitalized
foodborne diseases were only available with severe diarrhoea represented the
from a limited number of countries, two pathogen prevalence among diarrhoeal
approaches were used depending on the deaths [50]. To derive aetiological
level of development of the country. The proportions for children <5 years of age,
approaches have been described by a it was assumed that the distribution of
key accompanying publication [40]. pathogens in outpatient and community
The first approach, based on national studies represented the pathogen
estimates of the incidence of foodborne prevalence among diarrhoeal episodes
diseases, was applied to the 61 countries for those who did not die. The same
in low-mortality (EUR and other assumption was made for persons
subregion A) countries [41– 49]. For *5 years of age but due to sparseness of
countries with national estimates of data, inpatient studies were also included.
incidence and mortality, these data For some pathogens it was assumed
were used. The median and associated that different aetiological agents, such
uncertainty intervals for diarrhoeal as Shigella spp., NTS and Campylobacter
diseases for the subregion were used spp. had similar clinical profiles.
to estimate incidence and mortality of Initial estimates for the 61 countries in
diarrhoeal diseases for other countries low-mortality (EUR and other subregion
within these subregions without national A) countries had been prepared
data [40]. using this second “CHERG approach”.
The second approach was applied to However, it was recognized that for
the remaining 133 countries worldwide. 3
http://www.who.int/gho/en/ Accessed 6 June 2014
WHO Estimates of the global burden of foodborne diseases
33
(GDB2010) data were used to estimate region except those free from M. bovis
the burden of disease for typhoid, in cattle were assumed to have the same
paratyphoid and hepatitis A [58]. proportion of human TB infections due
IHME provided country-specific, age- to M. bovis. To account for internationally
standardized prevalence data for typhoid acquired infections, all countries free
and paratyphoid fever. These data were of M. bovis in cattle were assigned the
converted to incidence by dividing by lowest observed proportion of human
duration, and partitioned into typhoid TB infections due to M. bovis (0.3%).
and paratyphoid assuming a 1.0 to 0.23 To derive estimates of human M. bovis
ratio [59]. Country-specific hepatitis A incidence, WHO country-specific human
mortality data, stratified by age and sex, TB incidences were multiplied by the
were converted to incidence assuming a estimate of the proportion of human
CFR of 0.2%. TB infections that were due to M. bovis
Rates of iNTS are highly correlated with [65]. To estimate mortality associated
HIV prevalence and malaria risk [60]. To with M. bovis that accounted for HIV
estimate iNTS incidence globally, age- co-morbidity estimates were used of
mortality due to human TB in HIV-
HAZARD SPECIFIC
specific estimates of incidence from a
METHODOLOGY
systematic review [60] were used to negative persons (WHO data). Mortality
construct a random effect log linear data were adjusted by assuming that the
model using covariates of country- CFR for M. bovis was 20% lower than
specific HIV and malaria deaths, and human TB, as M. bovis infections are
the log of Gross Domestic Product. As more likely to be extrapulmonary [66].
data were sparse, incidence for all ages To estimate the incidence and mortality
was predicted, which was converted to for brucellosis a systematic review was
age-specific incidence based on age updated and included additional data
profiles for iNTS cases in low and high on 32 countries that were considered
incidence settings [60]. From this, iNTS Brucella-free in livestock (free of B.
incidence was predicted among persons abortus in cattle and B. melitensis
not infected with HIV [61, 62]. To estimate in sheep and goats) [67]. Incidence
deaths, it was assumed that the CFR data were imputed to countries
for iNTS in non-HIV infected individuals without estimates using a Bayesian
was a uniform distribution with a range log-normal random effects model,
5– 20% in sub-region B-E countries except for countries that were Brucella-
and range 3.9– 6.6% in sub-region A free in livestock [68]. To account for
countries [63]. internationally acquired infections, all
Estimates for M. bovis infections were countries that were Brucella-free in
based on a systematic review where livestock were assigned the median
the proportion of human tuberculosis incidence of human brucellosis reported
(TB) infections due to M. bovis ranged from these countries. The CFR for
from 0.3% in AMR to 2.8% in AFR [64]. brucellosis was 0.5%, and 40% of cases
Fifty-one countries were identified that resulted in chronic infections, and 10% of
were free from M. bovis in cattle, based cases in males resulted in orchitis [69].
on European Union certification and The incidence and mortality for listeriosis
the World Animal Health Information were estimated using a systematic
System (WAHIS) of World Organization review that is described elsewhere [70].
for Animal Health.4 All countries in a In accordance with standard burden of
4
OIE – www.oie.int/wahis disease practice, stillbirths were excluded
WHO Estimates of the global burden of foodborne diseases
35
in our baseline burden estimates. The the other intestinal protozoa, as citation
CFR was 14.9% for perinatal cases and frequency had remained constant over
25.9% for other cases. the same period.
Incidence and mortality data for botulism Foodborne trematodes of high priority
were only available from countries were Fasciola spp., Clonorchis spp.,
in Europe and North America. The Opisthorchis spp., Paragonimus spp.
estimation was limited to the 55 countries and intestinal trematodes such as
in EUR and AMR subregion A, which was Fasciolopsis buski, Heterophyes spp. and
based on the median incidence derived Metagonimus spp. Three cestode species
from countries with national estimates were considered important: Echinococcus
of botulism. It was estimated that 35% of granulosus, E. multilocularis and Taenia
botulism cases were severe and that the solium. The cestode Taenia saginata
CFR of severe botulism was 15%. was considered likely to have a very
low burden for human health because
of the lack of serious sequelae resulting
4.3 Parasitic Hazards from intestinal taeniosis, and hence was
At the first formal meeting of FERG, excluded from the priority list. Foodborne
the PDTF initially reviewed all parasitic Chagas disease was also considered for
diseases that could be potentially possible inclusion at the second FERG
transmitted by food (Appendix 3) with meeting, but resources were not available
14 parasitic diseases selected as high to commission work on the foodborne
priority. The selection criteria of these transmission of a primarily vector-borne
14 diseases was based on: proportion of disease. Finally the nematode species
foodborne transmission; severity of illness believed to have high impact were
and/or sequelae; frequency of illness Anisakidae, Ascaris spp. and Trichinella
and/or sequelae causes; global relevance; spp. Disease caused by the Anisakidae
particular regional relevance; propensity was later considered to be an uncommon
to cause outbreaks; and availability foodborne disease and was subsequently
of existing evidence to derive burden removed from the priority list.
estimates (see meeting reports from The incidence of each of the parasitic
FERG 1 and 2). diseases was estimated where possible.
Three intestinal protozoa genera – For cysticercosis, the burden was
Cryptosporidium, Entamoeba and Giardia estimated from a proportion of the
– were considered a priority, as they prevalent epilepsy cases, i.e. the number
were likely to result in a high disease of actual cases of disease, as further
burden, and the frequency of citations detailed below. Those incident cases with
for these parasites had been markedly sequelae (or diseased individuals) were
increasing between 1990 and 2008. For assigned years of life lost (YLLs) if fatal,
methodological reasons, the burden of or years lived with disability (YLDs) with
the three priority intestinal protozoa that a DW that depended on the severity of
cause diarrhoeal disease was estimated the disease. For some diseases, such
by the EDTF as described above. as toxoplasmosis, many of the incident
cases do not have sequelae (i.e. they are
Toxoplasma gondii was also considered
sub clinical). Such cases were given a DW
to be of high priority because of the
of zero.
potential serious sequelae. Cyclospora
was also initially considered, but a Systematic reviews were undertaken to
decision was made to target resources to estimate the incidence, sequelae and
Hazard-specific methodology
36
mortality due to these diseases [71– 77]. between age of T and T+1 can be used to
Where possible, public health records estimate incidence.
describing numbers of cases presenting
Incidence estimates and clinical sequelae,
for treatment were reviewed. These data
for diseases caused by foodborne
were only available for some diseases
trematodes, were mainly based on the
in some countries. In others surveillance
results of two review articles [77, 78].
data were used (for example laboratory
Incidence rates for countries without
data on sero-conversion rates in
reported national prevalence were
the population).
imputed, but only where there were
For congenital toxoplasmosis (CT) reports of at least one autochthonous
a systematic search of nine major human infection, by using a hierarchical
databases for published and unpublished random-effects model and incidence
sources was conducted, alongside direct information from other countries as input
contact with the authors of source data [79]. In highly endemic zones, adult
materials. Searches were country specific. subjects either maintain the parasites
To be included, studies had to report acquired when young or can be newly
HAZARD SPECIFIC
on the incidence of CT, on positivity infected as the consequence of inhabiting
METHODOLOGY
to Toxoplasma-specific IgM in infants a zone of high infection risk. This
and pregnant women (including sero- suggests that, in those areas, the majority
conversion results) or on positivity to of infected adults should be chronically
Toxoplasma-specific IgG in the general infected. However, acute lesions by
population. Various modelling techniques repetitive infections are frequently
were used, depending on the country- superimposed on chronic disease [80].
specific data available, to estimate the Therefore, it is reasonable to assume
CT incidence and burden in each country. that such overlapping series of repeat
Reports of children born with CT, IgM infections result in life-long sequelae.
serology of infants and pregnant women, Thus the incidence of trematode infection
and age-stratified sero-prevalence in was estimated from the numbers of new
women and the general population, cases in each age cohort.
combined with fertility rates of specific
age groups, were used to directly To estimate the incidence of alveolar
estimate the incidence of CT, or the data echinococcosis (AE), due to infection
was used to input into models that were with the larval stage of Echinococcus
able to generate CT incidences from IgM- multilocularis, literature searches were
sero-positive rates in children or pregnant undertaken in any relevant databases
women, or from the IgG-sero conversion that could be accessed. These data
rates in women, combined with age- sources were synthesized to obtain
specific fertility rates. These data were estimates of the incidences of AE in
then synthesized into an estimate of the countries where E. multilocularis was
global incidence of CT and of the global known to be endemic. Further details of
burden of CT in disability-adjusted life the strategy to obtain the data, together
years (DALYs). Further details of the with the methodology to estimate
methodology, inclusion criteria, PRISMA incidences from the data, are described
statement and the modelling techniques in the report from FERG 2. For cystic
used are given in [76]. Data on sero- echinococcosis (CE), due to infection
prevalence were also used to estimate with the larval stage of E. granulosus, the
the incidence of acquired toxoplasmosis. results of a systematic review [73] and
Thus changes in sero-prevalence other databases were used.
WHO Estimates of the global burden of foodborne diseases
37
HAZARD SPECIFIC
METHODOLOGY
estimated were aflatoxin5, cyanide in [87, 88] and a low protein intake [89].
cassava, peanut allergen, dioxin and It is a disease of extreme poverty.
dioxin-like compounds6, methylmercury, Konzo mostly occurs in epidemics, but
lead, arsenic and cadmium. Only the sporadic cases are also reported. The
results for aflatoxin, cyanide in cassava, case definition includes the following
peanut allergen, and dioxin are presented criteria: (1) a visible, symmetrically spastic
here. The results for the metals will be abnormality of gait while walking and/
provided in a subsequent publication. or running; (2) a history of abrupt onset
For each of the four chemicals, a (less than one week), followed by a non-
systematic literature review was progressive course in a formerly healthy
conducted. It was concluded that burden person; and (3) bilaterally exaggerated
estimates could be developed for: knee and/or ankle jerks without signs of
(1) cyanide in cassava, and associated disease in the spine [89, 90].
konzo syndrome; (2) peanut (Arachis Because konzo mostly affects remote
hypogaea) allergy; (3) aflatoxin and rural areas where health infrastructure is
hepatocellular carcinoma (HCC); poor or non-existent, many cases remain
(4) dioxin and hypothyroidy; and undiagnosed or unreported, so the true
(5) dioxin and decrease in sperm count. burden of disease remains unknown. No
cases have been reported from urban
4.4.1 Cyanide in cassava areas. A total of 2376 konzo cases have
Cassava is an important staple for over been reported in 5 countries in Africa
800 million people in approximately (Cameroon, Central African Republic,
80 countries, mostly in sub-Saharan Democratic Republic of Congo (DRC),
Africa but also in Asia, the Pacific, Mozambique, and United Republic of
and South America [85]. Cassava Tanzania) [86], corresponding to 149
tubers contain a varying quantity of cases per year for 122 million people.
cyanogenic glucosides, which protect Dividing the average annual number
the root against attack by animals and of cases for each country by the
5
The term, "aflatoxin," refers to all aflatoxins. corresponding country population
6
The term, “dioxin,” refers to dioxins and dioxin- produces an observed incidence ranging
like PCBs. from 0.043 to 0.179 per 100 000. The
WHO Estimates of the global burden of foodborne diseases
39
52% of cases referred to a specialist derived [111] and global populations, the
allergy medical practice in Australia population attributable fractions (PAFs)
suffered from mild symptoms (skin and by country were estimated, and applied
subcutaneous tissue involvement only), to HCC incidence and mortality based on
42% from moderate symptoms (features information from WHO [112, 113].
suggestive of respiratory, cardiovascular A Bayesian log-normal random effects
or gastrointestinal involvement), and model [79] was used to extrapolate
6% from severe symptoms (cyanosis, available PAFs to countries without
hypotension, confusion, collapse, loss of data. Age-specific incidence estimates
consciousness, incontinence). were derived from a study in China
The DW for peanut allergy was assigned comparing age-specific incidence of
as a weighted average accounting for HCC in Qidong, a city in China with high
this severity distribution. GBD2010 aflatoxin exposure, and Beijing, a city
DWs [82] for the health states defined with low aflatoxin exposure [114]. The
in the category “Asthma: controlled” YLD and YLL envelopes for HCC available
(DW=0.009) are considered applicable from WHO were multiplied by the
for mild and moderate cases (94%), and proportion of the burden due to aflatoxin.
HAZARD SPECIFIC
METHODOLOGY
“Generic uncomplicated disease: anxiety Thus no DW was directly involved in
about the diagnosis” (DW=0.054) for the calculation.
severe cases (6%), because anxiety
is known to affect Quality of Life in 4.4.4 Dioxin
food allergic patients [107], leading to Dioxins are mainly by-products of
a severity-weighted DW of 0.012 for industrial processes, but can also
clinically relevant peanut allergy. Unlike result from natural phenomena, such
other childhood allergies, such as cow’s as volcanic eruptions and forest fires.
milk and egg allergy, peanut allergy rarely More than 90% of human exposure is
resolves [108, 109]. through food, mainly meat and dairy
products, fish and shellfish [115]. Due
4.4.3 Aflatoxin to the bio-accumulating and lipophilic
Aflatoxins are secondary metabolites characteristics of dioxins, daily dietary
of the fungi Aspergillus flavus and A. exposure leads to accumulation of these
parasiticus, and less frequently other compounds in human body fat. In adults
Aspergillus species such as A. nomius this accumulation is thought to reach
[110]. These species are prevalent in a constant level (i.e. a steady state).
food crops – particularly maize, peanuts Consequently, the dioxin body burden,
(groundnuts), oilseeds and tree nuts rather than the daily exposure, is taken as
– in tropical and subtropical regions the dose metric for chronic toxicity risk
worldwide [110]. It is believed that all and the assessment of dioxins [116– 121].
aflatoxin exposure results from food In this context the dioxin concentration
consumption. A multiplicative model in breast milk fat directly reflects the
concentration in body fat [121– 124].
was assumed for the effects of aflatoxin
exposure and hepatitis B virus (HBV) Many national authorities have
infection on hepatocellular carcinoma. programmes in place to monitor dioxin
Aflatoxin exposure by country is that in the food supply and breast milk
described by Liu and Wu [110]. To [124– 126]. Dioxin-induced pre-natal and
account for differences in background post-natal hypothyroidy and pre-natally
rates between the study population from induced reduced sperm production have
which the cancer potency factor was been found to be the most sensitive
WHO Estimates of the global burden of foodborne diseases
41
HAZARD SPECIFIC
considered in the disease models.
METHODOLOGY
studies [141]. To ensure comparability, the For diarrhoea, for instance, DWs were
CTF adopted the DWs that were used available for mild, moderate and severe
for WHO’s Global Health Estimates [4]. diarrhoea, although the disease models
These DWs were based on those derived only included diarrhoea as such. In those
for the GBD 2010 study [82], but with an cases, weighted averages were calculated
alternative value for primary infertility (i.e. based on published reviews of severity
0.056 instead of 0.011). The latter revision distributions, avoiding an over- or under-
was motiviated by an analysis showing estimation of YLDs that would occur if
that the GBD 2010 weights undervalued only one DW would have been selected.
the health states associated with fertility
WHO Estimates of the global burden of foodborne diseases
43
Table 1. FERG hazards, causally related health states and corresponding disability weights (DWs).
Details on the derivation of the DWs are provided in Appendix 4.
HAZARD SPECIFIC
METHODOLOGY
NEMATODES
Ascaris spp. Ascariasis infestation 0.030
Mild abdominopelvic problems due to
0.012
ascariasis
Severe wasting due to ascariasis 0.127
Trichinella spp. Acute clinical trichinellosis 0.637
Trematodes
Abdominopelvic problems due to heavy
Clonorchis sinensis 0.123
clonorchiosis
Abdominopelvic problems due to heavy
Fasciola spp. 0.123
fasciolosis
Abdominopelvic problems due to heavy
Intestinal flukes (2) 0.123
intestinal fluke infections
Abdominopelvic problems due to heavy
Opisthorchis spp. 0.123
opisthorchiosis
Central nervous system problems due to
Paragonimus spp. 0.420
heavy paragonimosis
Pulmonary problems due to heavy
0.132
paragonimosis
ORGANIC POLLUTANTS
Dioxin Infertility 0.056
Hypothyroidy due to pre-natal exposure 0.019
Hypothyroidy due to post-natal exposure 0.019
TOXINS AND ALLERGENS
Hepatocellular carcinoma: diagnosis and
Aflatoxin 0.294
primary therapy
Hepatocellular carcinoma: metastatic 0.484
Hepatocellular carcinoma: terminal phase
0.508
with medication
Hepatocellular carcinoma: terminal phase
0.519
without medication
Cyanide in cassava Konzo 0.065
Peanut allergens (1) Living with peanut-induced allergy 0.012
Notes: (1) Excluded from global burden assessments. (2) Includes Echinostoma spp., Fasciolopsis buski, Heterophyes spp.,
Metagonimus spp. and other foodborne intestinal trematode species.
WHO Estimates of the global burden of foodborne diseases
45
NO. OF
HAZARD GROUPS HAZARDS PANEL STRUCTUREa
PANELS
DIARRHEAL DISEASE
HAZARD SPECIFIC
Global 3
METHODOLOGY
multilocularis
CHEMICALS
Lead Global 1
Total 19
a
Experts on a global panel were asked to provide estimates for all 14 sub regions, whereas experts on a sub-regional panel could
choose a set of sub regions depending on their expertise.
matter expert, should be able to define a were designed to focus on questions that
narrow uncertainty range that captures are substantively related to foodborne
the value. Therefore, an alternative is to illness source attribution. Further, to
ask about selected data or values in the account for the wide range of scientific
topic domain, about which the expert backgrounds and experiences, seed
will not have perfect knowledge, nor questions covered a range of substantive
access to realization values at the time topics relevant to source attribution. Five
they are answering the seed questions, main categories of seed questions were
but for which the values are known to the identified for the panels on biological
analyst. Such “retrospective” questions hazards (diarrhoeal pathogens and
are frequently used in expert elicitations parasites): (1) dietary patterns and food
applying the Cooke Classical Model (see supply; (2) under 5 years mortality
e.g. [31, 148]). rate; (3) access to improved water and
sanitation; (4) disease surveillance; and
In the present case, the seed questions
(5) systematic reviews related to these
formulated were a mixture of
and other scientific topics relevant to
retrospective and prospective seed
source attribution. For the panel on lead,
HAZARD SPECIFIC
variables. It is possible that expert
METHODOLOGY
questions were categorized as: (1) mean
uncertainty judgments vary by subject
blood levels; (2) dietary exposure;
matter domain. In this study, the
and (3) dietary patterns and food
possibility of such biases relevant to
supply. Examples of seed questions are
foodborne illness source attribution was
presented in Table 3.
of concern. Therefore, the seed questions
All experts comprising each panel were the questionnaires did provide experts
asked the same set of seed questions, with an option to indicate additional
and several sets of seed questions were routes of transmission if they disagreed
used across panels. This allows some with the TF’s evaluation.
consistency checks to be performed
Experts were asked to complete a set
between panels on performance and
of tables for each assigned hazard and
scoring outcomes. The number of
subregion. Experts were provided with
questions varied from nine to twelve in
the tables at the end of the telephone
total, depending on the panel. Experts
interview during which the seed
were asked to provide a central judgment
questions were asked, and the facilitator
in terms of the median value, and a
went through several target questions
90% credible interval for each question.
Seed questions were administered with the experts to ensure that they
by facilitators through one-to-one understood the task. For the target items
telephone interviews. The experts were (but not the seed questions), the experts
not presented with the seed questions were free to consult any information
before the interview and they were asked sources they felt appropriate in the four-
to provide estimates based on their week period they were given to return
experience, knowledge and judgment, the target item spreadsheets.
without referring to other sources As with the seed questions, the experts
of information. were asked to provide their 5th, 50th
and 95th percentile values for each
4.6.6 Target questions question. Technically, the median values
Target questions are the substantive of a joint distribution do not need to add
questions of interest. In this study, for all up to 100%, but because we included
identified hazards, we enquired about the a category “other”, we asked about a
percentage of all human disease cases joint distribution that logically spanned
caused by exposure through each of a all possible exposure routes. Therefore,
number of indicated exposure routes. the experts’ median values for source
The point of exposure was chosen as the attribution percentages for a hazard
point of attribution, i.e. the experts were should sum to a value close to 100%.
asked to distribute the disease burden In individual cases, where these sums
on the sources that was the direct cause were found to differ significantly (i.e.
of human exposure. So, for example, outside 100% ± 10%), the experts
someone with a norovirus infection might concerned were asked to review their
be exposed by eating food contaminated responses.
with the virus, although the food may
have been contaminated by wastewater 4.6.7 Data analysis
at an earlier stage. Weights for individual experts were
Exposure routes varied between hazards, computed using the Classical Model
as indicated in Table 4. In order to formulation, using the EXCALIBUR
reduce the time and effort burden of software, by multiplying their calibration
the elicitation on expert panelists, the and informativeness scores, with
hazard-based TFs decided which hazard the products then jointly normalized
exposure routes were relevant for present to sum to unity over all experts in
purposes. So, for example, human-to- the group. An expert was positively
human transmission was excluded as an weighted only if his/her p-value was
exposure route for Brucella spp. However, above a certain threshold, chosen to
Hazard-specific methodology
50
optimize the combined score across all The quantiles corresponding to these
seed items. See [146, 149] for further random deviates were then obtained
details on the computation of expert via linear interpolation. To ensure that
performance weights. the random attributional proportions
summed to 100%, a “normalization” step
The normalized experts’ performance
was applied per iteration, in which each
weights were used to construct the joint
random value was divided by the sum
probability distribution complying with
of random values for each exposure
their assessments for individual target
pathway. The resulting 10 000 normalized
questions (i.e. attributable proportion
random attributional proportions were
of illness per pathway, subregion and
then summarized by their median and a
hazard). In a final step, 10 000 random
95% uncertainty interval defined by the
values from the marginal cumulative
2.5th and 97.5th percentiles. These final
distributions were simulated using
manipulations were performed in R 3.1.1
probability integral transform [34].
[150] using functions available in the
First, independent vectors of 10 000
‘FERG’ package [151].
random deviates from a Uniform (0,100)
HAZARD SPECIFIC
distribution, per exposure category
METHODOLOGY
within a hazard-subregion combination,
were generated.
WHO Estimates of the global burden of foodborne diseases
51
COOKWARE, POTTERY OR
GLASSWARE
AND WILD)
WATER
OTHER
PAINT
FOOD
TOYS
SOIL
AIR
HAZARD
DIARRHOEAL DISEASE
PARASITIC DISEASE
CHEMICALS
Notes: N/A = not applicable, meaning that these exposure routes were considered not possible or extremely unlikely by the
respective FERG TF.
defines the burden of a specific foodborne models, and not merely as biological
hazard as that resulting from the health disease models. While biological disease
states, i.e. acute and chronic manifestations models merely reflect the natural history
of disease, including death, that are causally of disease, computational disease models
related to the hazard transmitted through also reflect the input parameters needed
food, and which may become manifest to calculate incidence and mortality of
at different time scales. This approach each of the relevant health states. As
thus allows for a comprehensive estimate such, computational disease models are
of the burden of disease due to a certain a combination of disease biology and
hazard, including sequelae, which may have data availability.
a higher burden than acute illness alone
Computational disease models may be
[153– 155].
represented as directed acyclic graphs,
defined by parent and child nodes and
4.7.1 Disease models and
directed edges (arrows) defining the
epidemiological data
relationships between nodes. In the CTF
The starting point of the CTF workflow framework, parent nodes were either
was the outline of disease models
HAZARD SPECIFIC
incidence, mortality, YLD or YLL rates,
METHODOLOGY
for each of the included hazards (as while child nodes were multiplicative
chosen by the hazard-based TFs), and elements, such as proportions or ratios
the epidemiological data inputs that (reflecting, e.g. the probability of
parameterized these disease models. developing a specific symptom following
To obtain this information, systematic infection, or the proportion of illnesses
reviews were commissioned and attributable to the concerned hazard).
managed by three hazard-based TFs, i.e. A specific disease model “language” was
the Enteric Diseases Task Force (EDTF), developed to denote the relationship
the Parasitic Diseases Task Force (PDTF), and contribution of the different nodes.
and the Chemicals and Toxins Task Rectangles defined parent nodes, and
Force (CTTF). rounded rectangles defined child nodes.
The course of disease is characterized Grey nodes did not contribute directly
by various health states (e.g. acute to the DALYs, green nodes contributed
or chronic phases; short-term or YLDs, and red nodes contributed YLLs.
long-term sequelae), possibly having Nodes that contributed to the incidence
different severity levels. A disease of the index disease were identified by
model, also referred to as an outcome a thick border. Appendix 5 gives the
tree, is a schematic representation of disease models for all 36 FERG hazards.
the various health states associated In general, three main approaches can be
with the concerned hazard, and the distinguished for estimating the burden
possible transitions between these due to a specific hazard in food, i.e.,
states. A disease model for each categorical attribution, counterfactual
hazard was defined by the members analysis, and risk assessment. Table
and commissioned experts of each 5 gives an overview of the modelling
hazard-based TF, considering relevant strategy applied for each included
health outcomes identified in the hazard. As the choice of the modelling
respective reviews. strategy was mainly driven by the type
In the context of the CTF, disease models of data available, no sensitivity analyses
were defined as computational disease could be performed to triangulate
different modelling approaches.
WHO Estimates of the global burden of foodborne diseases
53
Table 5. Modelling strategies for the hazards included in the WHO global burden of foodborne
disease estimates
BURDEN
DISEASE MODEL FOODBORNE
HAZARD ATTRIBUTION IMPUTATION
APPROACH ATTRIBUTION
APPROACH
DIARRHOEAL DISEASE AGENTS
Categorical Modified CHERG Expert elicitation
Norovirus Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Campylobacter spp. Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Enteropathogenic E. coli Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Enterotoxigenic E. coli Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Shiga toxin-producing E. coli Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Non-typhoidal S. enterica Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Shigella spp. Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Vibrio cholerae Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Cryptosporidium spp. Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Entamoeba histolytica Attribution
attribution approach [40] [156]
Categorical Modified CHERG Expert elicitation
Giardia spp. Attribution
attribution approach [40] [156]
INVASIVE INFECTIOUS DISEASE AGENTS
Categorical Expert elicitation
Hepatitis A virus Direct: GBD2010 [3] N/A (1)
attribution [156]
Categorical Random effects Expert elicitation
Brucella spp. Transition
attribution [79, 151] [156]
Categorical Random effects
Listeria monocytogenes, perinatal Transition 100%
attribution [79, 151]
Categorical Random effects
Listeria monocytogenes, acquired Transition 100%
attribution [79, 151]
Categorical
Mycobacterium bovis Attribution N/A (1) 100%
attribution
Categorical Expert elicitation
Salmonella Paratyphi Direct: GBD2010 [3] N/A (1)
attribution [156]
Categorical Expert elicitation
Salmonella Typhi Direct: GBD2010 [3] N/A (1)
attribution [156]
Categorical Random effects Expert elicitation
Toxoplasma gondii, congenital Transition
attribution [79, 151] [156]
Categorical Random effects Expert elicitation
Toxoplasma gondii, acquired Transition
attribution [79, 151] [156]
ENTERIC INTOXICATIONS
Categorical
Bacillus cereus (2) Direct Uniform 100%
attribution
Categorical
Clostridium botulinum (2) Direct Uniform 100%
attribution
Categorical
Clostridium perfringens (2) Direct Uniform 100%
attribution
Hazard-specific methodology
54
BURDEN
DISEASE MODEL FOODBORNE
HAZARD ATTRIBUTION IMPUTATION
APPROACH ATTRIBUTION
APPROACH
Categorical
Staphylococcus aureus (2) Direct Uniform 100%
attribution
CESTODES
Echinococcus granulosus, cases Categorical Random effects Expert elicitation
Transition
seeking treatment attribution [79, 151] [156]
Echinococcus granulosus, cases Categorical Random effects Expert elicitation
Transition
not seeking treatment attribution [79, 151] [156]
Categorical Random effects Expert elicitation
Echinococcus multilocularis Transition
attribution [79, 151] [156]
Categorical
Taenia solium Attribution N/A (1) 100%
attribution
NEMATODES
Categorical Expert elicitation
Ascaris spp. Direct: GBD2010 [3] N/A (1)
attribution [156]
Categorical
Trichinella spp. Direct N/A (1) 100%
attribution
HAZARD SPECIFIC
METHODOLOGY
TREMATODES
Categorical Random effects
Clonorchis sinensis Direct 100%
attribution [79, 151]
Categorical Random effects
Fasciola spp. Direct 100%
attribution [79, 151]
Categorical Random effects
Intestinal flukes (3) Direct 100%
attribution [79, 151]
Categorical Random effects
Opisthorchis spp. Direct 100%
attribution [79, 151]
Categorical Random effects
Paragonimus spp. Direct 100%
attribution [79, 151]
ORGANIC POLLUTANTS
Random effects
Dioxin Risk assessment Direct 100%
[79, 151]
TOXINS AND ALLERGENS
Counterfactual Random effects
Aflatoxin Attribution 100%
analysis [79, 151]
Categorical
Cyanide in cassava Direct Uniform 100%
attribution
Categorical
Peanut allergens (2) Direct Uniform 100%
attribution
Notes: (1) N/A = not applicable as no imputation had to be performed because data were used that had already been imputed.
(2) Excluded from global burden assessments. (3) Includes Echinostoma spp., Fasciolopsis buski, Heterophyes spp.,
Metagonimus spp. and other foodborne intestinal trematode species.
WHO Estimates of the global burden of foodborne diseases
55
HAZARD SPECIFIC
sex distribution; and (4) if applicable,
METHODOLOGY
model as the default model for imputing
the duration. Using a drop-down menu, missing country-level incidence data.
different formats could be selected for We used the subregions as defined in
entering the input parameters, including Appendix 2 as the random effect or
a mean and 95% confidence interval; cluster variable. This model assumes
minimum, most likely and maximum that the log-transformed incidence rate
percentiles; the shape and rate of a in country belonging to subregion
Gamma distribution (for rates); and the arises from a Normal distribution with
shape parameters of a Beta distribution
subregion specific mean and a within-
(for proportions). Gamma and Beta
region (= between-country) variance
distributions were chosen because
their domains correspond to that of . Each subregion specific mean
rates and proportions, respectively, is in turn assumed to arise from a
and because their parameters have an Normal distribution with mean and a
intuitive interpretation (i.e., number of between-region variance :
cases and sample size, respectively,
number of positives and number of
negatives). Likewise, different levels of
stratification could be selected for the
duration parameters (i.e. none, by age
After fitting this hierarchical random
only, by sex only, by age and sex). Age
effects model to the available data,
distribution, sex distribution and duration
incidence values for countries with
were allowed to vary by country, by
no data were imputed based on
defining different “groups” and assigning
the resulting posterior predictive
countries to “groups”.
distributions. In other words, we
4.7.3 Imputation represented missing incidence data by
log-Normal distributions based on the
Extrapolation or imputation models may fitted mean and variance parameters. For
be needed when literature searches countries in a subregion where none of
cannot provide essential epidemiological the countries had data, the log-incidence
7
was imputed as multiple random draws
Microsoft Corp., Redmond, Washington, USA
WHO Estimates of the global burden of foodborne diseases
57
from a Normal distribution with mean while for other hazards we directly
equal to the fitted global intercept drew on GBD 2010 estimates (Table
and variance equal to the sum of the 5). For the remaining hazards, either
fitted between-region variance and epidemiological data were used that
did not need (further) correction, or the
the fitted within-region variance (thus underreporting factor was included in the
imputing the log-incidence as that of disease model (which was the case for
a “random” country within a “random” Trichinella spp. and cyanide in cassava).
subregion, with the uncertainty interval
describing the variability between and For aflatoxin, the same random effects
within subregions): model was used to extrapolate PAFs, but
now using logit-transformed instead of
log-transformed values.
For countries in a subregion where at The model was implemented in a
least one of the other countries had Bayesian framework, using independent
data, the log-incidence was imputed as Normal(0, 1e5) priors for and ;
multiple random draws from a Normal a Uniform(0, 10) prior for ; and a
distribution with mean equal to the
Folded-t(1) prior for , as suggested
fitted region-specific intercept and
by Gelman [160]. Sensitivity analyses
variance equal to the fitted within-region
using Gamma priors for the variance
variance (thus imputing the log- parameters did not yield meaningful
incidence as that of a “random” country differences. The model was run in JAGS
within the concerned subregion, with [161] through the ‘rjags’ package in R
the uncertainty interval describing the [162]. After a burn-in of 5000 iterations,
variability within subregions): another 5000 iterations were retained
for inference. Two chains were run,
and convergence was ascertained
When countries were considered free through density and trace plots,
from exposure through the food chain, and the multivariate potential scale
they were excluded from the imputation reduction factor (or Brooks-Gelman-
model and thus did not contribute to Rubin diagnostic).
the subregional estimates. This was the A crucial assumption made by this
case for Brucella spp., as discussed in imputation model is that missing data
[168], and foodborne trematodes and were considered “missing at random”
Echinococcus spp., as discussed in [261]. (MAR), meaning that missingness was
For countries with available incidence independent of the unobserved data,
data, no imputation was performed. The given the observed data [163, 164]. In our
incidence data used in the probabilistic case, this assumption implied that, within
burden assessments were thus a each subregion, countries with data
combination of actual data and imputed provided unbiased information on those
estimates. No additional step had to be without data, and that, across subregions,
included to correct incidence data for subregions with data provided unbiased
potential underreporting, as this was information on those without data.
already captured by the previous steps However, for five hazards (Bacillus cereus,
of the framework. Indeed, for the hazards Clostridium perfringens, Clostridium
that used an attributional model, disease botulinum, Staphylococcus aureus and
envelopes were used that had already peanut allergens), only data from high-
been corrected for underreporting,
Hazard-specific methodology
58
HAZARD SPECIFIC
defined by the lowest and highest to health care. We used the highest
METHODOLOGY
globally observed incidence or mortality projected LE for 2050 as normative LE
rates. To ensure consistency with results for calculating YLLs [4]. This LE table has
of the Child Health Epidemiology a LE at birth of 92, higher than that of the
Reference Group (CHERG), alternative LE tables used in the GBD studies, which
imputation approaches were applied for were based on current death rates [1, 6].
estimating aetiological fractions for the Since even for the lowest observed death
eleven diarrhoeal agents [40, 50, 168]. rates there are a proportion of deaths
For seven other hazards, no imputation which are preventable or avertable, the
had to be performed because data were highest projected LE for the year 2050
used that had already been imputed. was deemed to better represent the
This was the case for hepatitis A virus, maximum life span of an individual in
Salmonella Typhi, Salmonella Paratyphi, good health, while acknowledging that
Ascaris spp. and Taenia solium, for it may still not represent the ultimate
which GBD2010 data were used, and for achievable human life span [166].
Mycobacterium bovis and Trichinella spp.,
In line with current global burden of
for which other published data were used
disease assessments, no age weighting
[84, 165].
or time discounting was applied [4, 6].
HIV infected invasive salmonellosis cases
4.7.4 Probabilistic burden assessment
and deaths, and HIV infected M. bovis
For each hazard, incidence, mortality, deaths, were excluded from the burden
YLD, YLL and DALY rates were calculated estimates. No further corrections were
for 11 age groups (<1; 1-4; 5-14; 15-24; made for possible co-morbidities.
25-34; 35-44; 45-54; 55-64; 65-74;
75-84; *85) and both sexes. When Parameter uncertainty was taken into
necessary, age and sex specific rates account by performing the burden
were obtained by multiplying the overall assessments in a probabilistic framework.
rates with outcome specific age and sex Ten thousand Monte Carlo (parametric
distributions. The reference year for the bootstrap) simulations of the input
calculation of absolute numbers was parameters were generated to calculate
2010, with population estimates obtained 10,000 estimates of incidence, mortality,
from the 2012 revision of the United YLD, YLL and DALY rates. These 10,000
WHO Estimates of the global burden of foodborne diseases
59
HAZARD SPECIFIC
METHODOLOGY
WHO Estimates of the global burden of foodborne diseases
61
RESULTS
5
WHO Estimates of the global burden of foodborne diseases
63
RESULTS
In this section, the results of the global interest (2). Of the 100 experts enrolled,
expert elicitation study are reported first. 78 completed interviews with facilitators
An overview of global and regional DALY and 73 returned their spreadsheets with
estimates according to hazard follows. their responses to the target questions
Subsequent sections report more specific and seed questions. The single main
hazard-based estimates, and include reason for not completing the interview
estimates for some hazards for which and returning the spreadsheet was time
global estimates could not be derived constraints. All responses were reviewed
and only regional estimates are reported (e.g. checked for missing estimates, that
(peanut allergen; toxin-producing species sums across pathways were close to
of bacteria). 100%, and that the 5th percentile < 50th
percentile < 95th percentiles), and some
experts were contacted for clarification
5.1 Attribution
of the responses they had provided. One
A total of 299 potential experts were expert was dropped after not responding
asked by email of their interest in to requests for clarification. This resulted
participating in the study. Of these154 in the responses of 72 experts being
replied positively and they were included in the final dataset. Table 6
requested to forward their CV, a filled-in shows the distribution of experts across
expert sheet and a signed declaration of panels, and Figure 3 shows distribution of
interest. A total of 103 did that. the experts by their geographical areas
Of these, 3 were not included due to lack of expertise.
of experience (1) or possible conflicts of
Table 6. The number of experts enrolled, interviewed and finally included in the elicitation
across panels
DIARRHEAL DISEASE
CHEMICALS
Lead Global 10 9 6
b
Total 100 78 72
a
Due to the sub regional structure of these panels, the number of experts varied between 10 and 15 depending on the hazard and
sub region.
b
As several experts served on more panels, the number of experts per panel does not add up to the total number of individual
experts included.
Results
64
Figure 3. Geographical distribution of experts according to working experience (>3 years) per
subregion. Several experts had experience in more than one subregion.
18 23
6
5
5
14
10
6
11 5
5 14
Notes: The subregions are defined on the on the basis of child and adult mortality, as described by Ezzati et al. [5]. Stratum A =
very low child and adult mortality; Stratum B = low child mortality and very low adult mortality; Stratum C = low child mortality and
high adult mortality; Stratum D = high child and adult mortality; and Stratum E = high child mortality and very high adult mortality.
The use of the term ‘subregion’ here and throughout the text does not identify an official grouping of WHO Member States, and the
“subregions” are not related to the six official WHO regions.
RESULTS
5.1.1 Expert performance produced by applying the performance
In this study, there were 115 distinct weights. In contrast, the informativeness
panels (i.e. panels that differed in properties of the equal weights solutions
membership or seed questions) and, were much lower than those provided
overall, performance weight and equal by performance weights solutions
weight combinations showed acceptable (Figure 4). This “trade-off” between
statistical accuracy. Only in the case accuracy and informativeness when
of the panel considering lead was the applying equal weights or performance
p-value of the performance-based weights is often seen, because the
combination small enough to cast doubt least accurate experts are typically the
on the usual criterion for statistical most informative, and their narrow 90%
accuracy, with p = 0.045 (i.e. less than confidence bands often have little or no
the 0.05 criterion). With a set of 115 overlap. Moreover, the combined score
panels, at least one score this low would using performance-based weights was
be expected even if the performance- above that of the equal weights pooling
based combination was always in 62% of the cases. It was therefore
statistically accurate. decided to use the performance weights
combinations for constructing the joint
Results obtained by applying equal probability distributions for the pathway
weights pooling and performance attribution estimates, as long as the
weights pooling were compared. The statistical accuracy was acceptable. It
equal weights solutions tended to have should also be noted that the weight
higher statistical accuracy than those
WHO Estimates of the global burden of foodborne diseases
65
Figure 4. Statistical accuracy versus informativeness of the experts included, when using equal
weight (blue) or performance weight (red) combinations, respectively.
1,8
1,6
1,4
1,2
Informativeness
1,0
0,8
0,6
0,4
0,2
0,0
1,E-02 1,E-01 1,E+00
Statistical accuracy (p-value)
RESULTS
subregions in the African, American and important transmission route for T. gondii
Eastern Mediterranean regions (AFR, AMR and Ascaris spp. in most subregions,
and EMR), water was identified as the most but there was a clear tendency for soil
important transmission route. For ETEC, to increase in relative importance in less
the estimated foodborne proportions were developed subregions (subregions D and
quite similar for all subregions with medians E) (Table A7.4 in Appendix 7). Specifically
ranging from 0.33 to 0.43 (Table A7.2 in for Ascaris spp., the foodborne route was
Appendix 7), but the foodborne route was assessed to be particularly important in
only assessed by experts to be the more developed subregions (A subregions).
important route in European subregions. There was only little geographical variation
For Cryptosporidium spp. and Giardia spp., between the median estimates for each
the foodborne proportions were also quite of the transmission pathways for the two
similar across subregions, but generally Echinococcus species. For E. granulosus,
considered less important, with medians animal contact was clearly believed to be
below 0.20 (Table A7.2 in Appendix the most important route, with medians
7). Human-to-human and waterborne just over 0.50. For E. multilocularis, the
transmission were the more important foodborne route was considered most
routes for these infections in all subregions. important, with medians ranging from
0.43 in EMR B to 0.58 in AFR D and E,
Figure 7 shows the subregional estimates
AMR B and D, and SEAR B and D, but
of the proportion foodborne for Salmonella
the estimates had very large uncertainty
Typhi, Vibrio cholerae, Entamoeba
intervals (Table A7.4 in Appendix 7).
WHO Estimates of the global burden of foodborne diseases
67
AFR E AFR E
AMR A AMR A
AMR B AMR B
AMR D AMR D
EUR A EUR A
EUR B EUR B
EUR C EUR C
EMR B EMR B
EMR D EMR D
SEAR B SEAR B
SEAR D SEAR D
WPR A WPR A
WPR B WPR B
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Proportion Proportion
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Proportion Proportion
Shigella spp.
AFR D
AFR E
AMR A
AMR B
AMR D
EUR A
EUR B
EUR C
EMR B
EMR D
SEAR B
SEAR D
WPR A
WPR B
AFR E AFR E
AMR A AMR A
AMR B AMR B
AMR D AMR D
EUR A EUR A
EUR B EUR B
EUR C EUR C
EMR B EMR B
EMR D EMR D
SEAR B SEAR B
SEAR D SEAR D
WPR A WPR A
WPR B WPR B
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Proportion Proportion
RESULTS
EUR A EUR A
EUR B EUR B
EUR C EUR C
EMR B EMR B
EMR D EMR D
SEAR B SEAR B
SEAR D SEAR D
WPR A WPR A
WPR B WPR B
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Proportion Proportion
Notes: Indicated on the line plot are the 2.5th, 5th, 50th, 95th and 97.5th percentiles.
WHO Estimates of the global burden of foodborne diseases
69
AFR E AFR E
AMR A AMR A
AMR B AMR B
AMR D AMR D
EUR A EUR A
EUR B EUR B
EUR C EUR C
EMR B EMR B
EMR D EMR D
SEAR B SEAR B
SEAR D SEAR D
WPR A WPR A
WPR B WPR B
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Proportion Proportion
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Proportion Proportion
Hepatitis A virus
AFR D
AFR E
AMR A
AMR B
AMR D
EUR A
EUR B
EUR C
EMR B
EMR D
SEAR B
SEAR D
WPR A
WPR B
Notes: Indicated on the line plot are the 2.5th, 5th, 50th, 95th and 97.5th percentiles. HAV = Hepatitis A virus.
Results
70
AFR E AFR E
AMR A AMR A
AMR B AMR B
AMR D AMR D
EUR A EUR A
EUR B EUR B
EUR C EUR C
EMR B EMR B
EMR D EMR D
SEAR B SEAR B
SEAR D SEAR D
WPR A WPR A
WPR B WPR B
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Proportion Proportion
RESULTS
EUR C EUR C
EMR B EMR B
EMR D EMR D
SEAR B SEAR B
SEAR D SEAR D
WPR A WPR A
WPR B WPR B
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Proportion Proportion
Notes: Indicated on the line plot are the 2.5th, 5th, 50th, 95th and 97.5th percentiles.
WHO Estimates of the global burden of foodborne diseases
71
Figure 9. Subregional estimates of the proportion of disease caused by foodborne exposure to lead.
Lead
AFR D
AFR E
AMR A
AMR B
AMR D
EUR A
EUR B
EUR C
EMR B
EMR D
SEAR B
SEAR D
WPR A
WPR B
Notes: Indicated on the line plot are the 2.5th, 5th, 50th, 95th and 97.5th percentiles.
Figure 10. Subregional estimates (medians) of the proportion of disease caused by exposure
to lead through eight different exposure routes.
exposure
AFR D
AFR E food
AMR A water
AMR B soil
AMR D air
EMR B paint
EMR D cookware
EUR A toys
EUR B other
EUR C
SEAR B
SEAR D
WPR A
WPR B
Proportion
Results
72
RESULTS
see the Supplementary Information
S. enterica, 32,000 were in the two for the overview publication [167]. The
African subregions, and included 22,000 Supplementary Information also includes
deaths due to invasive disease by this the results for total illnesses, deaths,
bacterium. The major non-diarrhoeal DALYs, YLLs and YLDs by all exposure
causes of foodborne deaths were due to pathways, for all hazards that were
Salmonella Typhi (52,000), the helminth included in the source attribution expert
Taenia solium (28,000) and hepatitis A elicitation in Appendix 7.
virus (28,000) and aflatoxin with 20,000
(95% UI 8,000-51,000) deaths.
WHO Estimates of the global burden of foodborne diseases
73
Table 7. Median global number of foodborne illnesses, deaths, Years Lived with Disability (YLDs),
Years of Life Lost (YLLs) and Disability Adjusted Life Years (DALYs), with 95% uncertainty
intervals, 2010.
1 176 854
Shiga toxin-producing 128 3 486 9 454 12 953
(754 108–
E. coli– STEC (55–374) (1 741–6 996) (4 140–27 208) (5 951–33 664)
2 523 007)
67 182 645 492 354
5 558 57 536 432 316
Protozoa (35 794 977– (239 400–
(2 593–11 958) (30 526–102 608) (195 372–960 910)
120 556 797) 1 034 790)
8 584 805
3 759 8 155 287 690 296 156
Cryptosporidium spp. (3 897 252–
(1 520–9 115) (3 598–17 355) (114 012–711 990) (119 456–724 660)
18 531 196)
28 023 571
1 470 20 851 115 740 138 863
Entamoeba histolytica (10 261 254–
(453–5 554) (7 431–53 080) (32 070–476 144) (47 339–503 775)
68 567 590)
28 236 123
0 26 270 0 26 270
Giardia spp. (12 945 655–
(0–0) (11 462–53 577) (0–0) (11 462–53 577)
56 996 454)
Results
74
393 239 110 971
1 957 13 324 124 884
Brucella spp. (143 815– (37 470–
(661–45 545) (4 095–315 952) (43 153–2 910 416)
9 099 394) 2 583 081)
RESULTS
Protozoa (7 403 516– (561 297–
(333–1 300) (511 314–1 175 619) (30 575–119 512)
14 904 324) 1 264 567)
10 280 089 829 071
684 763 326 62 899
Toxoplasma gondii (7 403 516– (561 297–
(333–1 300) (511 314–1 175 619) (30 575–119 512)
14 904 324) 1 264 567)
12 928 944 3 367 987 2 428 929 5 810 589
45 226
Helminths (8 957 617– (2 840 638– (1 869 610– (4 864 518–
(34 143–59 035)
24 008 256) 4 358 741) 3 173 545) 7 367 619)
1 033 097 1 048 937
139 238 250 15 535
Paragonimus spp. (730 118– (743 700–
(95 610–195 078) (160–371) (9 971–23 035)
1 423 031) 1 438 588)
217 632 247 920 650 157 908 356
19 712
Chemicals and toxins (172 024– (196 490– (283 769– (506 112–
(8 171–51 664)
1 140 463) 1 410 260) 1 617 168) 2 714 588)
632 901 636 869
21 757 19 455 3 945
Aflatoxin (265 578– (267 142–
(8 967–56 776) (7 954–51 324) (1 551–10 667)
1 606 493) 1 617 081)
Figure 11. Ranking of foodborne hazards, based on Disability-Adjusted Life Years at the global
level, with 95% uncertainty intervals, 2010.
10,000,000
Foodborne Disability-Adjusted Life Years
1,000,000
100,000
10,000
1,000
100
p.
gr ia s e
ia Fa nul p.
on ol us
m Bru tog p.
lla es
to p.
s s
cc or Dio p.
m ium ins
on loc p.
ob A his ris
te ris p.
b .
at s
Pa a P go in
us i A
.
br A p.
ch irus
ba ET e
en ov .
so s
m
l S Ty C
a
C
st al ca
go ty ii
ct EC
nt i
m pp
Ta Nor pp
pa ge spp
. e ph
pl Afl ovi
ia iru
rc uke
ic
a
us rd id
ra ara nd
oi nel EPE
el ma ox
sp
In his sp
sp
sp
ac sca sp
Vi tis sp
a TE
liu
ce en
m sci os
er
a
pi tin lyti
ni ph
x
er
cc Gia yan
oc a s
riu s
s
or ul
V
ol
av S
ho fl
lla
ti lla
er
hi
da la
c
ne
m
a
c
y
Cl ulti
d
io
hi
i
i
He Sh
on as
s
ph o
ba
lo
ic
te
ss
ty lm
co sp
Tr
py
l
oe
Ca
us
xo
no pto
n- Sa
m
O
co
To
yc
Ca
ta
lm
no
er
Ec Cr
M
En
st
Sa
hi
Li
no
Ec
hi
RESULTS
Notes: White dots indicate the median burden, black boxes the inter-quartile range (50% UI), black lines the 5 and 95 percentiles
(90% UI) and grey lines the 2.5 and 97.5 percentiles (95% UI). Note that the y-axis is on a logarithmic scale. Abbreviations:
EPEC = Enteropathogenic Escherichia coli; ETEC = Enterotoxigenic E. coli; STEC = Shiga toxin-producing E. coli.
subregions (see Table 8 for the detailed burden. In the AMR B region, important
data for all hazards and all subregions). causes of FBD burden were T. solium
In the SEAR D and SEAR B subregions, (25 DALYs per 100,000 population)
diarrhoeal disease agents contributed and T. gondii (20 DALYs per 100,000
approximately 50% of the total disease population). In the AMR D region, the
burden, mainly caused by a range of burden of T. solium was particularly high
hazards including EPEC, norovirus, at 69 DALYs per 100,000 population;
non-typhoidal S. enterica, ETEC and the trematodes Paragonimus spp.
Campylobacter spp. In both of these and Fasciola spp. contributing 53 and
subregions, there was also a considerable 46 DALYs per 100,000 population,
burden of Salmonella Typhi (180 DALYs respectively to the overall disease
per 100,000 population in SEAR B and burden.
110 DALYs per 100,000 population in
The burden due to chemical hazards was
SEAR D). The burden of disease due to
also highly localized. Aflatoxin caused
the fluke Opisthorchis spp. was almost
the highest burden in AFR D, WPR B
exclusively concentrated in SEAR B
and SEAR B, whereas dioxins caused the
(40 DALYs per 100,000 population). In
highest burden in SEAR D, EMR B and
EMR D, diarrhoeal disease agents were
D and EUR A. The burden of cassava
responsible for approximately 70% of the
cyanide was limited to the AFR regions,
total burden of FBD, with Campylobacter
and was similar to that of aflatoxin in
spp. the leading cause in the region,
AFR D.
followed by EPEC, non-typhoidal S.
enterica, Shigella spp. and ETEC. Other In the three European subregions,
important hazards in this region were diarrhoeal disease agents contributed
Salmonella Typhi, aflatoxin and hepatitis to 49-68% of the total burden of FBD,
A virus. with non-typhoidal S. enterica and
Campylobacter spp. being the most
In the WPR B subregion, diarrhoeal
important hazards. Other important
disease agents accounted for
hazards included T. gondii in all European
approximately 14% of the FBD burden,
subregions, Brucella spp. in the EUR B
with Campylobacter spp. the leading
and Mycobacterium bovis in the EUR C
cause. In this region, the seafoodborne
subregions. In the WPR A region, 65%
trematodes Paragonimus spp. and
of the burden was caused by diarrhoeal
Clonorchis sinensis were important
disease agents, with T. gondii and
contributors to the FBD burden. In
hepatitis A virus also contributing. Finally,
the AMR B and AMR D subregions,
in the AMR A region, diarrhoeal disease
the contribution of diarrhoeal disease
agents contributed approximately 67%
agents to the total burden was smaller
of the total burden, with non-typhoidal
than in other subregions (approximately
S. enterica and Campylobacter spp.
40% and 20%, respectively), with
the most important hazards; T. gondii
Campylobacter spp., norovirus and
and L. monocytogenes were also
non-typhoidal S. enterica causing most
relatively important.
Results
78
Table 8. Median rates of foodborne Disability Adjusted Life Years (DALYs) per 100 000
population, by subregion, with 95% uncertainty intervals, 2010.
AFR AFR AMR AMR AMR EMR EMR EUR EUR EUR SEAR SEAR WPR WPR
HAZARD
D E A B D B D A B C B D A B
1 276 1 179 140 315 362 571 52 685 711 36 293
35 41 49
TOTAL (459– (726– (97– (243– (205– (325– (33– (360– (343– (23– (219–
(23–49) (29–64) (33–77)
2263) 1764) 274) 575) 582) 954) 136) 1291) 1335) 170) 406)
Diarrhoeal 889 824 277 393 330 380
23 60 72 28 25 24 23 41
disease (196– (447– (153– (217– (154– (159–
(13–33) (36–94) (40–117) (17–39) (14–37) (13–35) (14–32) (21–65)
agents 1 731) 1 326) 460) 644) 576) 717)
28 33 3 69 3
75 76 3 12 13 4 3 55 4
Viruses (0.5– (0.4– (0.08– (0.8– (0.09–
(6–222) (0–225) (0.6–8) (0–38) (0–43) (0–11) (0.2–9) (0–224) (0–19)
79) 90) 9) 263) 7)
28 33 3 69 3
75 76 3 12 13 4 3 55 4
Norovirus (0.5– (0.4– (0.08– (0.8– (0.09–
(6–222) (0–225) (0.6–8) (0–38) (0–43) (0–11) (0.2–9) (0–224) (0–19)
79) 90) 9) 263) 7)
787 712 237 347 247 285
19 45 54 24 21 20 19 34
Bacteria (186– (393– (124– (190– (107– (119–
(10–28) (26–68) (28–87) (14–32) (11–32) (10–29) (11–27) (17–54)
1 482) 1 160) 403) 576) 429) 506)
75 97
Campyloba- 71 70 9 15 15 10 8 8 37 33 9 10
(40– (54–
cter spp. (35–119) (33–117) (5–14) (8–23) (8–26) (6–14) (4–13) (4–12) (14–87) (2–84) (5–14) (4–17)
109) 143)
Enteropatho- 0.006 9 0.006 0.006
136 138 7 46 60 0.004 0.004 64 65 5
genic E. coli– (0.001– (0.9– (0.001– (0.001–
(11–329) (6–327) (0.4–21) (9–114) (8–151) (0–0.01) (0–0.01) (3–144) (1–162) (0.3–13)
EPEC 0.02) 26) 0.02) 0.02)
Enterotoxi- 107 105 0.003
7 9 29 37 0.004 0.004 0.004 42 42 0.003 4
genic E. coli– (26– (17– (0–
(1–19) (2–25) (6–78) (5–102) (0–0.01) (0–0.01) (0–0.01) (3–106) (2–111) (0–0.01) (0.3–11)
ETEC 245) 240) 0.009)
Shiga toxin- 0.009 0.08 0.2 0.2 0.2 0.07 0.1 0.2 0.01
0.4 0.5 0.6 0.2 0.4
producing (0.002– (0.02– (0.05– (0.09– (0.1– (0.02– (0.03– (0.007– (0.002–
(0.09–1) (0.2–1) (0.2–1) (0.02–1) (0.1–1)
RESULTS
E. coli 0.03) 0.2) 0.3) 0.4) 0.5) 0.2) 0.4) 1) 0.04)
Non- 338 193 59
9 11 14 50 67 12 12 11 58 9 9
typhoidal (94– (44– (22–
(4–16) (2–20) (4–26) (17–82) (26–112) (7–18) (6–21) (5–19) (0–162) (5–14) (4–16)
S. enterica 612) 336) 154)
25 25
37 37 0.2 2 2 27 37 0.09 0.1 0.2 0.2 4
Shigella spp. (0.6– (0.7–
(0–156) (0–148) (0–1) (0–8) (0–9) (0–109) (0–145) (0–0.9) (0–1) (0–1) (0–1) (0.1–11)
84) 90)
0.2 28 36 0.1
Vibrio 70 143 0 0 0 0 0 0 2 0
(0.009– (0.9– (0.2– (0.005–
cholerae (2–197) (4–383) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0.2–3) (0–0)
0.5) 96) 133) 0.4)
6
20 21 0.2 2 3 7 0.2 0.2 0.2 10 10 0.2 1
Protozoa (0.9–
(0–74) (5–66) (0.01–1) (0.4–6) (0.5–8) (1–33) (0–0.8) (0–0.9) (0–0.9) (2–35) (2–39) (0–0.9) (0.2–4)
27)
Crypto- 0.2 0.7
12 12 1 3 3 0.1 0.1 0.1 6 6 0.1 0.3
sporidium (0.01– (0.08–
(0–44) (0–45) (0.1–5) (0–21) (0–24) (0–0.8) (0–0.8) (0–0.8) (0–28) (0–32) (0–0.8) (0–3)
spp. 0.9) 3)
Entamoeba 5 5 0 0.4 0.4 2 2 0 0 0 2 2 0 0.3
histolytica (0–41) (0–41) (0–0) (0–2) (0–3) (0–12) (0–17) (0–0) (0–0) (0–0) (0–17) (0–18) (0–0) (0–1)
0.6
0.7 0.7 0.03 0.4 0.5 0.4 0.03 0.03 0.03 0.1 0.1 0.03 0.3
Giardia spp. (0.005–
(0–3) (0–3) (0–0.1) (0–2) (0.01–3) (0–2) (0–0.1) (0–0.1) (0–0.1) (0–0.9) (0–1) (0–0.1) (0–1)
2)
WHO Estimates of the global burden of foodborne diseases
79
AFR AFR AMR AMR AMR EMR EMR EUR EUR EUR SEAR SEAR WPR WPR
HAZARD
D E A B D B D A B C B D A B
Invasive
146 147 73 137 244
infectious 10 38 49 10 19 16 272 10 65
(46– (55– (32– (38– (38–
disease (6–14) (16–76) (19–144) (7–15) (9–61) (10–29) (71–721) (5–132) (19–145)
342) 343) 148) 334) 623)
agents
0.5 0.8 1
27 18 1 2 2 32 1 1 5 58 5
Viruses (0.07– (0.03– (0.07–
(4–77) (3–55) (0.1–4) (0.2–7) (0.2–5) (2–102) (0.2–3) (0.3–4) (0.6–15) (6–182) (0.3–17)
2) 2) 3)
0.5 0.8 1
Hepatitis A 27 18 1 2 2 32 1 1 5 58 5
(0.07– (0.03– (0.07–
virus (4–77) (3–55) (0.1–4) (0.2–7) (0.2–5) (2–102) (0.2–3) (0.3–4) (0.6–15) (6–182) (0.3–17)
2) 2) 3)
93 104 82 251 165
4 16 19 50 3 8 5 3 50
Bacteria (31– (40– (22– (59– (27–
(2–7) (4–47) (5–65) (16–121) (3–5) (3–39) (3–10) (1–126) (12–124)
259) 277) 241) 696) 490)
2 0.3 0.07 2 4 0.8 0.8 0.7 0.6 0.6
1 23 0.3 4
Brucella spp. (0.2– (0.007– (0.02– (0.2– (0.6– (0.07– (0.004– (0.003– (0.02– (0.09–
(0.3–16) (3–83) (0.07–1) (0.7–35)
53) 18) 0.6) 38) 68) 6) 112) 92) 125) 9)
Listeria 0.3
1 1 3 2 1 1 1 3 0.6 1 1 1 1
mono- (0.2–
(0–21) (0–21) (2–5) (0.2–17) (0–21) (0–21) (0–21) (2–4) (0.3–2) (0–21) (0–21) (0.7–2) (1–4)
cytogenes 0.8)
Myco- 0.03 0.4 0.08 0.1
25 34 2 1 13 0.6 3 11 14 3
bacterium (0.01– (0.2– (0.06– (0.08–
(15–39) (21–48) (0.8–4) (0.5–3) (6–25) (0.5–1) (2–5) (4–27) (6–27) (1–5)
bovis 0.06) 0.8) 0.1) 0.2)
2 0.01 26
Salmonella 11 12 0.1 2 3 10 0.02 0.3 42 0.1 8
(0.006– (0– (0.6–
Paratyphi A (0–39) (0–43) (0–0.4) (0–6) (0–12) (0–36) (0–0.1) (0–2) (7–120) (0–0.5) (1–22)
7) 0.06) 80)
8 184
Salmonella 47 52 0.4 7 14 45 0.09 2 0.04 113 0.6 36
(0.03– (32–
Typhi (0–169) (0–187) (0–2) (0–27) (0–51) (0–158) (0–0.6) (0–9) (0–0.3) (3–347) (0–2) (6–95)
29) 522)
21 20 5 20 27 20 18 6 10 10 13 9 5 9
Protozoa
(8–41) (9–37) (2–8) (9–33) (10–84) (10–35) (9–31) (3–9) (5–23) (5–18) (6–22) (2–19) (3–8) (4–14)
Toxoplasma 21 20 5 20 27 20 18 6 10 10 13 9 5 9
gondii (8–41) (9–37) (2–8) (9–33) (10–84) (10–35) (9–31) (3–9) (5–23) (5–18) (6–22) (2–19) (3–8) (4–14)
186 184 36 185 162
1 5 21 0.4 6 6 52 60 2
Helminths (125– (141– (27– (149– (131–
(0.9–4) (2–15) (12–40) (0.2–1) (3–27) (4–15) (42–64) (45–80) (1–3)
308) 240) 134) 229) 202)
172 178 0.4 0.2 0.03
25 71 1 0.7 4 4 3 46 45
Cestodes (112– (136– (0.3– (0.05– (0.007–
(19–34) (53–95) (0.2–10) (0.1–19) (2–25) (2–12) (2–5) (34–61) (25–65)
289) 235) 0.6) 0.5) 0.8)
0.4 0.01 0.3 0.1 0.02 0.3
Echinococcus 0.8 2 0.9 0.6 2 0.5 0.001 0.8
(0.06– (0.002– (0.02– (0.02– (0.001– (0.08–
granulosus (0.2–16) (0.4–8) (0.2–10) (0.1–19) (0.5–6) (0.09–1) (0–0.1) (0.2–3)
21) 0.03) 5) 0.4) 0.8) 0.9)
0.03 0.005 0.03 0.007 0.008
Echinococcus 0 0 0 0 0 2 2 0 18
(0.005– (0– (0.008– (0– (0.001–
multilocularis (0–0) (0–0) (0–0) (0–0) (0–0) (0.5–21) (0.5–11) (0–0) (0–37)
0.06) 0.05) 0.06) 0.04) 0.02)
170 176 0.4
25 69 0 0 0 0 0.9 3 45 0 27
Taenia solium (110– (134– (0.3–
(19–32) (51–91) (0–0) (0–0) (0–0) (0–0) (0.6–2) (2–5) (33–60) (0–0) (20–35)
283) 229) 0.6)
0.6 0.04 0.004
13 5 11 12 3 13 1 1 8 13 11
Nematodes (0.3– (0.02– (0.001–
(2–28) (1–11) (3–106) (3–24) (1–7) (4–20) (0.3–2) (0.4–2) (2–15) (4–26) (3–22)
0.9) 0.07) 0.007)
0.6
13 5 11 12 3 13 0 1 1 8 13 0 11
Ascaris spp. (0.3–
(2–28) (1–11) (3–106) (3–24) (1–7) (4–20) (0–0) (0.3–2) (0.3–2) (2–15) (4–26) (0–0) (3–22)
0.9)
0.001 0.001 0.009 0.009 0.009 0.04 0.04 0.04 0 0 0.004 0.004
Trichinella 0 0
(0– (0– (0.005– (0.005– (0.005– (0.02– (0.02– (0.02– (0– (0– (0.001– (0.001–
spp. (0–0) (0–0)
0.002) 0.002) 0.01) 0.01) 0.01) 0.07) 0.07) 0.07) 0.001) 0.001) 0.007) 0.007)
Results
80
AFR AFR AMR AMR AMR EMR EMR EUR EUR EUR SEAR SEAR WPR WPR
HAZARD
D E A B D B D A B C B D A B
0.06 0.02 0.2 0.1 0.3 0.2 0.2
101 7 1 40 0.7 2 106
Trematodes (0.02– (0.008– (0.04– (0.04– (0.2– (0.05– (0.05–
(74–135) (4–10) (0.8–1) (32–50) (0.2–2) (1–2) (85–131)
0.2) 0.07) 3) 0.5) 0.5) 0.6) 0.6)
0.04 0.01 0.04 0.05
Clonorchis 0 0 0 0 0 0 0 0 0 31
(0.03– (0.003– (0.01– (0.01–
sinensis (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (26–38)
0.04) 0.04) 0.2) 0.2)
0.02 0.01 0.04 0.04 0.2 0.07 0.06 0.04 0.02 0.05 0.07
46 7 0.9
Fasciola spp. (0.008– (0.005– (0.001– (0.02– (0.1– (0.02– (0.02– (0.01– (0.008– (0.02– (0.01–
(27–75) (4–10) (0.1–8)
0.07) 0.04) 2) 0.1) 0.3) 0.2) 0.2) 0.1) 0.05) 0.1) 0.4)
0.01 0.1 0.06 0.06 0.08 0.03 0.05 0.09 0.2 0.1
Intestinal 0 0 1 9
(0.005– (0.04– (0.02– (0.02– (0.03– (0.009– (0.02– (0.03– (0.1– (0.03–
flukes* (0–0) (0–0) (0.9–2) (7–11)
0.04) 0.5) 0.2) 0.2) 0.2) 0.09) 0.2) 0.2) 0.5) 0.4)
0.07 0.05
Opisthorchis 0 0 0 0 0 0 0 0.9 40 0.4 0 3
(0.02– (0.01–
spp. (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0.6–1) (32–50) (0.1–1) (0–0) (2–4)
0.3) 0.3)
0.03 0.008 0.04 0.04 0.02 0.03 0.05 0.06 0.05
Paragonimus 53 0 0 0 60
(0.008– (0.002– (0.004– (0.01– (0.008– (0.01– (0.008– (0.02– (0.02–
spp. (38–73) (0–0) (0–0) (0–0) (43–83)
0.08) 0.02) 0.6) 0.1) 0.07) 0.1) 0.5) 0.2) 0.2)
2 0.9 0.3
Chemicals 30 7 0.4 3 0.8 9 2 2 20 18 18
(0.09– (0.4– (0.06–
and toxins (8–85) (3–21) (0.2–3) (0.7–16) (0.3–14) (4–66) (1–22) (2–9) (4–75) (13–52) (3–71)
159) 25) 13)
0.04 2 0.3 0.2
28 3 3 0.7 5 0.6 0.5 18 4 17
Aflatoxin (0.006– (0.07– (0.1– (0.04–
(7–78) (1–8) (0.6–9) (0.2–3) (1–17) (0.3–1) (0.2–2) (3–52) (0.6–15) (3–69)
0.2) 137) 0.7) 0.8)
Cassava 1 3 0 0 0 0 0 0 0 0 0 0 0 0
cyanide (0.1–3) (0.3–9) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0) (0–0)
0.2 0.2 0.1 0.2 0.09 0.3 0.2 0.1 0.06
0.3 3 2 2 14
Dioxin (0.05– (0.09– (0.03– (0.01– (0.004– (0.09– (0.005– (0.02– (0.006–
(0.1–3) (2–56) (1–22) (1–8) (12–40)
6) 9) 11) 23) 11) 24) 45) 12) 5)
RESULTS
Notes: * Includes selected species of the families Echinostomatidae, Fasciolidae, Gymnophallidae, Heterophyidae, Nanophyetidae,
Neodiplostomidae and Plagiorchiidae (depending on data availability).
Figure 12. The global burden of foodborne disease (DALYS per 100 000 population) by hazard
groups and by subregion, 2010.
1200
Foodborne DALYs per 100,000 inhabitants
800
400
0
AFR D AFR E AMR A AMR B AMR D EMR B EMR D EUR A EUR B EUR C SEAR B SEAR D WPR A WPR B
Figure 13. Relative contribution of Years of Life Lost due to premature mortality (YLL) and Years Lived
with Disability (YLD) to the global burden of 31 hazards in food, 2010.
1.00
0.75
0.50
Proportion
0.25
0.00
s
n
C
or nt s
a
.
m aty hi
oc A
ET s
EC
yc pa ella us
a s
m a h yan .
lo sto de
ct ica
ST .
on nu EC
Ta chi us
ho so .
Tr rch lium
p.
pl car pp.
.
i
Fa dia p.
st la s .
flu .
Di es
s
fla ae
co lla ella pp
ac is A spp
ra ma pp
p
al pp
di
e
in
Ec alm Salm idiu eric
xi
Cr hoid cyt PE
ar
En Ca ruc bov
us ar Typ
sp
sp
pi nia sp
ne p
ar sp
In sci sp
k
to l S gen
til hi
or los
r
riu Vir
ni gon
i
er
ox
ba lyt
to
He hig ovi
no one on m s
hi is s
s
Pa as is s
ul
p
E
ol
er
As lla
u s
m
ta ss ell
o
r
ch
o
c
o
sp . e
m
A
Cl ra
N
in
oe va
i
io
ob tit
g
ul
Gi
e
o
te
P
go
ic
br
us
a
on
te
b
st
py
Vi
cc
xo
m
O
m
co
cc
To
no eria
yp
p
Ca
no
ty
M
st
n-
hi
S
Li
Ec
hi
YLL YLD
Results
82
Figure 14. Age-distribution of disability adjusted life years for 31 hazards contributing to the
global burden of foodborne disease, 2010.
1.00
0.75
Proportion
0.50
0.25
0.00
.
di pp
ct ica
p.
m C
p.
br a so es
st cho m
av is e
.
Sh l flu s
la s
m As EP .
C
nt ii
pa or ica
A us
lm lm mu us
er ella ella pp.
on at phi
Ta to i A
co no ya .
us chi de
ob ruc ulo .
te lla s
.
m ella is
oc p.
at is
in
Di p
ist pp
no Clo a c spp
pp
co Tr ium pp
u
in oxin
ke
. e nd
a
E
id ETE
m TE
us in bov
Afl lar
ox
sp
sp
til sp
sp
sp
u
n
s
Sa Sa oni Vir
ir
Ca hor ler
s
ni
h
ba lyt
N er
Ty
m a h is s
Li on on s s
gr s s
s
en ge
Pa tit ov
l S go
l
oc p
Gi ola
u
S
a
er
y
o
r
B n
h
el
iu
da a
a
e
ci
oe ca
a
c
ar
ra is
y
ig
i
m ar
r
s
ul
Fa
st
ph as
cc ich
or
i
lo
te
ss
b
g
pl
ac
sp
py
Vi
cc
In
pi
n- oxo
He
oi
to
O
ia
yp
yc
T
Ca
ta
ty
Cr
M
En
st
no
hi
no
Ec
hi
Ec
5+ <5
RESULTS
under five years old bear 40% of the intervals. On the basis of this plot,
foodborne disease burden (including, hazards were divided by two criteria
for some hazards, the life-long burden with arbitrary cut-offs as indicated
of sequelae). More than 75% of the by grey-shaded areas in the Figure. V.
burden of four hazards (Fasciola spp., cholerae, T. solium and Paragonimus
Giardia spp., dioxins, and intestinal spp. were in the H/H category. All other
flukes) occurred among children under diarrheal disease agents were in the H/L
five (Figure 14). Prenatal infections category, except STEC, E. histolytica
accounted for 21% of the burden of and Giardia spp. (L/L). The L/L category
L. monocytogenes and for 32% of further included Trichinella spp. The L/H
the burden of Toxoplasma gondii. By category contained agents that are of
contrast, more than 75% of the burden of relatively low global impact but have
11 hazards occurred among people over a high impact on affected individuals.
five years old. These included different parasites,
particularly E. multilocularis, the invasive
Figure 15 presents a scatterplot of the
bacteria Brucella spp., L. monocytogenes
burden at individual level (DALYs per
and M. bovis. In subregions where the
case, a measure for disease severity)
burden is higher than the global average,
and the burden at population level
these agents are of specific relevance to
(foodborne DALYs per 100,000
policy makers.
population, also accounting for disease
WHO Estimates of the global burden of foodborne diseases
83
Figure 15. Scatterplot of the global burden of foodborne disease per 100 000 population and per
incident case.
100
Emult
A
Afla
Cass
C Clon
Clon
Opis
O
10 Lmono
Parag Tsol
Fasc Flukes
u
Mbov
M
Vchol
V
DALYs per foodborne case
Diox
Egran
1
SPara STyph
Bruc
Trich EPEC
0.1 HAV
V
Toxo
xo
Asc NTS
Cryp
ETEC Camp
Shig
NoV
STEC
0.01
Ehist
0.001 Giar
Abbreviations: NoV = Norovirus; Camp = Campylobacter spp.; EPEC = Enteropathogenic Escherichia coli; ETEC = Enterotoxigenic
E. coli; STEC = Shiga toxin-producing E. coli; NTS = non-typhoidal Salmonella enterica; Shig = Shigella spp.; Vchol; Vibrio cholerae;
Ehist = Entamoeba histolytica; Cryp = Cryptosporidium spp.; Giar = Giardia spp.; HAV = Hepatitis A virus; Bruc = Brucella spp.;
Lmono = Listeria monocytogenes; Mbov = Mycobacterium bovis; SPara = Salmonella Paratyphi A; STyph = Salmonella Typhi; Toxo
= Toxoplasma gondii; Egran = Echinococcus granulosus; Emult = E. multilocularis; Tsol = Taenia solium; Asc = Ascaris spp.; Trich
= Trichinella spp.; Clon = Clonorchis sinensis; Fasc = Fasciola spp.; Flukes = Intestinal flukes; Opis = Opisthorchis spp.; Parag =
Paragonimus spp.; Diox = Dioxin; Afla = Aflatoxin.
Results
84
RESULTS
333 000 chronic infections and 83 300 Overall, the 22 diseases in the enteric
episodes of orchitis. L. monocytogenes diseases study resulted in 351 000
resulted in 14 200 illnesses which (95% UI 240 000–524 000) deaths due
included 7830 cases of septicaemia, to contaminated food in 2010; with 33%
3920 cases of meningitis, and 666 cases (95% UI 27–40%) in children <5 years of
with neurological sequelae. age. The enteric pathogens resulting in
Overall, 29% (95% UI 23–36%) of all the most foodborne deaths were S. Typhi,
22 diseases were estimated to be EPEC, norovirus, iNTS, non-typhoidal
transmitted by contaminated food, Salmonella spp. and hepatitis A. The
equating to 582 million (95% UI mortality rates of foodborne diseases
400–922 million) foodborne cases in were consistently highest in the African
2010; of which 38% (95% UI 24–53%) in subregions, followed by the South
children <5 years of age. The pathogens Eastern Asian subregions (Table A8.2
resulting in the most foodborne cases in Appendix 8) Among the 11 diarrhoeal
were norovirus, Campylobacter spp., diseases due to contaminated food, the
ETEC, non-typhoidal Salmonella spp., rate ratio of deaths in children <5 years
and Shigella spp. A high proportion of age compared with those >5 years of
of foodborne infections caused by V. age was 8.37 (95% UI 5.90–11.4). For all
cholerae, S. Typhi, and S. Paratyphi 22 diseases, the rate ratio of deaths in
A occurred in the African region children <5 years of age compared with
(Table A8.2 in Appendix 8). A high those >5 years of age was 4.85 (95%
UI 3.54–6.59).
WHO Estimates of the global burden of foodborne diseases
85
It was estimated that the 22 diseases in age. Figure 17 shows the relative burden
the enteric diseases study caused 78.7 of foodborne enteric infections, if iNTS
million (95% UI 65.0–97.7 million) DALYS and non-typhoidal Salmonella spp. were
in 2010, of which 43% (95% UI 38–48%) grouped together. The pathogen resulting
in children <5 years of age. The in the most foodborne DALYs was non-
pathogens resulting in the most typhoidal Salmonella spp., if iNTS were
DALYs were norovirus, S. Typhi, EPEC, included (4.07 million). Other pathogens
V. cholerae, ETEC, and hepatitis A resulting in substantial foodborne DALYs
(Table A8.1 in Appendix 8). included: S. Typhi, EPEC, norovirus and
Campylobacter spp. The rates of DALYs
The rates of foodborne DALYs per
for foodborne diseases were highest in
100 000 population are shown by hazard
the African subregions. Overall, the 22
and by subregion in Table A8.2
diseases transmitted by contaminated
in Appendix 8.
food resulted in 10.8 million (95% UI
It was estimated that the 22 diseases 7.59–15.3 million) DALYs in children <5
in the enteric diseases study resulted years of age, compared with 14.3 million
in 25.2 million (95% UI 17.5–37.0 million) (95% UI 9.42–22.5 million) DALYs in those
DALYs due to contaminated food; 43% >5 years of age.
(95% UI 36–50%) in children <5 years of
Figure 16. The global burden of foodborne disease by subregion (DALYS per 100 000
population) caused by enteric hazards, 2010.
1000
Foodborne DALYs per 100,000 inhabitants
750
500
250
0
AFR D AFR E AMR A AMR B AMR D EMR B EMR D EUR A EUR B EUR C SEAR B SEAR D WPR A WPR B
Figure 17. Disability Adjusted Life Years for each pathogen acquired from contaminated food
ranked from lowest to highest with 95% Uncertainty Intervals, 2010.
10,000,000
1,000,000
Global foodborne DALYs
100,000
10,000
EC
p.
es
p.
p.
p.
ae
EC
p.
EC
hi
al
ph
vi
iru
ic
id
p
sp
sp
sp
sp
sp
en
is
er
bo
yt
ST
ET
EP
ty
Ty
ho
ov
tit
ol
og
a
la
er
ol
ra
m
pa
di
or
la
yp
ch
el
el
st
iu
ct
Pa
yt
el
riu
ar
N
uc
ig
hi
He
id
-t
ba
io
oc
on
Gi
Sh
la
on
or
te
Br
ba
br
lo
on
el
lm
ac
sp
,n
py
Vi
oe
on
m
Sa
ob
to
la
m
m
lm
ia
yp
el
yc
Ca
ta
er
Sa
on
Cr
M
En
st
lm
Li
Sa
Note figure is on a logarithmic scale. The figure shows the median (white dot); Inter-Quartile Range = 50% UI = 25%/75% percentiles
RESULTS
(thick black line); 90% UI = 5%/95% percentiles (thin black line); 95% UI = 2.5%/97.5% percentiles (thin grey line). Note, four
foodborne intoxications due to Clostridium botulinum, Cl. perfringens, S. aureus, and Bacillus cereus due to a lack of data for
global estimation. In addition, data for non-typhoidal Salmonella enterica infections and invasive non-typhoidal S. enterica have
been combined.
Figure 18A. The relative contribution to the DALY incidence by each agent for each of the
subregions. This includes enteric protozoa to complete the picture on foodborne parasitic
diseases. However the detail is reported in the accompanying manuscript on foodborne enteric
pathogens [168].
200
Foodborne DALYs per 100,000 inhabitants
150
100
50
0
AFR D AFR E AMR A AMR B AMR D EMR B EMR D EUR A EUR B EUR C SEAR B SEAR D WPR A WPR B
Figure 18B. Disability Adjusted Life Years for each parasite acquired from contaminated food
ranked from lowest to highest with 95% Uncertainty Intervals, 2010. This includes enteric
protozoa to complete the picture on foodborne parasitic diseases. However the detail is reported
in the accompanying manuscript on foodborne enteric pathogens [168].
10,000,000
Foodborne Disability-Adjusted Life Years
1,000,000
100,000
10,000
1,000
100
is
sis
is
is
is
sis
is
is
sis
sis
sis
is
sis
is
sis
os
os
os
os
os
os
os
os
llo
bo
io
co
io
io
co
cc
od
di
ol
hi
sm
m
id
ch
ar
ne
er
oc
oe
rc
ar
ci
as
ni
co
or
sc
at
r
la
ic
hi
oc
go
ho
no
Gi
pl
Fa
p
m
op
no
st
ic
ta
os
xo
in
st
lo
ra
Cy
tre
Tr
ox
hi
En
ch
pi
C
t
Pa
to
yp
ec
lt
al
re
d
Cr
ita
in
ic
ire
la
st
st
en
eo
qu
te
Cy
ng
lv
In
RESULTS
Ac
A
Co
Figure 19. The relative proportion of the burden of each of the foodborne parasitic diseases
contributed by YLLs and YLDs
1.00
0.75
0.50
Proportion
0.25
0.00
is
is
sis
is
sis
is
sis
sis
is
sis
is
sis
is
is
sis
os
os
os
os
os
os
os
os
io
bo
io
co
lo
io
lo
od
di
cc
cc
hi
sm
m
rid
ch
el
ar
io
ar
er
oe
at
rc
ni
as
co
co
in
sc
sc
or
Gi
la
ic
po
m
go
ho
m
ch
pl
Fa
op
no
no
A
st
on
tre
ta
s
xo
st
ra
i
Cy
to
Tr
ox
i
hi
Cl
En
ch
pi
Pa
to
yp
al
ec
lt
O
re
in
d
Cr
ita
ic
ire
st
la
st
en
te
eo
qu
Cy
In
ng
lv
Ac
A
Co
YLL YLD
WHO Estimates of the global burden of foodborne diseases
89
Figure 20. The relative contribution to the DALY incidence by each of four chemicals for each of
the WHO Regions.
30
Foodborne DALYs per 100,000 inhabitants
20
10
0
AFR D AFR E AMR A AMR B AMR D EMR B EMR D EUR A EUR B EUR C SEAR B SEAR D WPR A WPR B
Notes: Peanut allergy burden was estimated only for the AMR A (United States of America, Canada and Cuba); EUR A (primarily
countries in western Europe); and WPR A (Australia, Brunei Darussalam, Japan and New Zealand) subregions. Burden estimates for
RESULTS
cyanide in cassava are provided only for the African region (AFR), and assumed to be zero for other regions.
Figure 21. The relative contributions from YLLs and YLDs for each of four chemicals.
1.00
0.75
0.50
Proportion
0.25
0.00
e
ut
in
n
id
i
ox
ox
an
an
at
Di
Pe
cy
Afl
a
av
ss
Ca
YLL YLD
WHO Estimates of the global burden of foodborne diseases
91
Figure 22. Disability Adjusted Life Years for each of four chemicals from contaminated food
ranked, from lowest to highest, with 95% uncertainty intervals
10,000,000
Foodborne Disability-Adjusted Life Years
1,000,000
100,000
10,000
1,000
100
e
ut
in
id
i
an
ox
ox
an
Pe
at
Di
cy
Afl
a
av
ss
Ca
Notes: The dot in the middle of each box represents the median, the box the 50% uncertainty interval, the dark bar the 95%
uncertainty interval, and the light bar the 95% uncertainty interval.
Results
92
RESULTS
WHO Estimates of the global burden of foodborne diseases
93
DISCUSSION
6
WHO Estimates of the global burden of foodborne diseases
95
DISCUSSION
This study estimates that 31 foodborne in the FERG estimates whereas in the
hazards resulted in 33 million DALYs GBD 2010 study they were recorded in
in 2010, which shows the considerable different disease categories. Furthermore,
public health impact of contaminated the GBD 2010 study used a different life
food. Importantly, children <5 years of table than FERG and more extensive
age experienced 40% of the foodborne mathematical modeling to account for
disease burden, despite representing only data gaps, which smoothed the data
9% of the global population. The study considerably, resulting in narrower
provides a substantial expansion of the uncertainty intervals than in our study.
available data on the public health impact The GBD 2010 and FERG studies used
of FBDs. the same set of disability weights,
but the FERG included some updates
Several high-income countries have
as recommended by WHO. Neither
published national estimates of FBD.
study applied time discounting or age-
Estimates of food-related illnesses and
weighting in their baseline estimates.
deaths in the USA were estimated in the
late 1990s [169] and updated to cover The GBD 2010 study, which looked at
the period 2000–2008 [170, 171]. Similar all sources of disease, found that the
studies are available from Australia [173], key hazards and risk factors for disease
Canada [175], France [174] and the UK burden were dietary risk factors (254
[172]. Some countries have extended this million DALYs), unimproved water and
work to estimate DALYs, including Greece sanitation (211 million DALYs), HIV/AIDS
[177], the Netherlands [178], New Zealand (82 million DALYs), malaria (82 million
[176] and the USA [179]. While the range DALYs), air pollution (76 million DALYs)
of hazards covered in these studies and tuberculosis (49 million DALYs).
differed from those in the FERG studies, Recently published findings from WHO1
the focus was on enteric diseases and a for 2012 were: HIV/AIDS (92million
limited number of invasive and parasitic DALYs); malaria (55 million DALYs) and
diseases. The FERG data, by contrast, tuberculosis (44 million DALYs). Hence,
cover numerous countries across the the burden of FBD (33 million DALYs)
globe and provide a more complete is of a similar order of magnitude as
picture of FBD. the ‘big three’ infectious diseases (HIV/
Comparisons of the FERG estimate of AIDS, malaria and tuberculosis) and
the burden of FBD with other estimates, air pollution, but clearly lower than
such as those of the Institute for Health the burden of dietary risk factors or
Metrics and Evaluation’s GBD 2010 study unimproved water and sanitation.
[81], must be made with care because The FERG estimate of 29,000 deaths due
of differences in the methodology and to foodborne transmission of invasive
data used. For example, the GBD 2010 non-typhoidal S. enterica only included
study used prevalence-based DALYS, infections in non-HIV infected individuals.
whereas our study used incidence-based Ao et al. [180] estimated there were
DALYs. As a consequence, the impact of approximately 680,000 deaths due to
sequelae such as Guillain-Barré syndrome invasive non-typhoidal salmonellosis in
(due to Campylobacter spp.), hemolytic 2010. Of these, approximately 350,000
uremic syndrome (due to Shiga toxin- would be due to foodborne transmission,
producing E. coli) and invasive disease assuming 52% of all non-typhoidal
(due to non-typhoidal S. enterica) were 1
http://apps.who.int/gho/data/node.main.
attributed to the diarrheal disease agents DALYNUMWORLD?lang=en; accessed July 22, 2014
Discussion
96
DISCUSSION
For some hazards (e.g. M. bovis and cause other adverse health effects than
E. multilocularis, aflatoxin and dioxin), the ones considered (e.g. diarrhoeal
incident illness is related to past diseases [185] and aflatoxin as causes
exposures due to long incubation of malnutrition and stunting, dioxin and
times of disease. For such hazards, the immune effects or cancer), for which
estimated burden reflects exposure data were not available to allow disease
dating back to the average incubation burden estimates.
period of the disease rather than current A further limitation of this study is that
exposure. For some hazards (e.g. DALYs do not quantify the full societal
dioxins), the impact on the child depends impact of FBD. The economic burden
on the lifelong exposure of the mother. (cost-of-illness, losses in the agricultural
The FERG estimates of the FBD and food sectors and trade impacts)
burden are probably conservative, is also an important factor to consider
i.e. underestimates rather than in national and international decision-
overestimates. Limited resources and making. Also, the process of food
data obliged us to focus on only a subset production can cause human diseases
of more than 100 hazards of potential by mechanisms other than direct
WHO Estimates of the global burden of foodborne diseases
97
DISCUSSION
several pathogens, there was a clear
and– when combined with estimates pattern that the experts considered the
of incidence, severity, duration and foodborne route less important in low-
mortality– allows estimation of the and middle-income subregions, where
burden of foodborne disease. Attempting other routes (animal contact, water
to estimate foodborne transmission at and soil) were believed to contribute
the subregional level is an ambitious relatively more in comparison with high-
goal. However, this was vital given income subregions. This is consistent
the geographically localized nature of with data showing lower levels of access
exposure to many pathogens. The results to improved water and sanitation in
are significant due to the global nature less developed regions compared
of the estimation, the number of experts with high-income countries. This
participating, and the rigorous approach ranking of subregions across different
taken to assessing and including expert pathogens provides some confidence
performance in the final estimates. in the results, as the estimates were
2
http://www.who.int/campaigns/world-health- done independently and partly by
day/2015/en/ different experts.
WHO Estimates of the global burden of foodborne diseases
99
contaminated water sources and poor food or water depending on the point-
hygiene in developing regions [27], so of-attribution. It is clear that there exists
these transmission routes are likely to no single “right” way of delineating the
be relatively more important among foodborne route from other transmission
cases that have been travelling to these routes; however, it is critical that point of
regions. This is probably what is reflected attribution be clearly defined. Definition
in the attribution results in this study. of point-of-attribution should depend on
the objective and focus of the specific
The operational definition of different
study carried out, and could involve many
transmission routes, in particular the
factors, including the foodborne hazard
food and waterborne routes, will
in question, the food production systems
affect attribution estimates. Hazards
and routines in place, the geographical
transmitted by multiple routes can
occurrence of the hazard, sanitation and
“change” source or vehicle during the
hygiene in the region, and consumption
transmission from primary source to
patterns. If point of attribution is clearly
humans, meaning that the burden of
defined, then additional modelling or
illness caused by a particular hazard
further research can be used to adjust
attributed to a specific transmission route
attribution to exposure at other points of
may vary, depending on the point-of-
interest in the transmission chain. If the
attribution chosen [7, 21]. The choice of
point of attribution is not unambiguously
point-of-attribution seems particularly
defined, then not only are the results of
critical for delineating foodborne and
the study unclear, but it will be difficult to
waterborne diseases. This is because
use them to model other relationships in
water in itself is ingested just as a food,
the transmission chain.
is used for irrigation of food plants, for
washing and cleaning of food during FERG agreed that the point of human
preparation and constitutes an essential exposure was the most simple and
ingredient in many food products. In understandable point-of-attribution to
addition and particularly related to be used across all hazards for delineating
zoonotic diseases, the water source the major transmission routes (Figure 23).
itself is often contaminated by an animal FERG recognizes that for other purposes,
reservoir, including food-producing e.g. for risk management, other points
DISCUSSION
animals. Another situation relates to of attribution may be more appropriate
the consumption of water-based foods (e.g. primary production, processing and
such as shellfish harvested from areas retail, or preparation). Yet, the FERG’s
where the water is contaminated with definition of point of attribution for
pathogenic organisms, such as Vibrio major transmission routes directly links
spp. or enteric viruses. The burden attribution to disease incidence, as it is
related to the consumption of foods that the end of the transmission chain. Further
have had contact with contaminated modelling can then be used to work
water at some stage of the production backward from exposure to identify the
may, therefore, be attributed to either important points of contamination.
WHO Estimates of the global burden of foodborne diseases
101
Figure 23. Major transmission routes of human foodborne diseases indicate two points of
attribution: the reservoir level and the exposure level.
Processing Processing
Preparation
Consumption
Foodborne
Human-
human
DISCUSSION
brucellosis and M. bovis infections were our attention on the burden due to
highest in the Middle Eastern and African pathogens that were known to be
regions. transmitted by contaminated food.
To develop these comprehensive “When we examined the human health
estimates of the disease burden of impact of different pathogens, various
foodborne diseases, we adopted an serotypes of Salmonella resulted in
innovative approach to incorporate the greatest foodborne burden. If
the highest quality data available for we consider the combined burden
each foodborne disease [40]. Due to attributable to S. enterica from all
their quality, we gave highest priority invasive (including iNTS, Salmonella
to studies with national estimates of Typhi and Salmonella Paratyphi A)
foodborne diseases. Since studies with and diarrheal infections, there were
national estimates were only available in an estimated 21.2 million (95% UI 12.8–
a few countries, we adapted the CHERG 35.8 million) DALYs from all transmission
approach for estimating the disease sources and 8.76 million (95% UI 5.01–
burden of diarrhoeal diseases [50, 51]. 15.6 million) DALYs from contaminated
This approach was facilitated by the food. This highlights the significant
WHO Estimates of the global burden of foodborne diseases
103
DISCUSSION
be averted by improvements in food
safety. For some other pathogens in our diseases [40, 205].
estimates, such as L. monocytogenes
and Cryptosporidium spp., due to a 6.3 Parasites
lack of reliable data we were unable to
In this study, we estimate for the first
account for the excess mortality due to
time the disease burden imposed by
HIV infection.
foodborne parasites. The results highlight
Another important limitation of our the significant burden in low- and
attempt to quantify the disease burden middle-income countries, where cycles
due to foodborne disease is the inherent of parasitic infection are highly specific
difficulty in estimating the proportion of to food sources. In addition to those
illness acquired from food [7]. We relied detailed here, a further 357 million cases,
on a structured expert elicitation study. 33 900 deaths and 2.94 million DALYs
We were unable to estimate differences are due to enteric protozoa, of which 67.2
in mode of transmission by age, despite million cases, 5560 deaths and 492 000
this potentially being important. Expert DALYs are attributable to foodborne
elicitation studies can result in highly transmission (see [168] and Table 1),
WHO Estimates of the global burden of foodborne diseases
105
completing the picture for the foodborne in the present study. The former study
parasitic diseases, given data available. [206] also undertook age weighting and
discounting, which we decided not to
We used the best evidence available
incorporate into this study. In addition
combined with the natural history of
different life tables were used. Our use
the disease to obtain estimates of the
of DWs was guided by GBD2010 and
incidence, mortality and sequelae of
the results of a systematic review of
each parasitic disease. Several of the
the clinical manifestations of CE [75].
diseases were included in GBD2010 [81].
However, a median estimate in excess
In a number of cases our estimates for
of 188 000 cases of CE per year, with
the global burden of disease differ quite
the possibility of up to 1.77 million new
substantially from those of GBD2010.
cases, indicates a substantial burden.
The estimate for echinococcosis (which
With a low case-fatality rate the burden
combined AE and CE in one estimate)
in terms of DALYs is highly dependent on
in GBD2010 is 144 000 DALYs [81].
the DW and duration of illness. Neither of
This is less than a fifth of our combined
these is defined with certainty. The lack
median estimate of 871 000 DALYs. This
of defined DWs specific for the differing
discrepancy probably reflects different
sequelae of CE must be seen as a major
methodologies between the two studies.
data gap.
GBD2010 relied heavily on vital records
for mortality attributed to these diseases, When arriving at the estimates for AE,
whereas we used an approach based it was assumed that in excess of 90%
on the natural history of the disease. of cases outside of Europe would be
Our choice of approach was strongly fatal. This assumption was supported by
influenced by the chronic nature of these survival analyses, confirming that in the
diseases, and that often only prevalence absence of aggressive treatment of this
data were available. In addition, these disease, including chemotherapy, most
diseases often have their highest impact cases die [207, 208]. Our results suggest
in low income countries, where vital it is possible that the global burden of AE
records are likely to be poor and hospital may be somewhat higher than that of CE,
treatment unavailable. Our estimates for which may at first sight seem surprising
the global burden of CE would arguably as there are many more cases of CE
be more consistent with an earlier globally and the parasite has a more
estimate [206], if there had been no cosmopolitan distribution. Hence, we
substantial methodological differences. have a median estimate of CE incidence
The earlier report suggested a median that is ten times higher than the median
estimate of 285 000 DALYs assuming no estimate of AE incidence. The high case
under-reporting, rising to 1 million DALYs fatality ratio of AE, results in the loss
where under-reporting was assumed. The of 37 DALYs per case compared with
earlier report also used DWs ranging from 0.98 DALYs for each case of CE. Thus
0.2 to 0.809, depending on the severity the global burden of AE was driven by
of the disease. In the present study we the large number of YLLs. For CE it was
used a maximum DW of 0.221, and that driven by the YLDs.
was only applied to the relatively small
Our estimates for cysticercosis were
number of neurological cases.
higher than that of GBD2010, as a result
Echinococcosis of the abdominal organs of assigning a substantial proportion
– the most common presentation – had a of epilepsy burden to cysticercosis,
DW of 0.123 for treatment-seeking cases based on the results of a systematic
Discussion
106
review [74]. Furthermore, a subsequent and where human food habits suggest
systematic review has largely confirmed transmission to humans to be likely. We
our findings in terms of the fraction tried to correct for this lack of data by
of epilepsy attributable to NCC [209]. imputing incidence rates for all countries
However, our results are not inconsistent with at least one autochthonous human
with GBD2010 [81] because we have infection reported in the reviewed
allocated some of the burden from literature. Nevertheless, and in line with
epilepsy to a specific aetiological agent. the original study [78], very conservative
Nevertheless, the present estimate in this estimates from the imputation were
report may still underestimate the burden accepted in an attempt to avoid inflating
of cysticercosis, as there are other the burden estimates for human
important clinical symptoms associated foodborne trematode infections based on
with neurocysticercosis, such as chronic unclear evidence.
headache, hydrocephalus, stroke and
Some diseases, such as toxoplasmosis,
depressive disorders [73]. Better
were not estimated in GBD2010 and
estimates of the role that cysticercosis
will inevitably have been included
plays in stroke and depressive
in other syndromes. For example,
disorders globally could considerably
congenital defects in GBD2010 will have
increase its burden estimates, since
incorporated the DALYs for congenital
these conditions are ranked third and
toxoplasmosis that we have estimated in
eleventh, respectively, in the GBD2010
the present study.
[81] estimates. It is also unclear how
GBD2010 arrived at their estimates It can be argued that congenital
for cysticercosis. If, for example, it was toxoplasmosis is a vertically transmitted
assumed that cysticercosis-related disease rather than foodborne. However
epilepsy can only be attributed in public health measures are largely
individuals who are serologically undertaken to prevent maternal (i.e.
positive for cysticercosis, this would horizontal) infection, which will, as a
lead to substantive underestimates. A consequence, reduce the risk of foetal
large proportion of cases of epilepsy infection. There is relatively little evidence
attributed to cysticercosis, as shown that treatment to prevent vertical
DISCUSSION
by imaging studies, are nevertheless transmission (such as antiprotozoal
sero-negative. For example Montano treatment of acutely infected pregnant
et al. [210] describes 15 cases of women) is effective in reducing disease
epilepsy aetiologically confirmed as burden [211].Thus it was considered a
neurocysticercosis, but only 7 of these horizontally transmitted infection to the
were sero-positive. mother, although the burden of disease is
suffered mostly by the foetus, following
Likewise, the estimates for the burden of
subsequent vertical transmission.
foodborne trematode infections may also
Accordingly the proportion of foodborne
represent underestimates. Our estimates
disease suffered by the foetus is the
were based on the results of an earlier
proportion of the horizontal transmission
study, which used estimation methods
to susceptible women that occurs
that were conservative [78]. Often,
through food.
population-level information on human
foodborne trematode infections were With the exception of NCC, we have used
completely lacking from areas where an incidence approach to estimating
the parasites are endemic, as indicated the YLDs. This is where the YLD part of
by substantial rates of animal infections the DALY was estimated from number
WHO Estimates of the global burden of foodborne diseases
107
of incident cases per year multiplied by methods to deal with data deficiencies.
the DW and duration. This is in contrast Until these are published we will only be
to the GBD2010 approach, which used able to hypothesize the reasons for some
a prevalence approach to YLDs, where of the differences in the estimates.
YLDs were estimated by number of
The limitations in this study are similar
prevalent cases multiplied by the DW.
to others in this series. There were often
For acute disease in generally stable
substantial data gaps that had to be
epidemiological situations (i.e. no
filled by imputation and suffer from
considerable shifts in the epidemiological
the uncertainties that surround such
key indicators of prevalence, incidence,
models. Excluding stillbirths is consistent
duration, severity, remission and
with the approach used to estimate the
mortality) and settings with more or
burden due to enteric pathogens [168].
less stable population size, the approach
Congenital toxoplasmosis is the only
makes little difference [2]. But for chronic
pathogen investigated that could result
diseases in populations that are rapidly
in a substantial incidence of stillbirths.
increasing, the prevalence approach may
However, an estimate for the burden
underestimate the numbers of YLDs.
of congenital toxoplasmosis, which
Parasitic diseases are often chronic
includes stillbirths as equivalent to
and are often of highest incidence in
neonatal deaths, has been reported as
low income countries with increasing
1.2 million DALYs per annum [76]. FERG
populations. Many parasitic diseases
has assumed that acquired toxoplasmosis
have durations of many years, or in
usually results in a relatively mild acute
the case of congenital toxoplasmosis,
illness, with some cases suffering
the sequelae are usually lifelong. Thus,
fatigue for a few months [212]. Although
as we adopted the GBD2010 data for
fatal cases have been recorded [213],
epilepsy to estimate the burden of NCC,
these were assumed to be uncommon
the YLDs will be prevalence-based.
and hence zero YLLs were estimated.
Nearly all of the burden of NCC is in
We have also assumed that although
low income countries, which usually
acquired chorioretinitis occurs following
have increasing populations. Therefore
toxoplasmosis, it only occurs in a small
the cohort at the time of infection, with
proportion of cases (see Appendix 4).
the burden attributed in an incidence-
This results in approximately 1.15 million
based approach, will be larger than
DALYs in 2010 from an estimated 20.7
earlier cohorts that are still affected by
million people having clinical disease
NCC but are reported in the prevalence-
following exposure to the pathogen for
based approach. Accepting this
the first time. However, there is increasing
limitation means that the estimates for
evidence that acquired toxoplasmosis
epilepsy attributed to NCC will result in
may result in a number of neurological
a further under-estimate of the burden
or psychiatric diseases, such as
of cysticercosis.
schizophrenia and epilepsy. In GBD2010
We have summarized the differences these diseases resulted in 15.0 million
between the estimates for GBD2010 and and 17.4 million DALYs, respectively.
the FERG estimates for these pathogens, From two meta-analyses [214, 215] and
including the enteric protozoa in Table 9. a large cross-sectional study conducted
In addition, an issue that appears in China [216], it is possible to estimate
common to many hazards is that that the population-attributable
GBD2010 [9] has not published many of fraction of schizophrenia associated
the search strategies used, or modelling with seropositivity to toxoplasmosis is
Discussion
108
approximately 9%, which on a crude detect and remove the parasite from the
level could account for approximately 1.3 food chain [217]. Likewise, other cestode
million additional DALYs. zoonoses, where the adult tapeworm is
located in the gastrointestinal tract (e.g.
There were also some notable omissions
Diphyllobothrium spp.) with few clinical
from our study. Taenia saginata,
signs, were also excluded. In contrast,
which causes human taeniosis and
trichinellosis was considered to be an
is transmitted solely from beef, was
important foodborne pathogen with
not considered because the parasite
potentially serious disease. However,
produces very mild, unapparent clinical
this study has suggested that the global
disease in affected humans, which would
burden of trichinellosis is small. This is
result in a DW of close to zero and hence
discussed elsewhere [84]. For reasons
a very low burden of human disease.
of resource limitations, we were not able
However, it is accepted that this parasite
to consider foodborne Chagas disease,
generates substantial economic damage
although it was suggested as a possible
because of meat inspection and trade
priority pathogen during the second
regulations required in many countries to
FERG meeting.
Table 9. Comparisons of the total burden of parasitic diseases (foodborne and non-foodborne)
with 95% uncertainty intervals, estimated by FERG and by GBD2010 [9]
DISCUSSION
183 573 GBD2010 used vital records, which are often missing in low
Echinococcus
(88 082– resource countries. FERG used a natural history approach based
granulosus 152 000 1 590 846) on surveillance data. GBD2010 used prevalence-based YLDs, which
(60 000–
687 823 will underestimate burden for a chronic disease like echinococcosis.
Echinococcus 359 000)
(409 190– Methods for imputation of missing data were different. GBD2010 has
multilocularis not published their modelling methods for missing data.
1 106 320)
GBD2010 used vital records relying on a diagnosis of cysticercosis.
514 000 2 788 426 FERG assigned a substantial proportion of the epilepsy envelope to
Taenia solium (398 000– (2 137 613– cysticercosis in resource-poor, pork-consuming communities, based
650 000) 3 606 582) on evidence from a systematic review and meta-analysis. GBD2010
has not published their modelling methods for missing data.
1 315 000 1 317 535 Only subtle differences as FERG and GBD2010 used the same source
Ascaris spp. (713 000– (1 182 187– data, but FERG estimated incidence-based YLDs whereas GBD2010
2 349 000) 2 700 572) used prevalence-based.
550
Trichinella spp Not estimated
(285–934)
1 875 000 2 024 592 Only subtle differences as FERG and GBD2010 used the same source
Foodborne
(708 000– (1 652 243– data, but FERG estimated incidence-based YLDs whereas GBD2010
Trematodes
4 837 000) 2 483 514) used prevalence-based.
DISCUSSION
species of molluscs. This limits their were health-related criteria and 3 non-
distribution to specific regions where health criteria. This weighting of the
suitable life-cycle hosts are endemic, different criteria may be responsible for
which may be adapted to specific the FAO/WHO report having a different
climatic and hydrological conditions ranking order for the various parasites.
[227]. The human disease is further For example, E. granulosus has a global
limited to populations that are likely to distribution and a relatively important
consume the raw fish or undercooked measure in the FAO/WHO ranking. In
aquatic vegetables that are the sources contrast, E. multilocularis is only found in
of transmission. Consequently, although the northern hemisphere.
we are reporting the global burden of
these parasitic diseases, this is often
borne almost completely by relatively 6.4 Chemicals
small populations in limited geographical The assessment of the burden of disease
areas. Therefore, in such communities, from chemicals in food is a challenge
these diseases have a major impact on on several levels. There are thousands
the health of the population. of chemicals in production and many
WHO Estimates of the global burden of foodborne diseases
111
naturally occurring toxins. How many of the disease that is attributable to the
of these chemicals and toxins make it exposure. A probabilistic version of this
into the food supply is unknown. The method, which is applied in chemical risk
health effects of chemicals may not be assessment, was used for dioxin [127, 128].
observed for years following exposure
The two approaches would result in
(e.g. aflatoxin and liver cancer; lead and
the same outcome if perfect data were
cardiovascular disease). Longitudinal
available, and if it can be assumed that
studies of these effects are expensive and
the risk of exposure to a chemical is
time-consuming. Sufficient information is
additive to the background risk from
available, however, to make estimates of
other causes. In reality, the available data
the burden for arsenic, cadmium, methyl
for both approaches are limited and
mercury and lead, and possibly for other
there is insufficient information to decide
chemicals and toxins (e.g. fish toxins,
conclusively whether risks are additive,
aristolochic acid). Other chemicals (e.g.
multiplicative or otherwise. This may result
Persistent Organic Pollutants) may not
in considerable discrepancies between
require elaborate epidemiological studies
results from these methods. In this study,
because the burden can be derived from
bio-monitoring data in combination with FERG chose a “top-down” approach for
relevant toxicity data. Estimates of the aflatoxin because the cancer potency
burden for these chemicals will provide a factor derived by JECFA [111] was based
much more comprehensive understanding on a multiplicative model, and there is
of the impact that chemicals in the food evidence for a high background rate in the
supply have on the burden of disease. study population underlying this estimate
and the global population (see Appendix
As the relevant disease endpoints due 4). Using the population-attributable
to foodborne chemicals may arise from fraction approach, it was estimated
different causes, various approaches there were approximately 22 000 (95%
are possible for estimating incidence UI 9 000–57 000) cases of aflatoxin-
and mortality. A “top-down” approach related HCC in 2010. A dose-response
uses an existing estimate of morbidity approach [110] estimated that, annually,
or mortality of the disease endpoint by 25 200–155 000 cases of HCC might
all causes (the “envelope”) as a starting be attributable to aflatoxin exposure.
point. A population-attributable fraction Even though the uncertainty intervals
is then calculated for the hazard under overlap, the differences between these
consideration, and applied to the envelope two approaches are considerable. There
to estimate the hazard-specific incidence. is evidence for a high background rate
This method, which is the standard in in the study population underlying this
Global Burden of Disease estimations, was estimate and the global population (see
used for aflatoxin. Appendix 4), which may result in over-
A “bottom-up” or dose-response approach estimation of mortality by the dose-
uses dose-response and exposure response approach. In contrast, the
information. The approach begins with global liver cancer envelope may be
selection of the appropriate dose-response underestimated, particularly in Africa
relationship between the chemical and [228, 229], leading to underestimation
the particular disease. This dose-response of the aflatoxin-attributable incidence.
relationship is then combined with the Hence, there is considerable data and
distribution of exposure within a population model uncertainty in our estimates, which
to derive an estimate of the incidence should be addressed by further studies.
Discussion
112
DISCUSSION
WHO Estimates of the global burden of foodborne diseases
113
COUNTRY
7
STUDIES
WHO Estimates of the global burden of foodborne diseases
115
COUNTRY STUDIES
Building on the FERG framework, the In addition, the prevalence of liver fluke
policy situation analysis provided an infection (Opisthorchis viverrini, a locally
overview of the food safety policies and important foodborne hazard transmitted
systems in Japan. As a Japan-specific via fish) and the incidence of rotavirus
issue, a rigorous policy situation analysis infection have been estimated.
of the management of risks associated Food safety regulatory activity in
with possible radioactive substances in Thailand is led by the Bureau of Food and
food, due to the nuclear power plant Water Sanitation, Department of Health,
accident in Fukushima after the Great Ministry of Public Health. The popularity
East Japan Earthquake in 2011, was of street food has led to the development
also completed. of a sanitation standard for vendors.
7.4.3 Thailand 7.4.4 Uganda
The Thai country study focused on The Ugandan country study established
the incidence of diarrhoeal disease, teams to separately address enteric,
using data from the National Notifiable parasitic and chemical hazards, and
Disease Surveillance System maintained source attribution [240]. A detailed
by the Bureau of Epidemiology of situation analysis was prepared, which
the Thai Ministry of Public Health.
described the context for food safety
COUNTRY
STUDIES
a situation analysis (for an example see that they address risk management
[243]), and early and continuous efforts questions. Such questions are best
formulated and delivered by policy-
to recognize and incorporate knowledge
makers at the commencement of
translation and risk communication to the research, but researchers should
audiences identified in the situation always expect to address questions
analysis. Differences in experience and of effectiveness, cost, and high
perspectives can make collaboration risk groups.
between the social scientists and
7.6 Discussion
epidemiological/food safety technical
COUNTRY
STUDIES
identified challenges in both process and a situation analysis (such as the existing
information, including the recognition national food control system). In addition,
that data collection and analysis, the WHO initiative sought to foster the
development of situation analysis, and knowledge translation of burden of
on-going knowledge translation and risk disease data into policy through on-
communication, require commitment of going cross-agency communication.
time and financial resources. Such activities are best undertaken by
The WHO initiative has provided burden people from within a country.
of foodborne disease estimates from National burden of foodborne disease
global and regional perspectives. These studies, particularly in developing
estimates provide context and can fill countries, now have an opportunity to
many of the data gaps for individual fill data gaps, and assign aetiology and
countries undertaking foodborne attribution to the incidence of foodborne
burden-of disease studies. In particular, diseases, using the data from the WHO
the provision of aetiology estimates initiative to augment local data. Such
for syndromic surveillance data, and local data can also be used as a cross
attribution estimates for foodborne check to validate national estimates
disease, will be particularly difficult derived from regional estimates. This
for studies in developing countries to should allow the generation of at least
address individually. preliminary burden estimates to inform
The Global Burden of Disease 2010 Study national policy. The effective delivery of
(GBD2010), undertaken by IHME, Seattle, this information can be guided by the
USA, covers a broad range of disease considerations and tools provided in the
and injuries, and has published country- SA/KT/RC Manual. In the longer term,
specific estimates for these on its website burden of foodborne disease information
[245]. Foodborne diseases are a subset should be a fundamental component of
of these estimates, although estimates a systematic approach to food safety,
are typically not stratified by transmission such as the risk management framework
route. National foodborne disease studies advocated by Codex [246]. Such an
as promoted by WHO and FERG include approach can enhance both public health
consideration of the national context in and trade.
Country studies
122
COUNTRY
STUDIES
WHO Estimates of the global burden of foodborne diseases
123
8
CONCLUSION
WHO Estimates of the global burden of foodborne diseases
125
CONCLUSION
This report presents the first global army of more than a hundred scientists,
and regional estimates of the burden specialized in their own fields, it turned
of foodborne diseases. The large out to be possible to present the first
disease burden from food highlights the ever estimates of the global burden of
importance of food safety, particularly in foodborne disease. The process took
Africa, South-East Asia and other more eight years and an uncounted number
greatly affected regions. Our results of hours. All involved donated their time
indicate that some hazards, such as and experience to WHO, finding own
non-typhoidal S. enterica, are important sources of funding in addition to the
causes of FBD in all regions of the limited means available and invested
world, while others – such as certain liberal amounts of personal time. In
parasitic helminths and aflatoxin – are particular the Core Group (Task Force
of highly focal nature resulting in high chairs and senior advisers) spent their
local burden. time in numerous teleconferences at
Despite the data gaps and limitations of sometimes highly inconvenient hours,
these initial estimates, it is apparent that in particular for the colleagues from
the global burden of FBD is considerable, Australia and New Zealand. Initially
and affects individuals of all ages, but annual meetings were organized, creating
particularly children <5 years of age and momentum and commitment. The global
persons living in low-income regions financial crisis inevitably hit FERG, and
of the world. By incorporating these much more reliance was placed on
estimates into policy development at teleconferences and other means of
both national and international levels, remote communication, slowing down
all stakeholders can contribute to the process and limiting the involvement
improvements in safety throughout the to the Core Group mainly. Nevertheless,
food chain. These results will also help to all FERG members and resource advisers
direct future research activities. continued to believe in and support
the Initiative.
and on the number of Years of Life Lost needs for burden of illness estimates are
(YLL) due to premature mortality. Many high, and crucial information was often
foodborne hazards are not exclusively lacking, particularly for some of the
transmitted by food, and a separate world’s most populous countries, such
effort was set up for the attribution of as China, India and Russia. FERG used
exposure to different sources, including statistical models and expert input to
food, the environment and direct contact estimate some missing data. In particular,
between humans or with animals. As Bayesian regression modelling has
many data are lacking for attribution, been used to estimate missing disease
it was decided to apply structured incidence data.
expert elicitation to provide a consistent Due to the limitations in data availability,
set of estimates. The global expert FERG decided to present its estimates
elicitation study involved 73 experts and on a regional level, even though all
11 elicitors, and was one of the largest, calculations were made on a national
if not the largest study, of this kind ever level. The regional estimates are
undertaken. Combining all streams of considered more robust as they build
data resulted in estimates of the global on data from several countries in most
burden of foodborne disease. regions. Yet, the regional estimates
Unlike previously completed national do not reflect the diversity of risks
burden of illness studies, FERG decided between countries in a region, or even
to also include chemical hazards. The within a country. Maps are therefore
inclusion of chemical hazards was not presented as it was considered
particularly challenging, and it was that these would not adequately reflect
only through determined efforts by regional heterogeneity.
the Chemicals and Toxins Task Force The results of the FERG project are
(CTTF) that several chemical hazards presented in several formats. A PLOS
could be included. Whereas WHO collection entitled “The World Health
committees such as the Joint FAO/WHO Organization Estimates of the Global
Expert Committee on Food Additives Burden of Foodborne Diseases”, which
(JECFA) and Joint FAO/WHO Meeting can be accessed at a dedicated website.1
on Pesticide Residues (JMPR) typically The website presents the key results in
use a risk assessment approach, a a series of seven peer-reviewed papers,
counterfactual attribution approach and also provides access to a large
is commonly applied in global burden and growing number of reviews and
estimates of cancer, cardiovascular and description of methods that have been
other diseases. Deciding which of these published in different peer-reviewed
approaches was most appropriate for journals. This large body of evidence
FERG was a difficult, and as yet not fully reflects the considerable support given to
resolved, process. As a result, burden FERG by the global scientific community.
estimates for several important chemical These papers are also accessible through
contaminants (methylmercury, lead, a dedicated WHO website.2 WHO has
arsenic and cadmium) are expected to be also produced this report, documenting
presented at a later stage. the results and the process of estimating
CONCLUSION
1
Even though all efforts were made to http://collections.plos.org/ferg-2015 accessed 2
December 2015
include the best available science in the 2
http://www.who.int/foodsafety/areas_work/
estimates, FERG is fully aware of the foodborne-diseases/ferg/en/ accessed 3
limitations of the current work. Data November 2015
WHO Estimates of the global burden of foodborne diseases
127
the global burden of foodborne disease (as documented for aflatoxin, see
and an interactive website allowing Section 6.4).
stakeholders to explore the results from
Countries who want to build their
different perspectives.
national food safety strategies are
Even though the currently presented advised to combine the global estimates
burden of foodborne disease is with national data. It is our experience
substantial, it was not feasible to that a vast amount of additional data
document the full burden, which is likely exist but has not yet been mined because
to be considerably higher. Not all relevant it is not available in easily accessible
contaminants could be included, and for databases but rather in paper form.
those that were included, not all relevant Building on such data may provide
endpoints could be taken into account. sources of validation for any estimates
FERG selected a shortlist of hazards at derived from FERG numbers. As a next
the onset, reducing a list of more than step, further development of national
100 contaminants to 40. Exclusions laboratory-based surveillance programs,
were based on initial judgments about should be a priority.
the importance of the global or regional
A crucial element of the initiative, often
burden, but also on data availability. Of
taking a back seat during the huge
the 40 contaminants selected, analyses
effort in generating global and regional
have not been completed for lead,
burden estimates, was therefore the
methyl mercury, arsenic and cadmium
promotion of foodborne burden of
for inclusion in this report. Of potentially
disease studies and capacity building in
relevant endpoints, only Guillain-Barré
individual countries. FERG was only able
syndrome and haemolytic uraemic
to make limited progress towards this
syndrome and invasive salmonellosis
objective, in the form of pilot studies in
were included as outcomes for diarrhoeal
four countries. Since some of these pilot
diseases, but not irritable bowel
studies encountered significant resource
syndrome or other functional bowel
barriers and data shortages, it is hoped
disorders that are increasingly linked to
that one legacy of the initiative would
diarrhoeal disease in developed countries
be to help overcome these through local
and are associated with a substantial
use of regional estimates. Individual
burden. FERG estimates do not include
countries can evaluate and apply the
the effects of foodborne diseases
FERG regional burden estimates to
on malnutrition and development in
generate national DALY-based burden
low- and middle-income countries,
data for foodborne illness prioritization.
and invasive salmonellosis in HIV co-
Such a process should include local data
morbid cases was also excluded, even
for validation where available, and be
though a major proportion of these
undertaken by local scientists with an
infections may be foodborne. No
awareness of the food safety context in
stillbirths were included for listeriosis
their country. FERG has also sought to
and toxoplasmosis, but many would
promote knowledge translation of burden
be preventable by appropriate food
of disease estimates into food safety
safety interventions. The counterfactual
policy at a national level.
approach for chemicals produces lower
estimates than risk assessment approach
Conclusion
128
CONCLUSION
WHO Estimates of the global burden of foodborne diseases
129
APPENDICES
WHO Estimates of the global burden of foodborne diseases
131
APPENDIX 1.
Formal Description of the Project and Participants
A1.1 Terms of Reference for A1.2 Foodborne Disease Burden
WHO’s Foodborne Disease Epidemiology Reference Group
Burden Epidemiology Reference The Foodborne Disease Burden
Group (FERG) Epidemiology Reference Group
The Foodborne Disease Burden (FERG) is composed of internationally
Epidemiology Reference Group renowned experts in a broad range of
(FERG) will act as an advisory body to disciplines relevant to global foodborne
WHO on matters of global foodborne disease epidemiology. Members were
diseases epidemiology. appointed by the WHO Director-
General, Dr Margaret Chan, following a
Functions: transparent selection process.
The FERG shall have the The expert group is charged to:
following functions: f assemble, appraise and report on
f To review epidemiological data on the current, the projected, as well
foodborne disease burden. as the averted burden of foodborne
f To identify technical gaps and priorities disease estimates;
for research activities. f conduct epidemiological reviews for
f To make recommendations to WHO mortality, morbidity and disability in
on the establishment of FERG TFs and each of the major foodborne diseases;
other means through which scientific f provide models for the estimation of
and technical matters are addressed. FBD burden where data are lacking;
f develop cause attribution models to
Composition:
estimate the proportion of diseases that
f FERG members shall serve in their
are foodborne; and, most importantly,
personal capacities to represent the f use the FERG models to develop user-
broad range of disciplines relevant to friendly tools for burden of foodborne
global foodborne disease epidemiology. disease studies at country level.
f Members of the FERG, including the
Chair, shall be selected by the Director- To estimate the global human health
General. burden (expressed in Disability-Adjusted
f Members of the FERG, including the Life Years– DALYs), FERG will initially
Chair, shall be appointed to serve for a focus on microbial, parasitic, zoonotic
period of one year, and shall be eligible and chemical contamination of food with
for re-appointment. an emphasis on:
Operation: f diseases whose incidence and severity
The FERG shall usually meet at least is thought to be high; and
f pathogens and chemicals that are most
twice a year. WHO shall provide any
necessary scientific, technical and other likely to contaminate food, and that
support for the FERG, including for the have a high degree of preventability.
preparation of meeting reports. FOS shall
provide secretarial support.
Appendices
132
A1.3 Participants Formerly
National Institute for Public Health and
FERG Members the Environment
f Formally appointed by the WHO Bilthoven
Director-General, following a Netherlands
selection procedure. Vice-chair
f Allocated to Core Group and TFs.
2007 - 2010
f Have full participation rights in all
Nilanthi DE SILVA
technical discussions.
Dean and Professor of Parasitology
Resource advisers Faculty of Medicine, University
f Not formally appointed by the of Kelaniya
Director General. Ragama, Sri Lanka
f Allocated to TFs on an ad hoc basis
(as required). 2011 - 2015
f Have full participation rights in Alejandro CRAVIOTO
technical discussions. Global Evaluative Sciences
Seattle, WA.
WHO Secretariat and other USA
UN Organizations
f Have full participation rights in
Members
technical discussions. Gabriel O Adegoke
f Allocated to TFs on an ad hoc basis. Department of Animal Science, National
University of Lesotho
Observers Lesotho
f Nominated by FERG members (one
per member). Reza Afhshari
f No ‘formal’ right of intervention Head, Development of Research and
in plenary. Education Development
f Participation in TFs, as appropriate. Mashhad University of Medical Sciences
Stakeholders Iman Rez Hospital
Iran (Islamic Rep.)
f Invited by WHO to designated sessions.
f Formal right of intervention in
Frederick J. ANGULO
designated sessions.
Associate Director for Science
f No participation in technical discussions
Division of Global Health Protection
to avoid conflicts of interest.
Center for Global Health,
CDC, Atlanta GA USA
A1.4 Members of the Foodborne
Disease Burden Epidemiology Janis BAINES
Reference Group (FERG) (past Manager
and present) Food composition, Evaluation and
Modelling Section
Chair Food Standards Australia New Zealand
Arie HAVELAAR Canberra BC ACT
Emerging Pathogens Institute Australia
University of Florida
Gainesville, FL
USA
APPENDICES
WHO Estimates of the global burden of foodborne diseases
133
Observers
Ermias Woldemariam AMENE Patricia GRIFFIN
School of Veterinary Medicine CDC
University of Wisconsin-Madison, Atlanta
WI, USA Georgia, USA
SATF: CSTF:
Chair: Tine Hald Chair: Niko Speybroeck (2009 - 2011),
Rob Lake (2012 - 2015)
SATF and FERG members: Arie Havelaar,
Fred Angulo, Fumiko Kasuga, Nilanthi CSTF and FERG members: Gabriel
de Silva, Muhammad Rokni, David Adegoke, Fred Angulo, David Bellinger,
Bellinger, Robert Black, Herman Gibb, Dr Alejandro Cravioto, Dörte Döpfer, Nicolas
Wan Mansor Bin Hamzah, Dörte Döpfer, Praet, Herman Gibb, Arie Havelaar,
Rob Lake Fumiko Kasuga, Bocar Kouyate, George
Nasinyama, Robert Buchanan, Catterina
Commissioned scientists/resource
Ferriccio, Bocar Kouyate, Myron Levine,
advisers/consultants: Sara Pires, Roger
Sarah O’Brien, Nilanthi de Silva, Paul
Cooke, Sandra Hoffmann, Willie Aspinall
Torgerson
CTF:
Consultants: Brecht Devleesschauwer,
Chair: Nicolas Praet Charline Maertens de Noordhout,
CTF and FERG members: Brecht Juanita Haagsma
Devleesschauwer, Paul Torgerson, Aamir KTPG:
Fazil, David Bellinger, Arie Havelaar, Rob
Chair: Pierre Ongolo-Zogo (2010 - 2011),
Lake, Dörte Döpfer, Niko Speybroeck,
John Ehiri (2012 - 2015)
Eric Fèvre
Members: Deena Alasfoor,
Commissioned scientists/resource
Nasreen Jessani, Helen Jensen,
advisers/consultants: Dana Cole, Sara
Tanja Kuchenmüller, Haichao Lei,
Pires, Juanita Haagsma, Kate Thomas,
Bocar Kouyate
Scott McDonald, Felicia Wu
Consultant: Sandy Campbell
APPENDICES
WHO Estimates of the global burden of foodborne diseases
143
APPENDIX 2.
Subregions
SUBREGION(1) [5] WHO MEMBER STATES
AFR D Algeria; Angola; Benin; Burkina Faso; Cameroon; Cabo Verde; Chad; Comoros; Equatorial Guinea; Gabon;
Gambia; Ghana; Guinea; Guinea-Bissau; Liberia; Madagascar; Mali; Mauritania; Mauritius; Niger; Nigeria;
Sao Tome and Principe; Senegal; Seychelles; Sierra Leone; Togo.
AFR E Botswana; Burundi; Central African Republic; Congo; Côte d’Ivoire; Democratic Republic of the Congo;
Eritrea; Ethiopia; Kenya; Lesotho; Malawi; Mozambique; Namibia; Rwanda; South Africa; Swaziland;
Uganda; United Republic of Tanzania; Zambia; Zimbabwe.
AMR A Canada; Cuba; United States of America.
AMR B Antigua and Barbuda; Argentina; Bahamas; Barbados; Belize; Brazil; Chile; Colombia; Costa Rica;
Dominica; Dominican Republic; El Salvador; Grenada; Guyana; Honduras; Jamaica; Mexico; Panama;
Paraguay; Saint Kitts and Nevis; Saint Lucia; Saint Vincent and the Grenadines; Suriname; Trinidad and
Tobago; Uruguay; Venezuela (Bolivarian Republic of).
AMR D Bolivia (Plurinational State of); Ecuador; Guatemala; Haiti; Nicaragua; Peru.
EMR B Bahrain; Iran (Islamic Republic of); Jordan; Kuwait; Lebanon; Libya; Oman; Qatar; Saudi Arabia; Syrian
Arab Republic; Tunisia; United Arab Emirates.
EMR D Afghanistan; Djibouti; Egypt; Iraq; Morocco; Pakistan; Somalia; South Sudan(2); Sudan; Yemen.
EUR A Andorra; Austria; Belgium; Croatia; Cyprus; Czech Republic; Denmark; Finland; France; Germany; Greece;
Iceland; Ireland; Israel; Italy; Luxembourg; Malta; Monaco; Netherlands; Norway; Portugal; San Marino;
Slovenia; Spain; Sweden; Switzerland; United Kingdom.
EUR B Albania; Armenia; Azerbaijan; Bosnia and Herzegovina; Bulgaria; Georgia; Kyrgyzstan; Montenegro;
Poland; Romania; Serbia; Slovakia; Tajikistan; The Former Yugoslav Republic of Macedonia; Turkey;
Turkmenistan; Uzbekistan.
EUR C Belarus; Estonia; Hungary; Kazakhstan; Latvia; Lithuania; Republic of Moldova; Russian Federation;
Ukraine.
SEAR B Indonesia; Sri Lanka; Thailand.
SEAR D Bangladesh; Bhutan; Democratic People’s Republic of Korea; India; Maldives; Myanmar; Nepal; Timor-
Leste.
WPR A Australia; Brunei Darussalam; Japan; New Zealand; Singapore.
WPR B Cambodia; China; Cook Islands; Fiji; Kiribati; Lao People’s Democratic Republic; Malaysia; Marshall Islands;
Micronesia (Federated States of); Mongolia; Nauru; Niue; Palau; Papua New Guinea; Philippines; Republic
of Korea; Samoa; Solomon Islands; Tonga; Tuvalu; Vanuatu; Viet Nam.
Notes: (1) The subregions are defined on the basis of child and adult mortality as described by Ezzati et al. [5]. Stratum A = very
low child and adult mortality; Stratum B = low child mortality and very low adult mortality; Stratum C = low child mortality and high
adult mortality; Stratum D = high child and adult mortality; and Stratum E = high child mortality and very high adult mortality. The
use of the term ‘subregion’ here and throughout the text does not identify an official grouping of WHO Member States, and the
“subregions” are not related to the six official WHO regions, which are AFR = African Region; AMR = Region of the Americas; EMR =
Eastern Mediterranean Region; EUR = European Region; SEAR = South-East Asia Region; WPR = Western Pacific Region.
(2) South Sudan was re-assigned to the WHO African Region in May 2013. As this study relates to time periods prior to this date,
estimates for South Sudan were included in the WHO Eastern Mediterranean Region.
Appendices
144
APPENDIX 3.
Preliminary hazards considered by each task force
At the FERG 1 meeting (26–28 November Dientamoeba fragilis
2007), each of the hazard-based TFs Diphyllobothrium latum
considered a comprehensive list of potential Echinococcus spp.
foodborne hazards for the development of Echinostoma spp.
burden estimates. During the course of the Entamoeba histolytica
project these lists had to be reduced, largely Fasciola spp.
for practical reasons concerning the ability Fasciolopsis buski
to generate burden estimates. For reference, Gastrodiscoides hominis
the complete list is given here. Giardia spp.
EDTF Gnathostoma spinigerum
Adenovirus Heterophyes heterophyes
Aeromonas spp. Hymenolepis nana
Astrovirus Isospora belli
Bacterial toxins (B. cereus) Linguatula serata
Bacterial toxins (C. perfringens) Metagonimus yokogawai
Bacterial toxins (S. aureus) Nanophytes salmincola
Brucella spp. Opisthorchis felineus
Campylobacter spp. Opisthorchis viverrini
Clostridium botulinum Paragonimus spp.
Enteroaggerative E. coli (EAggEC) Sarcocystis hominis
Entero-pathogenic E. coli (EPEC) Taenia saginata
Entero-toxigenic E. coli (ETEC) Taenia solium
Enterovirus Toxocara spp.
Helicobacter pylori Toxoplasma gondii
Hepatitis A virus Trichinella spp.
Hepatitis E virus Trichostrongylus spp.
Leptospira spp. Trichuris trichiura
Listeria monocytogenes CTTF
Mycobacterium bovis Elemental contaminants (e.g. lead,
Non cholera Vibrios mercury, cadmium, manganese, arsenic)
Norovirus Mycotoxins (e.g. aflatoxins, ochratoxins,
Prions fumonisin, trichothocenes)
Rotavirus
Food additives (e.g. sulphites, nitrites/
Salmonella (non-typhoidal) spp.
nitrates, benzoic acid)
Salmonella (typhoid) spp.
Pesticides (e.g. organophosphates,
Shiga-toxin producing E. coli (STEC)
carbamates, DDT, pyrethrins)
Shigella spp.
Vibrio cholerae 01/0139 Organic industrial pollutants (e.g.
Yersinia spp. persistent organic pollutants)
PDTF Veterinary drugs/residues (e.g.
antibiotics, hormones–but not
Ancylostoma duodenale antimicrobial residues)
Angiostrongylus cantonensis
Seafood toxins (e.g. tetrodotoxin,
Angiostrongylus costaricensis
ciguatera, shellfish toxins, DSPs,
Anisakis simplex
PSPs, histamines)
Ascaris spp.
Blastocystis hominis Process contaminants (e.g. acrylamide,
Capillaria philippinensis PAHs, choropropanol)
Clonorchis sinensis Allergens (e.g. peanuts)
Cryptosporidium spp. Natural toxicants (e.g. cyanide in
Cyclospora spp. cassava, aminoglycosides)
APPENDICES
APPENDIX 4.
Hazard-specific input parameter sources and methods
A4.1 Brucellosis brucellosis incidence estimates for
81 countries. The FERG Computational
Incidence Task Force imputation model was
There were 32 countries identified as then used to impute an incidence of
“free of brucellosis in livestock”, using human brucellosis in all countries with
2006–2012 data reported to the World missing incidence.
Organisation for Animal Health (OIE)
[248], and a list of European countries Clinical Outcomes
recognized by the European Union as The FERG-commissioned systematic
“officially brucellosis free” in cattle, sheep review assisted in determining the clinical
and goats in 2010 [249]. Using 2001– outcomes for human brucellosis [254].
2004 OIE data, a previous review [250] These were: acute brucellosis (severe);
estimated human brucellosis incidence acute brucellosis (moderate); chronic
for 9 of the countries identified as free brucellosis; brucellosis orchitis; and
of brucellosis in livestock. The median brucellosis death. For acute brucellosis,
human brucellosis incidence from these it was assumed that 50% of cases were
9 countries free of brucellosis in livestock severe, 50% of cases were moderate,
was used as the estimated human 40% of brucellosis cases resulted in
brucellosis incidence for each of the chronic brucellosis, and 10% of brucellosis
32 countries free of brucellosis in livestock. cases in males resulted in orchitis [254].
A FERG-commissioned systematic review
was then used to screen 2385 articles Duration
[251] and a literature review for national Acute brucellosis: duration 14 days (min.
human brucellosis incidence estimates 7 days–max. 21 days). Chronic brucellosis:
[174, 175, 187, 188, 252, 253], to extract duration 6 months (min. 3 months– max.
brucellosis national incidence estimates 24 months). Brucellosis orchitis: duration
for 17 countries (Argentina, Canada, 6 months (min. 3 months–max.
Chad, China, Egypt, France, Greece, Iraq, 24 months) [254].
Iran, Italy, Kyrgyztan, Jordan, Mexico,
Oman, Saudi Arabia, Turkey, and the Disability weight
United States of America). The human Acute brucellosis (severe): GBD2010
brucellosis incidence estimates in each disability weight of 0.210 (95% UI
of these countries were compared with 0.139–0.298) for infectious disease,
human brucellosis incidence estimates acute episode, severe. Acute brucellosis
in the same country in a previous review, (moderate): GBD2010 disability weight
which used 2001–2004 OIE data [250], to of 0.053 (95% UI 0.033–0.081) for
estimate a multiplier (mean=5.4, range infectious disease, acute episode, mild.
1.6–15.4) to account for under-reporting. Chronic brucellosis: GBD2010 disability
This multiplier was used to estimate weight 0.079 (95%UI 0.053–0.115)
national human brucellosis incidence for for musculoskeletal problems, legs,
countries with OIE human brucellosis data moderate. Brucellosis orchitis: GBD2010
in the previous review but without national disability weight of 0.097 (95% UI 0.063–
human brucellosis incidence estimates 0.0137) for epididymo-orchitis [82].
identified in the current systematic review
or literature review. By multipling the Mortality
human brucellosis incidence reported to Acute brucellosis and chronic brucellosis
OIE by the multiplier, there were 32 such case fatality ratio 0.5% (min. CFR 0.25%–
countries. These steps yielded human max. CFR 0.75%) [255, 256].
Appendices
146
Incidence Mortality
There were 51 countries identified as Deaths from M. bovis were estimated
“free of Mycobacterium bovis in cattle” following the same approach for
using 2005–2012 data reported to OIE estimating M. bovis cases after
[248] and a list of European countries reducing the mortality by 20% due to
recognized by the European Union as the recognition from another FERG-
“officially free of bovine tuberculosis” commissioned review that M. bovis
in 2010 [249]. A FERG-commissioned infections are more likely to result in
systematic review screened 1203 articles extrapulmonary infections [259] and that
[258] with data from 91 countries, and extrapulmonary infections have a lower
estimated the median proportion of case-fatality ratio (CFR) than pulmonary
human tuberculosis cases due to M. bovis tuberculosis infections; a 20% reduction
at the region level as 2.8% for AFR, 0.4% in mortality was based on a review of
the United States of America national
for EUR and 0.3% for AMR; the overall
surveillance data from 2009–2010, which
median proportion from studies in the
found that the CFR for extrapulmonary
review (1.0%) was used in the three
tuberculosis infections was approximately
other regions. These proportions were
20% lower than the CFR for pulmonary
applied to all countries in each respective
tuberculosis infections. Therefore,
region except for the 51 countries free of
country-level human tuberculosis
M. bovis in cattle. The lowest observed mortality rates of tuberculosis among
proportion (0.3%) was assigned to the persons not infected with HIV were
51 countries free of M. bovis in cattle. abstracted from the WHO Global
Country-level human tuberculosis Tuberculosis Report [165], reduced by
incidence was abstracted from the WHO 20%, and then multiplied by population
Global Tuberculosis Report [165] and estimates and the proportion of human
multiplied by population estimates and tuberculosis cases due to M. bovis to
the proportion of human tuberculosis estimate M. bovis deaths.
cases due to M. bovis to estimate human
M. bovis cases.
APPENDICES
WHO Estimates of the global burden of foodborne diseases
147
A4.8 Clostridium Mortality
perfringens intoxication Estimates of mortality were only
conducted for the 61 EUR and other
Incidence subregion A (low mortality) countries.
Estimates of incidence were only National estimates of Clostridium
conducted for the 61 EUR and other perfringens intoxications cases and
subregion A (low mortality) countries. deaths were available from Australia
Based on a literature review for articles [272], France [174], Netherlands [154],
with national estimates of foodborne New Zealand [252], and the United States
diseases that included Clostridium of America [188]; the median case fatality
perfringens intoxications, we identified ratio (CFR) from these five countries
national incidence estimates for was the New Zealand (0.0030% [95%CI:
Clostridium perfringens intoxications 0.0024%-0.0038%]), therefore the CFR
from seven countries: Australia [272], from New Zealand was used as the
Canada [175], France [174], Netherlands CFR for all EUR and other subregion
[154], New Zealand [252], United A countries.
Kingdom [48], and the United States of
America [188]. The median C. perfringens Age distribution
intoxication incidence from these Acute gastroenteritis and deaths due to
seven countries was from the United Clostridium perfringens intoxication age
States of America, therefore the C. distribution: 1% <5 years; 13% 5–14 years;
perfringens intoxication incidence from 59% 15–54 years; 27% *55 years [273].
the United States of America (324.19
per 100 000 population with a 95% Sex distribution
confidence interval of 126.14–833.44 per Acute gastroenteritis and deaths due to
100 000) was used as the C. perfringens Clostridium perfringens intoxication sex
intoxication incidence for all EUR and distribution: 63% male [273].
other subregion A countries.
epilepsy, treated with recent seizures; 5–14 years; 10% 15–34 years; 6% 35–44
and (4) 0.76 for motor impairment, years; 7% 45–54 years; 13% 55–64 years;
moderate. 20% 65–74 years; 20% 75–84 years; 18%
– For listeriosis neurological sequelae: *85 years.
a DW of 0.292 (95% UI 0.272–0.316)
derived from a multiplicative Sex distribution
methodology and expert elicitation The sex distribution of listeriosis cases
(with bootstrap analysis for CI) using and deaths was determined from
a combination of following DWs: published papers during the FERG-
(1) 0.047 resulting from average of commissioned review [70]. The sex
all 10 DWs involving hearing loss; distribution for listeriosis cases and
(2) 0.087 resulting from average of all deaths was: 50% male.
5 DWs for vision loss; and (3) 0.303
resulting from average of all 4 DWs
for stroke, long-term consequence A4.12 Non-typhoidal
[70, 82]. Salmonella infection
Mortality Incidence
Listeriosis case fatality ratios were The incidence of diarrhoeal non-typhoidal
estimated following the same approach Salmonella (NTS) was estimated
for estimating clinical outcomes of separately for middle and high mortality
listeriosis cases; using probabilities from countries, and low mortality countries.
the FERG-commissioned review and a For the 133 middle to high mortality
random effects meta-regression model. countries, we used a modification of the
For each study identified in the review, a Child Health Epidemiology Reference
weight was assigned reflecting the study Group (CHERG) approach [50]. To
quality; these weights were included as a derive “envelopes” of diarrhoea cases,
fixed effect in the meta-regression model. for children <5 years of age we used
The case fatality ratio for perinatal cases estimates of diarrhoea incidence from
was 14.9% (mimimum 11.3% - max. 18.5%); a CHERG systematic review [51] and
9.2% (mimimum 7.5% - max. 10.9%) for persons >5 years of age we used
resulted in neonatal deaths and 5.7% a FERG-commissioned systematic
(minumum 3.8% - max. 7.6%) resulted in review [52]. We then estimated the
stillbirths; stillborns were not included in aetiological proportions of diarrhoeal
the final FERG estimates of deaths and illnesses due to NTS and the 10 other
DALYs. The case fatality ratio for non- diarrhoeal pathogens1 in children <5
perinatal cases was 25.9% (mimimum years of age using CHERG and FERG
23.8% - max. 29.0%). systematic reviews of aetiology studies
among outpatients and persons in the
Age distribution community [40], and the aetiological
The age distribution of listeriosis cases proportion of diarrhoeal illnesses due
and deaths was determined from to NTS and the other 10 diarrhoeal
published papers during the FERG- pathogens in persons >5 years of age
commissioned review [70]. The age 1
The 11 diarrhoeal pathogens are: non-typhoidal
distribution for perinatal listeriosis cases Salmonella, Campylobacter, Shigella, norovirus,
and deaths was: 100% <1 month. The age enterotoxigenic E. coli (ETEC), enteropathogenic
E. coli (EPEC), Cryptosporidia, Giardia, Entamoeba
distribution for non-perinatal cases and histolytica, other diarrhoeal agents not known
deaths was: 0% <1 year; 2% 1–4 years; 4% to be foodborne (rotavirus and astrovirus), and
unspecified agents.
Appendices
156
States: 0.18 per 100 000 with a range of to Shigella and the 10 other diarrhoeal
0.11– 0.28. We assumed no deaths among pathogens in persons >5 years of age
vomiting-only norovirus cases. using an updated FERG systematic
review of aetiology studies among
Age distribution inpatients, outpatients and persons in
In middle-to-high mortality countries the community [40, 274]. The shigellosis
we estimated incidence and mortality aetiological proportions were extracted
of norovirus seperately for children <5 from studies, and regional median
years of age and persons >5 years of shigellosis aetiological proportions
age. In low mortality countries the age calculated. We modified the CHERG
distribution for norovirus was 40% <5 approach by dropping regional median
years; 10% 5–14 years; 30% 15–44 years; shigellosis aetiological proportion
10% 45–64 years; and 10% outliers that were >5 times greater
>65 years [284]. than the global median shigellosis
aetiological proportion, and replacing
Sex distribution missing regional shigellosis aetiological
Norovirus sex distribution: 50% male. proportions with the global median.
Furthermore, for children <5 years
of age, we proportionally decreased
A4.16 Shigellosis the aetiological proportions for all 11
diarrhoeal pathogens in each region
Incidence so that the sum of the aetiological
The incidence of shigellosis was proportions for all diarrhoeal pathogens
estimated separately for middle-to- in a region equalled 1. The resultant
high mortality countries, and low regional shigellosis aetiological
mortality countries. For the 133 middle- proportions were multiplied by the
to-high mortality countries, we used a regional estimates of diarrhoea incidence,
modification of the CHERG approach and the resultant regional shigellosis
[50]. To derive “envelopes” of diarrhoea incidence was applied to all countries in
cases, for children <5 years of age we that region.
used estimates of diarrhoea incidence
In the 61 low mortality countries (EUR
from a CHERG systematic review [51]
and other subregion “A” countries), we
and for persons >5 years of age we
used a literature review that identified
used a FERG-commissioned systematic
national incidence estimates for
review [52]. We then estimated the
shigellosis from five countries: Australia
aetiological proportions of diarrhoeal
[272], Canada [175], France [174], New
illnesses due to Shigella and the 10 other
Zealand [252], and the United States of
diarrhoeal pathogens4 in children <5
America [188]. These national estimates
years of age using a CHERG and FERG
were based on systematic reviews,
systematic review of aetiology studies
national surveillance data, and expert
among outpatients and persons in the
judgment. In these five countries, we
community [40], and the aetiological
used the estimated national shigellosis
proportion of diarrhoeal illnesses due
incidence (and range) for that country.
4
The 11 diarrhoeal pathogens are: non-typhoidal
For low mortality countries without a
Salmonella, Campylobacter, Shigella, norovirus,
enterotoxigenic E. coli (ETEC), enteropathogenic national estimate, we used the median
E. coli (EPEC), Cryptosporidia, Giardia, Entamoeba shigellosis incidence from the five
histolytica, other diarrhoeal agents not known national studies. The median incidence
to be foodborne (rotavirus and astrovirus), and
unspecified agents. was from Canada (which was increased
Appendices
164
diarrhoea, 8.5% of EPEC cases resulted <5 years of age using a CHERG and FERG
in moderate diarrhoea, and 91% of EPEC systematic review of aetiology studies
cases resulted in mild diarrhoea. among inpatients [40], and the aetiological
proportions of diarrhoeal deaths due
Duration to EPEC and the 10 other diarrhoeal
In children <5 years of age, duration pathogens in persons >5 years of age using
of severe diarrhoea was 8.4 days, an updated FERG systematic review of
moderate diarrhoea was 6.4 days, and aetiology studies among inpatients [40,
mild diarrhoea was 4.3 days [188]. Based 274]. The EPEC aetiological proportions
on the assumed distribution of severe, were extracted from studies, and regional
moderate and mild diarrhoea cases, the median EPEC aetiological proportions
duration of EPEC diarrhoea cases in calculated. We modified the CHERG
children <5 years of age was estimated approach by dropping regional median
to be 4.9 days (min. 4.3 days–max. 8.4 EPEC aetiological proportion outliers that
days). In persons >5 years of age, the were >5 times greater than the global
duration of diarrhoea was 2.8 days [266]. median EPEC aetiological proportion,
and replacing missing regional EPEC
Disability weight aetiological proportions with the global
– Acute EPEC diarrhoea (severe): median. Furthermore, for children <5 years
GBD2010 disability weight of 0.281 of age, we proportionally decreased the
(95% UI 0.184–0.399) for diarrhoea, aetiological proportions for all 11 diarrhoeal
severe. pathogens in each region so that the
– Acute EPEC diarrhoea (moderate): sum of the aetiological proportions for all
GBD2010 disability weight of 0.202 diarrhoeal pathogens in a region equalled 1.
(95% UI 0.133–0.299) for diarrhoea, The resultant EPEC aetiological proportions
moderate. were multiplied by the regional estimates of
– Acute EPEC diarrhoea (mild): diarrhoea deaths, and the resultant regional
GBD2010 disability weight of 0.061 EPEC mortality was applied to all countries
(95% UI 0.036–0.093) for diarrhoea, in that region. We estimated no EPEC
deaths in the 61 low mortality countries
mild [82].
(EUR and other subregion “A” countries).
Mortality
The mortality of EPEC was estimated Age distribution
separately for middle-to-high mortality In middle-to-high mortality countries
countries, and low mortality countries. we estimated incidence and mortality
For the 133 middle-to-high mortality of EPEC separately for children <5 years
countries, we used a modification of the of age and persons >5 years of age. In
CHERG approach [50]. We received low mortality countries, no information
envelopes of diarrhoeal deaths from WHO; was available on the age distribution of
because this estimate was not available EPEC cases; we therefore used the age
with an uncertainity interval, we used the distribution for Campylobacter diarrhoea
uncertainity range from the GBD2010 cases as a proxy, which was 11% <5 years;
estimate of diarrhoeal deaths (81.7% to 8% 5–14 years; 10% 15–24 years; 57%
114.6% around the point estimate) [58]. We 25–64 years; and 14% >65 years.
then estimated the aetiological proportions
of diarrhoeal deaths due to EPEC and the Sex distribution
10 other diarrhoeal pathogens in children EPEC sex distribution: 50% male.
APPENDICES
WHO Estimates of the global burden of foodborne diseases
169
AGE
DISTRIBUTION <5 5-14 15-24 25-34 35-44 45-54 55-64 65-74 75-84 85+
(YEARS)
G1Mean 0.26 0.34 0.19 0.11 0.055 0.027 0.013 0.0057 0.0015 0.00016
G2 Mean 0.21 0.3 0.2 0.13 0.078 0.043 0.023 0.011 0.0029 0.0002
G3 Mean 0.09 0.18 0.19 0.15 0.14 0.12 0.071 0.041 0.02 0.0059
G4Mean 0.075 0.14 0.14 0.14 0.14 0.13 0.11 0.064 0.039 0.016
G5Mean 0.072 0.11 0.17 0.32 0.21 0.086 0.021 0.0049 0.00098 0.00019
APPENDICES
WHO Estimates of the global burden of foodborne diseases
177
AGE DISTRIBUTION
<5 5-14 15-24 25-34 35-44 45-54 55-64 65+
(YEARS)
G1Mean 4.6 10.4 17.1 21.2 18.1 12.8 8.7 7.1
Clinical Outcomes
A4.26 Alveolar echinococcosis Abdominopelvic problems followed by
recovery after treatment, or death.
Incidence
Full details of the methodology of Duration
estimating the incidence of alveolar Europe: 10 years. Other: 8 years.
echinococcosis (AE) can be found in
[72]. In addition, this data has been Disability weight
updated because of subsequent reports
– 0.123
Appendices
178
Clinical Outcomes
Abdominal pelvic discomfort, carcinoma.
Appendices
180
AGE
DISTRIBUTION 0-1 1-4 5-9 10-14 15-19 20-24 25-34 35-44 45-54 55-64 65-74 75-84 85+
(YEARS)
G1Mean 5.0 8.1 11.5 13.4 22.7 12.4 13.9 8.1 2.9 1.1 0.6 0.2 0.0
G2 Mean 4.3 7.5 10.4 12.0 18.8 13.5 16.7 8.0 3.8 2.4 1.4 0.8 0.3
G3 Mean 4.1 7.2 9.1 13.0 32.9 18.6 12.9 1.6 0.3 0.1 0.1 0.1 0.1
G4Mean 6.8 8.8 11.2 12.4 23.5 12.2 13.4 7.6 2.6 0.9 0.5 0.2 0.0
G5Mean 5.1 8.2 11.4 13.2 22.6 12.5 14.0 8.0 2.9 1.1 0.6 0.3 0.1
Age distribution
AGE
DISTRIBUTION 0-1 1-4 5-9 10-14 15-19 20-24 25-34 35-44 45-54 55-64 65-74 75-84 85+
(YEARS)
G1 Mean 49.1 26.9 11.6 4.6 2.3 1.4 1.8 1.0 0.6 0.3 0.2 0.1 0.0
G2 Mean 19.5 33.1 29.5 10.4 4.3 1.4 1.0 0.4 0.2 0.1 0.0 0.0 0.0
G3 Mean 75.3 14.8 5.1 1.8 0.9 0.6 0.7 0.4 0.2 0.1 0.1 0.0 0.0
G4 Mean 21.9 52.6 16.1 1.3 0.5 0.7 1.5 1.3 1.1 0.8 0.7 0.7 0.7
G5 Mean 59.1 22.0 9.4 3.6 1.8 1.1 1.3 0.7 0.4 0.2 0.2 0.1 0.0
Age distribution
AGE
DISTRIBUTION 0-1 1-4 5-9 10-14 15-19 20-24 25-34 35-44 45-54 55-64 65-74 75-84 85+
(YEARS)
G1 Mean 4.2 11.8 13.9 12.4 10.7 8.7 13.4 10.3 7.1 3.9 2.4 0.9 0.2
G2 Mean 4.3 13.4 15.8 15.4 12.7 8.5 10.9 8.1 5.5 2.3 2.1 0.9 0.2
G3 Mean 2.3 6.8 8.4 9.6 11.6 9.6 13.7 11.6 11.1 6.2 5.6 2.9 0.5
G4 Mean 4.2 11.7 13.8 12.4 10.7 8.7 13.4 10.3 7.2 4.0 2.4 1.0 0.2
Sex distribution
Male = 55.9%
APPENDICES
WHO Estimates of the global burden of foodborne diseases
183
Sex distribution
Male = 55%.
Appendices
184
Duration Mortality
It is assumed that peanut allergy is a The limited data on the mortality rate of
lifelong disease. It is important to note peanut-induced anaphylaxis show values
that the duration of allergic symptoms is ranging from 0 to 0.006 deaths per
very short. 100 000 person-years [102]. To reflect
this uncertainty, the mortality rate of
Disability Weight peanut-induced anaphylaxis in subregion
Mullins et al. [103] reported that “A” countries was modelled as a Uniform
52% of cases referred to a specialist distribution ranging from 0/100 000
allergy medical practice in Australia to 0.006/100 000. Given the lack of
suffered from mild symptoms (skin and data, no estimates were generated for
subcutaneous tissue involvement only), other countries.
42% from moderate symptoms (features
suggestive of respiratory, cardiovascular Age distribution
or gastrointestinal involvement), and The onset of peanut allergy is early in
6% from severe symptoms (cyanosis, life (median age 18–24 months, [107,
hypotension, confusion, collapse, loss 108], with continued prevalence in older
of consciousness, incontinence). We age groups. All incident cases of peanut
propose the DW for clinically relevant allergy were therefore assumed to
peanut allergy be a weighted average develop early in life, i.e. before the age
accounting for this severity distribution. of five.
GBD2010 DWs [82] for the health states
Deaths due to peanut allergen were
“Asthma: controlled” (DW = 0.009)
assumed to occur at all ages, with an
are considered applicable for mild and
average age of 37 years [102][1].
moderate cases (94%), and “Generic
uncomplicated disease: anxiety about the Sex distribution
diagnosis” (DW = 0.054) for severe cases
In the absence of information on the sex
(6%), leading to a severity-weighted
distribution of peanut allergy, a 50:50
DW of 0.012 for clinically relevant
age distribution was assumed.
peanut allergy.
Appendices
190
APPENDIX 5.
Disease models
Aflatoxin Dioxin
Disease Model Disease Model
RATIO—local RATIO—local
cases/100k births to cases/100k births to
PROB—local PROB—local PROB—local PROB—local
cases/100k population cases/100k population
PAF aflatoxin PAF aflatoxin PAF aflatoxin PAF aflatoxin
correction factor correction factor
INC
INC
symptomatic
intestinal symptomatic
trematodosis fasciolosis
INC INC
ascariosis-related ascariosis-related
severe wasting mortality
APPENDICES
WHO Estimates of the global burden of foodborne diseases
191
INC
C. botulinum
PROB—global
moderate/mild
disease
Appendices
192
PROB—global PROB—global
Chorioretinitis PROB—global
Chorioretinitis
later in life later in life Neonatal death
INC INC
Typhoid deaths Paratyphoid deaths
PROB—global
neurological
sequela
PROB—global INC
INC
chronic infection Hepatitis A
Hepatitis A
deaths
INC PROB—global
Brucella orchitis
PROB—global
case fatality ratio
APPENDICES
WHO Estimates of the global burden of foodborne diseases
193
INC PROB—local
diarrhea deaths <5 AF MRT Entam <5
INC
diarrhea deaths 5+M
PROB—local
INC AF MRT Entam 5+
diarrhea deaths 5+F
CHERG approach
INC
diarrhea 5-14
INC
INC PROB—local Cryptosporidium-
diarrhea 15-54 AF INC Crypto 5+ associated
diarrhea PROB—local
INC death
diarrhea 55+
INC
Death due to
INC PROB—local Cryptosporidium-
diarrhea deaths <5 AF MRT Crypto <5 associated
diarrhea
INC
diarrhea deaths 5+M
PROB—local National studies approach
INC AF MRT Crypto 5+
diarrhea deaths 5+F
CHERG approach
INC INC
INC PROB—local INC PROB—local Salmonella- Death due to
diarrhea <5 AF INC Shigella <5 diarrhea <5 AF INC Salmo <5 associated Salmonella-
diarrhea associated
INC INC diarrhea
diarrhea 5-14 diarrhea 5-14 National studies
INC approach
INC PROB—local INC PROB—local
diarrhea 15-54 Shigella-associated diarrhea 15-54 AF INC Salmo 5+
AF INC Shigella 5+
diarrhea
INC INC INC INC
diarrhea 55+ diarrhea 55+
Invasive non- Invasive non-
INC typhoidal typhoidal
PROB—local Death due to INC PROB—local salmonellosis <5 salmonellosis 5+
INC
AF MRT Shigella <5 Shigella-associated diarrhea deaths <5 AF MRT Salmo <5
diarrhea deaths <5
diarrhea
INC INC
diarrhea deaths 5+M diarrhea deaths 5+M
PROB—local National studies approach PROB—local PROB—global PROB—global
AF MRT Shigella 5+ INC AF MRT Salmo 5+ case fatality case fatality
INC
diarrhea deaths 5+F diarrhea deaths 5+F ratio <5 ratio 5+
CHERG approach CHERG approach
Appendices
194
STEC ETEC
Disease Model Disease Model
INC INC
INC PROB—local INC PROB—local Death due to
Campylobacter-
diarrhea <5 AF INC EPEC <5 diarrhea <5 AF INC Campy <5 Camylobacter-
associated associated
INC INC diarrhea diarrhea
diarrhea 5-14 diarrhea 5-14
INC National studies approach
INC PROB—local EPEC -associated INC PROB—local
diarrhea 15-54 AF INC EPEC 5+ diarrhea diarrhea 15-54 AF INC Campy 5+
INC
INC INC GBS
diarrhea 55+ INC diarrhea 55+
Death due to
PROB—local PROB—local PROB—global
INC EPEC -associated INC
diarrhea deaths <5 AF MRT EPEC <5 diarrhea deaths <5 AF MRT Campy <5 attributable fraction
diarrhea Campylobacter
INC INC
diarrhea deaths 5+M National studies approach diarrhea deaths 5+M
PROB—local PROB—local PROB—global
INC AF MRT EPEC 5+ INC AF MRT Campy 5+ GBS case fatality
diarrhea deaths 5+F diarrhea deaths 5+F ratio
INC PROB—local
INC
cassava cyanide diarrhea <5 AF INC
induced konzo Norovirus <5
INC
diarrhea 5-14
INC
INC PROB—local Norovirus-
PROB—global AF INC
diarrhea 15-54 associated
konzo incidence Norovirus 5+
diarrhea
expansion factor INC
diarrhea 55+
INC
PROB—local Death due to
INC Norovirus-
PROB—global PROB—global diarrhea deaths <5 AF MRT
PROB—global Norovirus <5 associated
konzo konzo diarrhea
identity factor
survival ratio case-fatality ratio INC
diarrhea deaths 5+M PROB—local National studies approach
AF MRT
INC Norovirus 5+
diarrhea deaths 5+F
CHERG approach
Peanut allergen
Disease Model
INC
INC
Death due to peanut-
Peanut-induced allergy
induced allergy
APPENDICES
WHO Estimates of the global burden of foodborne diseases
195
APPENDIX 6.
Derivation of Disability Weights
Figure A6.1 FERG hazards, causally related health states and corresponding disability weights
(DWs). The fourth column describes how the various DWs were derived from the Global Burden
of Disease Studies (GBD) and the World Health Organization Global Health Estimates (WHO/
GHE).
Trichinella spp. Acute clinical trichinellosis 0.637 Aggregate of Diarrhoea: moderate (DW =
0.202); Disfigurement: level 2, with itch or
pain (DW = 0.187); Musculoskeletal problems:
generalized, moderate (DW = 0.292); and
Infectious disease: acute episode, severe (DW
= 0.210) [84]
Trematodes
Clonorchis sinensis Abdominopelvic problems 0.123 Proxy health state of Abdominopelvic problem:
due to heavy clonorchiosis moderate
Fasciola spp. Abdominopelvic problems 0.123 Proxy health state of Abdominopelvic problem:
due to heavy fasciolosis moderate
(2)
Intestinal flukes Abdominopelvic problems 0.123 Proxy health state of Abdominopelvic problem:
due to heavy intestinal fluke moderate
infections
Opisthorchis spp. Abdominopelvic problems 0.123 Proxy health state of Abdominopelvic problem:
due to heavy opisthorchiosis moderate
Appendices
198
APPENDICES
WHO Estimates of the global burden of foodborne diseases
199
APPENDIX 7:
Attribution – Expert Elicitation Results
Table A7.1 Subregional estimates (median and 95% uncertainty interval) of the proportion of
illnesses caused by Campylobacter spp., non-typhoidal Salmonella spp., Shiga-toxin producing
Escherichia coli (STEC), Brucella spp. and Shigella spp. through each exposure pathway.
ANIMAL
HUMAN-
CONTACT
SUBREGION FOOD TO-HUMAN WATER SOIL OTHER
(DOMESTIC AND
CONTACT
WILD)
DIARRHOEAL DISEASE
Campylobacter spp.
0.57 0.18 0.04 0.09 0.00 0.06
AFR D
(0.31–0.77) (0.00–0.42) (0.00–0.22) (0.01–0.29) (0.00–0.12) (0.00–0.16)
0.57 0.17 0.04 0.09 0.00 0.06
AFR E
(0.29–0.77) (0.00–0.42) (0.00–0.23) (0.00–0.30) (0.00–0.12) (0.00–0.16)
0.73 0.10 0.00 0.11 0.00 0.00
AMR A
(0.38–0.91) (0.00–0.37) (0.00–0.20) (0.00–0.32) (0.00–0.11) (0.00–0.02)
0.68 0.11 0.03 0.08 0.00 0.06
AMR B
(0.41–0.82) (0.00–0.33) (0.00–0.21) (0.00–0.27) (0.00–0.11) (0.00–0.16)
0.67 0.12 0.03 0.08 0.00 0.06
AMR D
(0.37–0.81) (0.01–0.36) (0.00–0.21) (0.00–0.29) (0.00–0.15) (0.00–0.16)
0.67 0.11 0.03 0.07 0.00 0.06
EMR B
(0.38–0.82) (0.01–0.35) (0.00–0.27) (0.00–0.29) (0.00–0.15) (0.00–0.15)
0.67 0.11 0.03 0.07 0.00 0.06
EMR D
(0.41–0.82) (0.00–0.34) (0.00–0.22) (0.00–0.27) (0.00–0.20) (0.00–0.15)
0.76 0.08 0.01 0.06 0.01 0.00
EUR A
(0.44–0.93) (0.00–0.31) (0.00–0.13) (0.00–0.35) (0.00–0.09) (0.00–0.08)
0.66 0.11 0.03 0.12 0.03 0.00
EUR B
(0.34–0.87) (0.00–0.39) (0.00–0.21) (0.00–0.40) (0.00–0.13) (0.00–0.05)
0.66 0.11 0.03 0.12 0.03 0.00
EUR C
(0.34–0.87) (0.00–0.38) (0.00–0.23) (0.00–0.39) (0.00–0.19) (0.00–0.02)
0.57 0.13 0.11 0.05 0.03 0.02
SEAR B
(0.27–0.81) (0.00–0.36) (0.00–0.36) (0.00–0.35) (0.00–0.21) (0.00–0.06)
0.51 0.11 0.11 0.07 0.03 0.02
SEAR D
(0.03–0.79) (0.00–0.39) (0.01–0.41) (0.00–0.44) (0.00–0.32) (0.00–0.10)
0.68 0.13 0.00 0.11 0.00 0.00
WPR A
(0.40–0.89) (0.00–0.33) (0.00–0.23) (0.00–0.32) (0.00–0.08) (0.00–0.01)
0.57 0.17 0.06 0.05 0.03 0.02
WPR B
(0.25–0.82) (0.00–0.42) (0.00–0.34) (0.00–0.32) (0.00–0.15) (0.00–0.07)
Non-typhoidal Salmonella enterica
0.46 0.15 0.18 0.10 0.01 0.02
AFR D
(0.13–0.74) (0.00–0.43) (0.00–0.48) (0.00–0.39) (0.00–0.13) (0.00–0.06)
0.46 0.15 0.18 0.10 0.01 0.02
AFR E
(0.10–0.73) (0.00–0.42) (0.00–0.48) (0.00–0.40) (0.00–0.19) (0.00–0.08)
0.73 0.10 0.05 0.02 0.00 0.00
AMR A
(0.38–0.91) (0.00–0.39) (0.00–0.28) (0.00–0.22) (0.00–0.09) (0.00–0.05)
0.49 0.19 0.15 0.09 0.01 0.02
AMR B
(0.09–0.74) (0.00–0.45) (0.00–0.40) (0.00–0.32) (0.00–0.12) (0.00–0.05)
0.50 0.19 0.15 0.09 0.01 0.02
AMR D
(0.14–0.75) (0.00–0.46) (0.00–0.39) (0.00–0.31) (0.00–0.12) (0.00–0.05)
0.50 0.15 0.15 0.12 0.01 0.02
EMR B
(0.18–0.75) (0.00–0.43) (0.01–0.38) (0.00–0.33) (0.00–0.19) (0.00–0.04)
0.50 0.15 0.15 0.12 0.01 0.02
EMR D
(0.19–0.74) (0.00–0.43) (0.01–0.39) (0.00–0.32) (0.00–0.21) (0.00–0.05)
0.76 0.05 0.06 0.03 0.00 0.00
EUR A
(0.47–0.94) (0.00–0.30) (0.00–0.26) (0.00–0.21) (0.00–0.11) (0.00–0.14)
Appendices
200
ANIMAL
HUMAN-
CONTACT
SUBREGION FOOD TO-HUMAN WATER SOIL OTHER
(DOMESTIC AND
CONTACT
WILD)
0.62 0.10 0.11 0.07 0.02 0.00
EUR B
(0.31–0.84) (0.00–0.37) (0.01–0.32) (0.00–0.32) (0.00–0.12) (0.00–0.01)
0.62 0.10 0.10 0.07 0.02 0.00
EUR C
(0.32–0.84) (0.00–0.36) (0.00–0.32) (0.00–0.32) (0.00–0.12) (0.00–0.01)
0.58 0.06 0.10 0.11 0.02 0.00
SEAR B
(0.23–0.84) (0.00–0.32) (0.00–0.38) (0.00–0.40) (0.00–0.20) (0.00–0.03)
0.54 0.06 0.10 0.15 0.02 0.00
SEAR D
(0.00–0.85) (0.00–0.37) (0.00–0.42) (0.00–0.59) (0.00–0.29) (0.00–0.06)
0.74 0.09 0.04 0.01 0.00 0.00
WPR A
(0.45–0.93) (0.00–0.31) (0.00–0.28) (0.00–0.22) (0.00–0.08) (0.00–0.04)
0.57 0.10 0.12 0.08 0.02 0.00
WPR B
(0.25–0.82) (0.00–0.33) (0.00–0.35) (0.00–0.37) (0.00–0.21) (0.00–0.01)
Shiga toxin-producing E. coli
0.42 0.21 0.16 0.10 0.05 0.00
AFR D
(0.19–0.66) (0.04–0.46) (0.00–0.33) (0.00–0.30) (0.00–0.25) (0.00–0.03)
0.43 0.21 0.17 0.10 0.05 0.00
AFR E
(0.14–0.66) (0.04–0.46) (0.01–0.34) (0.00–0.34) (0.00–0.19) (0.00–0.03)
0.59 0.13 0.07 0.07 0.00 0.00
AMR A
(0.19–0.84) (0.00–0.41) (0.00–0.32) (0.00–0.31) (0.00–0.13) (0.00–0.27)
0.53 0.17 0.11 0.08 0.04 0.00
AMR B
(0.24–0.73) (0.01–0.44) (0.01–0.29) (0.00–0.32) (0.00–0.21) (0.00–0.03)
0.53 0.15 0.11 0.09 0.04 0.00
AMR D
(0.24–0.75) (0.00–0.43) (0.01–0.29) (0.00–0.32) (0.00–0.17) (0.00–0.03)
0.53 0.15 0.11 0.10 0.04 0.00
EMR B
(0.24–0.76) (0.02–0.43) (0.00–0.29) (0.00–0.37) (0.00–0.18) (0.00–0.03)
0.52 0.14 0.11 0.10 0.04 0.00
EMR D
(0.26–0.75) (0.01–0.42) (0.01–0.30) (0.00–0.37) (0.00–0.17) (0.00–0.03)
0.60 0.11 0.08 0.07 0.03 0.00
EUR A
(0.26–0.83) (0.01–0.37) (0.00–0.33) (0.00–0.33) (0.00–0.19) (0.00–0.14)
0.49 0.12 0.10 0.09 0.08 0.00
EUR B
(0.15–0.75) (0.00–0.42) (0.01–0.32) (0.00–0.38) (0.00–0.35) (0.00–0.01)
0.49 0.12 0.10 0.09 0.08 0.00
EUR C
(0.15–0.75) (0.00–0.42) (0.01–0.32) (0.00–0.36) (0.00–0.35) (0.00–0.01)
0.41 0.12 0.07 0.23 0.06 0.00
SEAR B
(0.10–0.70) (0.00–0.47) (0.00–0.31) (0.00–0.53) (0.00–0.26) (0.00–0.01)
0.40 0.13 0.06 0.23 0.06 0.00
SEAR D
(0.08–0.71) (0.00–0.47) (0.00–0.35) (0.00–0.53) (0.00–0.26) (0.00–0.02)
0.57 0.14 0.07 0.07 0.00 0.00
WPR A
(0.25–0.82) (0.00–0.36) (0.00–0.35) (0.00–0.29) (0.00–0.16) (0.00–0.24)
0.43 0.12 0.07 0.22 0.06 0.00
WPR B
(0.12–0.73) (0.00–0.44) (0.00–0.35) (0.00–0.46) (0.00–0.27) (0.00–0.01)
Brucella spp.
0.44 0.50 0.01 0.01 0.01
AFR D na
(0.10–0.68) (0.26–0.81) (0.00–0.08) (0.00–0.10) (0.00–0.06)
0.44 0.50 0.01 0.01 0.01
AFR E na
(0.06–0.70) (0.22–0.83) (0.00–0.12) (0.00–0.11) (0.00–0.06)
0.75 0.19 0.01 0.01 0.01
AMR A na
(0.28–0.93) (0.00–0.62) (0.00–0.04) (0.00–0.09) (0.00–0.12)
0.44 0.50 0.01 0.01 0.01
AMR B na
(0.09–0.69) (0.24–0.81) (0.00–0.08) (0.00–0.12) (0.00–0.08)
APPENDICES
WHO Estimates of the global burden of foodborne diseases
201
ANIMAL
HUMAN-
CONTACT
SUBREGION FOOD TO-HUMAN WATER SOIL OTHER
(DOMESTIC AND
CONTACT
WILD)
0.44 0.50 0.01 0.01 0.01
AMR D na
(0.09–0.72) (0.18–0.81) (0.00–0.12) (0.00–0.11) (0.00–0.06)
0.51 0.43 0.01 0.01 0.01
EMR B na
(0.08–0.80) (0.11–0.81) (0.00–0.07) (0.00–0.08) (0.00–0.11)
0.44 0.50 0.01 0.01 0.01
EMR D na
(0.07–0.70) (0.20–0.83) (0.00–0.14) (0.00–0.15) (0.00–0.06)
0.66 0.23 0.01 0.01 0.02
EUR A na
(0.23–0.90) (0.01–0.60) (0.00–0.04) (0.00–0.05) (0.00–0.35)
0.45 0.50 0.01 0.01 0.01
EUR B na
(0.09–0.71) (0.20–0.81) (0.00–0.07) (0.00–0.08) (0.00–0.06)
0.44 0.50 0.01 0.01 0.01
EUR C na
(0.10–0.73) (0.18–0.81) (0.00–0.06) (0.00–0.06) (0.00–0.06)
0.51 0.43 0.01 0.01 0.01
SEAR B na
(0.07–0.81) (0.10–0.81) (0.00–0.07) (0.00–0.07) (0.00–0.07)
0.45 0.50 0.01 0.01 0.01
SEAR D na
(0.07–0.70) (0.22–0.82) (0.00–0.08) (0.00–0.07) (0.00–0.06)
0.71 0.18 0.01 0.01 0.02
WPR A na
(0.28–0.92) (0.00–0.58) (0.00–0.09) (0.00–0.26) (0.00–0.30)
0.51 0.43 0.01 0.01 0.01
WPR B na
(0.07–0.80) (0.12–0.81) (0.00–0.07) (0.00–0.07) (0.00–0.07)
Shigella spp.
0.15 0.50 0.27 0.00 0.00
AFR D na
(0.00–0.52) (0.06–0.81) (0.03–0.62) (0.00–0.19) (0.00–0.13)
0.15 0.50 0.26 0.00 0.00
AFR E na
(0.00–0.51) (0.08–0.80) (0.05–0.61) (0.00–0.19) (0.00–0.16)
0.12 0.69 0.10 0.00 0.00
AMR A na
(0.00–0.46) (0.33–0.93) (0.00–0.41) (0.00–0.21) (0.00–0.06)
0.14 0.51 0.27 0.00 0.00
AMR B na
(0.00–0.52) (0.10–0.81) (0.03–0.61) (0.00–0.18) (0.00–0.06)
0.14 0.51 0.27 0.00 0.00
AMR D na
(0.00–0.52) (0.11–0.80) (0.02–0.60) (0.00–0.20) (0.00–0.02)
0.14 0.51 0.28 0.00 0.00
EMR B na
(0.00–0.52) (0.11–0.81) (0.03–0.61) (0.00–0.17) (0.00–0.02)
0.14 0.51 0.28 0.00 0.00
EMR D na
(0.00–0.52) (0.11–0.81) (0.02–0.61) (0.00–0.18) (0.00–0.02)
0.07 0.54 0.12 0.01 0.00
EUR A na
(0.00–0.46) (0.14–0.90) (0.00–0.52) (0.00–0.20) (0.00–0.55)
0.11 0.44 0.31 0.02 0.02
EUR B na
(0.00–0.50) (0.10–0.75) (0.04–0.60) (0.00–0.20) (0.00–0.21)
0.19 0.43 0.26 0.01 0.05
EUR C na
(0.00–0.51) (0.07–0.70) (0.02–0.53) (0.00–0.20) (0.00–0.22)
0.36 0.30 0.26 0.04 0.01
SEAR B na
(0.01–0.68) (0.01–0.65) (0.01–0.59) (0.00–0.21) (0.00–0.03)
0.34 0.25 0.29 0.04 0.01
SEAR D na
(0.01–0.69) (0.00–0.64) (0.01–0.65) (0.00–0.26) (0.00–0.06)
0.13 0.66 0.12 0.00 0.00
WPR A na
(0.00–0.50) (0.25–0.91) (0.00–0.42) (0.00–0.22) (0.00–0.19)
0.36 0.28 0.27 0.04 0.01
WPR B na
(0.01–0.70) (0.00–0.65) (0.01–0.60) (0.00–0.22) (0.00–0.03)
Appendices
202
ANIMAL CONTACT
SUB- HUMAN TO HUMAN
FOOD (DOMESTIC AND WATER OTHER
REGION CONTACT
WILD)
Enteropathogenic E. coli
AFR D 0.29 (0.02–0.62) 0.00 (0.00–0.33) 0.16 (0.00–0.51) 0.45 (0.12–0.76) 0.00 (0.00–0.01)
AFR E 0.29 (0.01–0.62) 0.00 (0.00–0.32) 0.16 (0.00–0.51) 0.46 (0.10–0.76) 0.00 (0.00–0.01)
AMR A 0.72 (0.20–0.97) 0.00 (0.00–0.31) 0.11 (0.00–0.53) 0.00 (0.00–0.57) 0.00 (0.00–0.01)
AMR B 0.29 (0.01–0.62) 0.00 (0.00–0.34) 0.16 (0.00–0.50) 0.46 (0.12–0.76) 0.00 (0.00–0.01)
AMR D 0.30 (0.03–0.61) 0.00 (0.00–0.33) 0.15 (0.00–0.47) 0.47 (0.13–0.74) 0.00 (0.00–0.01)
EMR B 0.31 (0.06–0.62) 0.00 (0.00–0.35) 0.14 (0.00–0.44) 0.46 (0.11–0.70) 0.00 (0.00–0.01)
EMR D 0.31 (0.05–0.62) 0.00 (0.00–0.37) 0.14 (0.00–0.44) 0.45 (0.10–0.70) 0.00 (0.00–0.01)
EUR A 0.64 (0.17–0.90) 0.05 (0.00–0.38) 0.17 (0.00–0.58) 0.03 (0.00–0.31) 0.00 (0.00–0.21)
EUR B 0.48 (0.06–0.81) 0.08 (0.00–0.41) 0.26 (0.00–0.65) 0.08 (0.00–0.43) 0.00 (0.00–0.01)
EUR C 0.48 (0.06–0.81) 0.09 (0.00–0.42) 0.26 (0.00–0.65) 0.08 (0.00–0.42) 0.00 (0.00–0.02)
SEAR B 0.29 (0.01–0.62) 0.09 (0.00–0.34) 0.29 (0.01–0.62) 0.27 (0.01–0.58) 0.00 (0.00–0.02)
SEAR D 0.29 (0.01–0.67) 0.09 (0.00–0.38) 0.27 (0.00–0.65) 0.27 (0.00–0.63) 0.00 (0.00–0.05)
WPR A 0.69 (0.16–0.94) 0.00 (0.00–0.34) 0.18 (0.00–0.66) 0.00 (0.00–0.30) 0.00 (0.00–0.02)
WPR B 0.30 (0.01–0.62) 0.14 (0.00–0.40) 0.23 (0.00–0.59) 0.26 (0.02–0.55) 0.00 (0.00–0.01)
Enterotoxigenic E. coli
AFR D 0.33 (0.09–0.65) 0.00 (0.00–0.33) 0.13 (0.00–0.44) 0.45 (0.12–0.71) 0.00 (0.00–0.01)
AFR E 0.33 (0.06–0.64) 0.00 (0.00–0.33) 0.13 (0.00–0.45) 0.45 (0.09–0.71) 0.00 (0.00–0.01)
AMR A 0.36 (0.12–0.63) 0.04 (0.00–0.32) 0.15 (0.00–0.37) 0.42 (0.11–0.66) 0.00 (0.00–0.19)
AMR B 0.34 (0.08–0.65) 0.00 (0.00–0.34) 0.12 (0.00–0.42) 0.46 (0.11–0.70) 0.00 (0.00–0.13)
AMR D 0.36 (0.07–0.68) 0.00 (0.00–0.32) 0.13 (0.00–0.43) 0.47 (0.10–0.72) 0.00 (0.00–0.01)
EMR B 0.34 (0.07–0.65) 0.00 (0.00–0.31) 0.13 (0.00–0.42) 0.49 (0.10–0.72) 0.00 (0.00–0.01)
EMR D 0.35 (0.05–0.66) 0.00 (0.00–0.31) 0.12 (0.00–0.41) 0.48 (0.12–0.73) 0.00 (0.00–0.01)
EUR A 0.42 (0.09–0.73) 0.05 (0.00–0.31) 0.26 (0.01–0.60) 0.18 (0.00–0.53) 0.00 (0.00–0.08)
EUR B 0.43 (0.05–0.73) 0.05 (0.00–0.34) 0.31 (0.02–0.66) 0.14 (0.00–0.47) 0.00 (0.00–0.18)
EUR C 0.43 (0.06–0.72) 0.05 (0.00–0.34) 0.31 (0.02–0.66) 0.14 (0.00–0.47) 0.00 (0.00–0.20)
SEAR B 0.38 (0.03–0.73) 0.05 (0.00–0.32) 0.09 (0.00–0.51) 0.39 (0.02–0.71) 0.00 (0.00–0.02)
SEAR D 0.37 (0.02–0.73) 0.06 (0.00–0.34) 0.09 (0.00–0.52) 0.38 (0.03–0.73) 0.00 (0.00–0.11)
WPR A 0.38 (0.10–0.72) 0.04 (0.00–0.29) 0.20 (0.00–0.53) 0.33 (0.00–0.61) 0.00 (0.00–0.01)
WPR B 0.38 (0.03–0.72) 0.04 (0.00–0.29) 0.08 (0.00–0.50) 0.39 (0.04–0.71) 0.00 (0.00–0.20)
Cryptosporidium spp.
AFR D 0.15 (0.00–0.44) 0.06 (0.00–0.27) 0.38 (0.01–0.72) 0.35 (0.01–0.68) 0.01 (0.00–0.16)
AFR E 0.15 (0.00–0.47) 0.05 (0.00–0.26) 0.36 (0.01–0.72) 0.37 (0.01–0.71) 0.01 (0.00–0.17)
AMR A 0.16 (0.01–0.44) 0.10 (0.01–0.42) 0.30 (0.03–0.64) 0.37 (0.08–0.72) 0.00 (0.00–0.09)
AMR B 0.11 (0.01–0.38) 0.20 (0.02–0.47) 0.35 (0.07–0.66) 0.26 (0.05–0.61) 0.00 (0.00–0.09)
AMR D 0.16 (0.01–0.44) 0.21 (0.03–0.49) 0.34 (0.07–0.66) 0.20 (0.03–0.59) 0.00 (0.00–0.08)
EMR B 0.09 (0.00–0.41) 0.14 (0.00–0.46) 0.31 (0.02–0.65) 0.36 (0.05–0.69) 0.01 (0.00–0.17)
EMR D 0.08 (0.00–0.36) 0.13 (0.00–0.43) 0.32 (0.01–0.66) 0.38 (0.06–0.71) 0.01 (0.00–0.17)
APPENDICES
WHO Estimates of the global burden of foodborne diseases
203
ANIMAL CONTACT
SUB- HUMAN TO HUMAN
FOOD (DOMESTIC AND WATER OTHER
REGION CONTACT
WILD)
EUR A 0.10 (0.00–0.39) 0.14 (0.00–0.44) 0.30 (0.01–0.65) 0.38 (0.03–0.70) 0.01 (0.00–0.09)
EUR B 0.11 (0.00–0.39) 0.16 (0.00–0.46) 0.28 (0.01–0.64) 0.37 (0.02–0.68) 0.01 (0.00–0.08)
EUR C 0.09 (0.00–0.40) 0.15 (0.00–0.48) 0.29 (0.01–0.64) 0.36 (0.05–0.70) 0.01 (0.00–0.09)
SEAR B 0.10 (0.00–0.37) 0.13 (0.00–0.46) 0.31 (0.01–0.66) 0.38 (0.02–0.71) 0.01 (0.00–0.09)
SEAR D 0.10 (0.00–0.42) 0.13 (0.00–0.46) 0.30 (0.01–0.66) 0.37 (0.03–0.71) 0.01 (0.00–0.15)
WPR A 0.10 (0.00–0.40) 0.12 (0.00–0.46) 0.29 (0.01–0.66) 0.39 (0.03–0.72) 0.01 (0.00–0.09)
WPR B 0.10 (0.00–0.45) 0.10 (0.00–0.45) 0.29 (0.01–0.66) 0.39 (0.04–0.73) 0.01 (0.00–0.10)
Giardia spp.
AFR D 0.11 (0.00–0.43) 0.03 (0.00–0.27) 0.43 (0.01–0.75) 0.33 (0.05–0.69) 0.02 (0.00–0.18)
AFR E 0.11 (0.00–0.43) 0.03 (0.00–0.25) 0.44 (0.04–0.75) 0.32 (0.04–0.67) 0.02 (0.00–0.19)
AMR A 0.11 (0.00–0.39) 0.14 (0.00–0.41) 0.25 (0.00–0.64) 0.42 (0.05–0.75) 0.00 (0.00–0.12)
AMR B 0.12 (0.00–0.42) 0.18 (0.00–0.47) 0.32 (0.01–0.67) 0.30 (0.04–0.65) 0.00 (0.00–0.09)
AMR D 0.12 (0.00–0.42) 0.18 (0.00–0.46) 0.36 (0.01–0.69) 0.26 (0.03–0.63) 0.00 (0.00–0.10)
EMR B 0.13 (0.00–0.50) 0.02 (0.00–0.15) 0.45 (0.03–0.77) 0.32 (0.03–0.71) 0.01 (0.00–0.19)
EMR D 0.13 (0.00–0.47) 0.02 (0.00–0.25) 0.39 (0.02–0.73) 0.35 (0.03–0.71) 0.01 (0.00–0.18)
EUR A 0.11 (0.00–0.44) 0.02 (0.00–0.15) 0.47 (0.02–0.79) 0.32 (0.03–0.72) 0.01 (0.00–0.14)
EUR B 0.12 (0.00–0.47) 0.02 (0.00–0.15) 0.44 (0.02–0.77) 0.34 (0.02–0.73) 0.01 (0.00–0.12)
EUR C 0.12 (0.00–0.48) 0.02 (0.00–0.15) 0.44 (0.02–0.77) 0.34 (0.04–0.74) 0.01 (0.00–0.13)
SEAR B 0.13 (0.00–0.48) 0.02 (0.00–0.23) 0.41 (0.02–0.74) 0.35 (0.02–0.72) 0.01 (0.00–0.17)
SEAR D 0.13 (0.00–0.48) 0.02 (0.00–0.22) 0.41 (0.02–0.76) 0.35 (0.03–0.72) 0.01 (0.00–0.16)
WPR A 0.12 (0.00–0.45) 0.02 (0.00–0.31) 0.46 (0.02–0.78) 0.29 (0.01–0.68) 0.01 (0.00–0.18)
WPR B 0.14 (0.00–0.49) 0.02 (0.00–0.29) 0.43 (0.02–0.75) 0.30 (0.03–0.69) 0.01 (0.00–0.19)
Appendices
204
HUMAN-TO-HUMAN
SUBREGION FOOD WATER OTHER
CONTACT
Salmonella Typhi
AFR D 0.24 (0.00–0.58) 0.22 (0.00–0.54) 0.51 (0.13–0.82) 0.00 (0.00–0.09)
AFR E 0.24 (0.00–0.58) 0.22 (0.00–0.53) 0.51 (0.16–0.81) 0.00 (0.00–0.10)
AMR A 0.26 (0.00–0.64) 0.11 (0.00–0.48) 0.57 (0.14–0.87) 0.00 (0.00–0.37)
AMR B 0.23 (0.00–0.59) 0.21 (0.00–0.53) 0.52 (0.14–0.82) 0.00 (0.00–0.10)
AMR D 0.23 (0.00–0.56) 0.21 (0.00–0.52) 0.53 (0.18–0.81) 0.00 (0.00–0.09)
EMR B 0.24 (0.00–0.58) 0.21 (0.00–0.53) 0.52 (0.15–0.82) 0.00 (0.00–0.10)
EMR D 0.24 (0.00–0.58) 0.21 (0.00–0.53) 0.52 (0.15–0.83) 0.00 (0.00–0.10)
EUR A 0.10 (0.00–0.53) 0.23 (0.00–0.72) 0.41 (0.00–0.83) 0.01 (0.00–0.66)
EUR B 0.08 (0.00–0.43) 0.47 (0.16–0.78) 0.35 (0.04–0.62) 0.02 (0.00–0.21)
EUR C 0.08 (0.00–0.43) 0.47 (0.15–0.78) 0.35 (0.03–0.62) 0.02 (0.00–0.21)
SEAR B 0.43 (0.11–0.82) 0.12 (0.00–0.49) 0.40 (0.01–0.70) 0.00 (0.00–0.03)
SEAR D 0.40 (0.01–0.81) 0.13 (0.00–0.54) 0.42 (0.00–0.80) 0.00 (0.00–0.10)
WPR A 0.33 (0.00–0.84) 0.11 (0.00–0.55) 0.48 (0.00–0.86) 0.00 (0.00–0.36)
WPR B 0.49 (0.10–0.84) 0.13 (0.00–0.51) 0.33 (0.01–0.66) 0.00 (0.00–0.03)
Vibrio cholerae
AFR D 0.21 (0.01–0.57) 0.02 (0.00–0.31) 0.72 (0.29–0.94) 0.00 (0.00–0.03)
AFR E 0.21 (0.01–0.56) 0.02 (0.00–0.30) 0.72 (0.33–0.94) 0.00 (0.00–0.04)
AMR A 0.30 (0.01–0.95) 0.02 (0.00–0.43) 0.59 (0.00–0.93) 0.00 (0.00–0.37)
AMR B 0.25 (0.00–0.58) 0.02 (0.00–0.27) 0.70 (0.33–0.95) 0.00 (0.00–0.34)
AMR D 0.25 (0.00–0.57) 0.02 (0.00–0.29) 0.69 (0.34–0.94) 0.00 (0.00–0.29)
EMR B 0.23 (0.01–0.64) 0.02 (0.00–0.30) 0.69 (0.25–0.94) 0.00 (0.00–0.03)
EMR D 0.23 (0.01–0.65) 0.02 (0.00–0.31) 0.70 (0.23–0.94) 0.00 (0.00–0.03)
EUR A 0.31 (0.00–0.85) 0.03 (0.00–0.44) 0.44 (0.00–0.86) 0.01 (0.00–0.57)
EUR B 0.46 (0.01–0.86) 0.11 (0.00–0.47) 0.36 (0.00–0.77) 0.00 (0.00–0.36)
EUR C 0.46 (0.02–0.86) 0.11 (0.00–0.47) 0.36 (0.00–0.76) 0.00 (0.00–0.38)
SEAR B 0.36 (0.04–0.78) 0.14 (0.00–0.50) 0.45 (0.02–0.79) 0.00 (0.00–0.02)
SEAR D 0.25 (0.00–0.75) 0.08 (0.00–0.50) 0.58 (0.04–0.91) 0.00 (0.00–0.02)
WPR A 0.25 (0.01–0.92) 0.04 (0.00–0.64) 0.56 (0.00–0.93) 0.00 (0.00–0.05)
WPR B 0.29 (0.01–0.74) 0.13 (0.00–0.49) 0.51 (0.04–0.83) 0.00 (0.00–0.30)
Norovirus
AFR D 0.15 (0.01–0.40) 0.68 (0.37–0.89) 0.07 (0.00–0.38) 0.04 (0.00–0.23)
AFR E 0.15 (0.00–0.40) 0.68 (0.38–0.89) 0.07 (0.00–0.37) 0.04 (0.00–0.24)
AMR A 0.23 (0.04–0.50) 0.50 (0.18–0.79) 0.22 (0.00–0.49) 0.00 (0.00–0.22)
AMR B 0.14 (0.00–0.42) 0.72 (0.36–0.90) 0.06 (0.00–0.40) 0.04 (0.00–0.24)
AMR D 0.15 (0.00–0.46) 0.72 (0.36–0.89) 0.06 (0.00–0.41) 0.04 (0.00–0.23)
EMR B 0.15 (0.00–0.40) 0.72 (0.43–0.89) 0.07 (0.00–0.30) 0.04 (0.00–0.22)
EMR D 0.15 (0.00–0.40) 0.72 (0.42–0.89) 0.06 (0.00–0.32) 0.04 (0.00–0.23)
EUR A 0.26 (0.00–0.73) 0.43 (0.00–0.83) 0.17 (0.00–0.58) 0.00 (0.00–0.36)
EUR B 0.23 (0.01–0.57) 0.32 (0.02–0.67) 0.33 (0.00–0.65) 0.04 (0.00–0.34)
APPENDICES
WHO Estimates of the global burden of foodborne diseases
205
HUMAN-TO-HUMAN
SUBREGION FOOD WATER OTHER
CONTACT
EUR C 0.23 (0.01–0.57) 0.33 (0.02–0.67) 0.33 (0.01–0.63) 0.04 (0.00–0.33)
SEAR B 0.12 (0.00–0.48) 0.53 (0.13–0.83) 0.21 (0.00–0.53) 0.00 (0.00–0.42)
SEAR D 0.15 (0.00–0.55) 0.46 (0.00–0.79) 0.29 (0.00–0.72) 0.00 (0.00–0.35)
WPR A 0.22 (0.01–0.52) 0.48 (0.12–0.77) 0.22 (0.00–0.51) 0.00 (0.00–0.32)
WPR B 0.15 (0.00–0.55) 0.46 (0.00–0.79) 0.28 (0.01–0.68) 0.00 (0.00–0.34)
Hepatitis A
AFR D 0.36 (0.07–0.63) 0.40 (0.10–0.68) 0.17 (0.00–0.49) 0.04 (0.00–0.10)
AFR E 0.29 (0.07–0.57) 0.36 (0.08–0.64) 0.30 (0.06–0.59) 0.02 (0.00–0.06)
AMR A 0.42 (0.06–0.77) 0.46 (0.04–0.78) 0.01 (0.00–0.19) 0.10 (0.00–0.32)
AMR B 0.31 (0.03–0.60) 0.46 (0.16–0.74) 0.11 (0.00–0.39) 0.09 (0.00–0.21)
AMR D 0.32 (0.03–0.61) 0.35 (0.11–0.65) 0.26 (0.04–0.57) 0.04 (0.00–0.09)
EMR B 0.35 (0.04–0.61) 0.42 (0.17–0.69) 0.15 (0.02–0.34) 0.09 (0.00–0.20)
EMR D 0.32 (0.02–0.59) 0.36 (0.11–0.66) 0.22 (0.00–0.49) 0.08 (0.00–0.23)
EUR A 0.42 (0.02–0.75) 0.46 (0.10–0.79) 0.01 (0.00–0.17) 0.10 (0.00–0.32)
EUR B 0.35 (0.12–0.59) 0.35 (0.18–0.61) 0.20 (0.01–0.36) 0.08 (0.00–0.19)
EUR C 0.34 (0.08–0.60) 0.42 (0.17–0.69) 0.14 (0.00–0.35) 0.09 (0.00–0.24)
SEAR B 0.34 (0.05–0.60) 0.35 (0.14–0.65) 0.23 (0.04–0.55) 0.04 (0.00–0.09)
SEAR D 0.29 (0.04–0.56) 0.37 (0.13–0.64) 0.29 (0.06–0.56) 0.02 (0.00–0.06)
WPR A 0.42 (0.03–0.76) 0.46 (0.10–0.79) 0.01 (0.00–0.16) 0.10 (0.00–0.29)
WPR B 0.34 (0.02–0.64) 0.36 (0.06–0.66) 0.21 (0.01–0.47) 0.08 (0.00–0.20)
Entamoeba histolytica
AFR D 0.30 (0.00–0.68) 0.37 (0.00–0.73) 0.25 (0.00–0.63) 0.04 (0.00–0.21)
AFR E 0.30 (0.00–0.68) 0.37 (0.00–0.72) 0.24 (0.00–0.62) 0.04 (0.00–0.22)
AMR A 0.25 (0.00–0.70) 0.34 (0.00–0.76) 0.33 (0.00–0.74) 0.00 (0.00–0.19)
AMR B 0.21 (0.00–0.62) 0.38 (0.02–0.76) 0.32 (0.00–0.70) 0.00 (0.00–0.20)
AMR D 0.17 (0.00–0.58) 0.37 (0.04–0.76) 0.37 (0.01–0.73) 0.00 (0.00–0.20)
EMR B 0.24 (0.00–0.62) 0.42 (0.01–0.76) 0.24 (0.00–0.62) 0.04 (0.00–0.22)
EMR D 0.28 (0.00–0.66) 0.39 (0.00–0.75) 0.25 (0.00–0.65) 0.04 (0.00–0.22)
EUR A 0.33 (0.00–0.71) 0.49 (0.03–0.83) 0.15 (0.00–0.51) 0.01 (0.00–0.16)
EUR B 0.30 (0.00–0.66) 0.42 (0.02–0.76) 0.20 (0.00–0.59) 0.04 (0.00–0.20)
EUR C 0.26 (0.00–0.64) 0.42 (0.02–0.76) 0.23 (0.00–0.61) 0.04 (0.00–0.19)
SEAR B 0.26 (0.00–0.65) 0.38 (0.00–0.75) 0.28 (0.00–0.68) 0.04 (0.00–0.18)
SEAR D 0.25 (0.00–0.63) 0.37 (0.00–0.72) 0.29 (0.01–0.69) 0.04 (0.00–0.19)
WPR A 0.25 (0.00–0.62) 0.41 (0.00–0.74) 0.26 (0.01–0.62) 0.04 (0.00–0.25)
WPR B 0.27 (0.00–0.63) 0.41 (0.00–0.73) 0.24 (0.01–0.62) 0.05 (0.00–0.23)
Appendices
206
ANIMAL
HUMAN-
CONTACT
SUBREGION FOOD TO-HUMAN WATER SOIL AIR OTHER
(DOMESTIC
CONTACT
AND WILD)
PARASITIC DISEASE
Toxoplasma gondii
0.48 0.01 0.11 0.36
AFR D na na na
(0.24–0.76) (0.00–0.20) (0.00–0.37) (0.07–0.57)
0.42 0.01 0.16 0.38
AFR E na na na
(0.20–0.70) (0.00–0.19) (0.02–0.41) (0.05–0.58)
0.60 0.01 0.19 0.19
AMR A na na na
(0.30–0.81) (0.00–0.28) (0.01–0.42) (0.00–0.46)
0.52 0.01 0.23 0.22
AMR B na na na
(0.27–0.77) (0.00–0.20) (0.01–0.45) (0.00–0.46)
0.53 0.01 0.23 0.22
AMR D na na na
(0.27–0.77) (0.00–0.21) (0.02–0.44) (0.00–0.45)
0.52 0.01 0.11 0.34
EMR B na na na
(0.27–0.80) (0.00–0.20) (0.01–0.29) (0.02–0.56)
0.53 0.01 0.23 0.22
EMR D na na na
(0.29–0.77) (0.00–0.20) (0.02–0.43) (0.00–0.42)
0.61 0.01 0.19 0.18
EUR A na na na
(0.35–0.82) (0.00–0.21) (0.02–0.36) (0.00–0.40)
0.45 0.01 0.15 0.37
EUR B na na na
(0.23–0.76) (0.00–0.20) (0.02–0.35) (0.01–0.58)
0.53 0.01 0.23 0.22
EUR C na na na
(0.31–0.78) (0.00–0.20) (0.03–0.41) (0.01–0.41)
0.52 0.01 0.23 0.22
SEAR B na na na
(0.26–0.77) (0.00–0.19) (0.03–0.45) (0.00–0.43)
0.43 0.01 0.27 0.26
SEAR D na na na
(0.09–0.73) (0.00–0.22) (0.03–0.58) (0.00–0.56)
0.60 0.01 0.19 0.18
WPR A na na na
(0.33–0.81) (0.00–0.21) (0.02–0.37) (0.00–0.43)
0.53 0.01 0.23 0.22
WPR B na na na
(0.29–0.77) (0.00–0.20) (0.04–0.43) (0.00–0.43)
Echinococcus granulosus
0.21 0.51 0.18 0.09 0.00 0.00
AFR D na
(0.07–0.42) (0.25–0.72) (0.01–0.34) (0.00–0.20) (0.00–0.06) (0.00–0.01)
0.20 0.52 0.18 0.09 0.00 0.00
AFR E na
(0.05–0.40) (0.27–0.73) (0.00–0.35) (0.00–0.19) (0.00–0.06) (0.00–0.06)
0.20 0.52 0.17 0.09 0.00 0.00
AMR A na
(0.03–0.40) (0.30–0.75) (0.00–0.31) (0.00–0.20) (0.00–0.14) (0.00–0.01)
0.20 0.52 0.18 0.09 0.00 0.00
AMR B na
(0.02–0.43) (0.28–0.73) (0.00–0.34) (0.00–0.22) (0.00–0.14) (0.00–0.01)
0.21 0.51 0.18 0.09 0.00 0.00
AMR D na
(0.05–0.41) (0.29–0.72) (0.01–0.35) (0.00–0.23) (0.00–0.13) (0.00–0.01)
0.21 0.51 0.17 0.09 0.00 0.00
EMR B na
(0.05–0.43) (0.28–0.73) (0.00–0.32) (0.00–0.19) (0.00–0.14) (0.00–0.06)
0.21 0.52 0.18 0.09 0.00 0.00
EMR D na
(0.06–0.41) (0.28–0.72) (0.00–0.32) (0.00–0.18) (0.00–0.14) (0.00–0.01)
0.21 0.51 0.18 0.09 0.00 0.00
EUR A na
(0.04–0.40) (0.29–0.72) (0.00–0.33) (0.00–0.20) (0.00–0.14) (0.00–0.01)
APPENDICES
WHO Estimates of the global burden of foodborne diseases
207
ANIMAL
HUMAN-
CONTACT
SUBREGION FOOD TO-HUMAN WATER SOIL AIR OTHER
(DOMESTIC
CONTACT
AND WILD)
0.21 0.52 0.18 0.09 0.00 0.00
EUR B na
(0.06–0.40) (0.27–0.73) (0.00–0.33) (0.00–0.19) (0.00–0.15) (0.00–0.01)
0.21 0.51 0.18 0.09 0.00 0.00
EUR C na
(0.04–0.40) (0.26–0.73) (0.00–0.35) (0.00–0.21) (0.00–0.15) (0.00–0.01)
0.21 0.51 0.18 0.09 0.00 0.00
SEAR B na
(0.03–0.44) (0.22–0.73) (0.00–0.35) (0.00–0.19) (0.00–0.13) (0.00–0.01)
0.20 0.52 0.18 0.09 0.00 0.00
SEAR D na
(0.06–0.40) (0.29–0.73) (0.00–0.34) (0.00–0.19) (0.00–0.14) (0.00–0.01)
0.20 0.53 0.18 0.09 0.00 0.00
WPR A na
(0.01–0.39) (0.30–0.75) (0.00–0.33) (0.00–0.20) (0.00–0.13) (0.00–0.01)
0.21 0.51 0.17 0.09 0.00 0.00
WPR B na
(0.05–0.43) (0.29–0.73) (0.00–0.32) (0.00–0.21) (0.00–0.14) (0.00–0.01)
Echinococcus multilocularis
0.58 0.02 0.20 0.20 0.00 0.00
AFR D na
(0.00–0.87) (0.00–0.42) (0.00–0.61) (0.00–0.63) (0.00–0.03) (0.00–0.00)
0.58 0.02 0.20 0.20 0.00 0.00
AFR E na
(0.00–0.87) (0.00–0.41) (0.00–0.62) (0.00–0.61) (0.00–0.03) (0.00–0.00)
0.51 0.03 0.17 0.16 0.00 0.00
AMR A na
(0.13–0.79) (0.00–0.50) (0.01–0.40) (0.01–0.38) (0.00–0.11) (0.00–0.03)
0.58 0.02 0.20 0.20 0.00 0.00
AMR B na
(0.00–0.87) (0.00–0.38) (0.00–0.62) (0.00–0.61) (0.00–0.03) (0.00–0.00)
0.58 0.02 0.19 0.20 0.00 0.00
AMR D na
(0.00–0.88) (0.00–0.41) (0.00–0.61) (0.00–0.60) (0.00–0.03) (0.00–0.00)
0.43 0.14 0.17 0.17 0.00 0.00
EMR B na
(0.09–0.73) (0.00–0.55) (0.00–0.42) (0.00–0.42) (0.00–0.06) (0.00–0.01)
0.48 0.12 0.20 0.20 0.00 0.00
EMR D na
(0.00–0.77) (0.00–0.49) (0.00–0.54) (0.00–0.53) (0.00–0.04) (0.00–0.00)
0.52 0.03 0.17 0.16 0.00 0.00
EUR A na
(0.15–0.79) (0.00–0.48) (0.01–0.40) (0.00–0.39) (0.00–0.11) (0.00–0.03)
0.45 0.13 0.18 0.17 0.00 0.00
EUR B na
(0.12–0.72) (0.00–0.52) (0.02–0.38) (0.00–0.37) (0.00–0.13) (0.00–0.03)
0.44 0.14 0.17 0.17 0.00 0.00
EUR C na
(0.12–0.72) (0.00–0.53) (0.01–0.38) (0.00–0.37) (0.00–0.12) (0.00–0.03)
0.58 0.02 0.20 0.20 0.00 0.00
SEAR B na
(0.00–0.88) (0.00–0.41) (0.00–0.61) (0.00–0.61) (0.00–0.03) (0.00–0.00)
0.58 0.02 0.20 0.20 0.00 0.00
SEAR D na
(0.00–0.88) (0.00–0.37) (0.00–0.62) (0.00–0.60) (0.00–0.05) (0.00–0.00)
0.51 0.04 0.16 0.16 0.00 0.00
WPR A na
(0.09–0.81) (0.00–0.52) (0.00–0.41) (0.00–0.40) (0.00–0.03) (0.00–0.01)
0.48 0.12 0.20 0.20 0.00 0.00
WPR B na
(0.00–0.78) (0.00–0.49) (0.00–0.54) (0.00–0.54) (0.00–0.12) (0.00–0.03)
Ascaris spp.
0.38 0.00 0.00 0.19 0.39 0.00
AFR D na
(0.10–0.66) (0.00–0.09) (0.00–0.08) (0.07–0.40) (0.07–0.65) (0.00–0.06)
0.38 0.00 0.00 0.19 0.39 0.00
AFR E na
(0.07–0.67) (0.00–0.09) (0.00–0.09) (0.07–0.41) (0.05–0.65) (0.00–0.06)
0.83 0.00 0.00 0.05 0.06 0.00
AMR A na
(0.43–0.97) (0.00–0.29) (0.00–0.08) (0.00–0.18) (0.00–0.42) (0.00–0.06)
0.55 0.00 0.00 0.19 0.22 0.00
AMR B na
(0.17–0.75) (0.00–0.13) (0.00–0.09) (0.06–0.40) (0.05–0.50) (0.00–0.04)
Appendices
208
ANIMAL
HUMAN-
CONTACT
SUBREGION FOOD TO-HUMAN WATER SOIL AIR OTHER
(DOMESTIC
CONTACT
AND WILD)
0.37 0.00 0.00 0.18 0.41 0.00
AMR D na
(0.07–0.68) (0.00–0.15) (0.00–0.08) (0.05–0.41) (0.04–0.69) (0.00–0.04)
0.55 0.00 0.00 0.20 0.22 0.00
EMR B na
(0.15–0.77) (0.00–0.10) (0.00–0.07) (0.02–0.44) (0.02–0.51) (0.00–0.06)
0.55 0.00 0.00 0.20 0.21 0.00
EMR D na
(0.18–0.75) (0.00–0.10) (0.00–0.09) (0.04–0.43) (0.04–0.51) (0.00–0.05)
0.85 0.00 0.00 0.05 0.06 0.00
EUR A na
(0.47–0.97) (0.00–0.25) (0.00–0.09) (0.00–0.18) (0.00–0.38) (0.00–0.06)
0.55 0.00 0.00 0.19 0.22 0.00
EUR B na
(0.13–0.76) (0.00–0.27) (0.00–0.10) (0.03–0.40) (0.02–0.50) (0.00–0.06)
0.55 0.00 0.00 0.19 0.22 0.00
EUR C na
(0.14–0.76) (0.00–0.25) (0.00–0.12) (0.03–0.40) (0.04–0.50) (0.00–0.05)
0.54 0.00 0.00 0.20 0.22 0.00
SEAR B na
(0.18–0.75) (0.00–0.14) (0.00–0.08) (0.03–0.44) (0.01–0.52) (0.00–0.05)
0.39 0.00 0.00 0.20 0.38 0.00
SEAR D na
(0.11–0.68) (0.00–0.12) (0.00–0.07) (0.04–0.44) (0.04–0.65) (0.00–0.06)
0.85 0.00 0.00 0.05 0.06 0.00
WPR A na
(0.47–0.97) (0.00–0.23) (0.00–0.09) (0.00–0.19) (0.00–0.37) (0.00–0.06)
0.54 0.00 0.00 0.20 0.21 0.00
WPR B na
(0.16–0.77) (0.00–0.24) (0.00–0.11) (0.02–0.43) (0.02–0.49) (0.00–0.06)
APPENDICES
Table A7.5 Subregional estimates (median and 95% uncertainty interval) of the proportion of illnesses caused by exposure to lead through
209
each pathway.
COOKWARE,
SUBREGION FOOD WATER SOIL AIR PAINT POTTERY OR TOYS OTHER
GLASSWARE
LEAD
Period 2006 1996-2006 2010 2008 2010 2010 2005 2010 2010
Formal expert Derived by the Formal expert Formal expert Derived by the Formal expert Formal expert Formal expert Formal expert
Method
elicitation authors b) elicitation elicitation authors b) elicitation elicitation elicitation elicitation
depended on
Only domestically acquired cases yes no yes yes no no yes no
the data used
HAZARDS
Campylobacter spp. 42 (16-84) 55 (30-80) 76 (44-93) 68 (54-82) 80 (73-86) 73 (38-91) 56 (26-82) 76 (70-80) 68 (40-89)
Non-typhoidal Salmonella spp. 55 (32-88) 95 (55-95) 76 (47-94) 80 (68-92) 94 (91-96) 73 (38-91) 60 (18-83) 71 (65-75) 74 (45-93)
Shiga toxin-producing E. coli 42 (21-78) 51 (40-90) 60 (26-83) 76 (60-91) 82 (75-87) 59 (19-84) 40 (6-95) 55 (30-75) 57 (25-82)
APPENDICES
211
APPENDIX 8.
DATA TABLES FOR INDIVIDUAL HAZARD CLASSES: ENTERIC, PARASITIC,
CHEMICAL1
Table A8.1 Median number of foodborne Illnesses, Deaths, and Disability Adjusted Life Years (DALYs), with 95% uncertainty
intervals, 2010.
ILLNESSES DEATHS DALYs PROPORTION FOODBORNE FOODBORNE ILLNESSES FOODBORNE DEATHS FOODBORNE
PATHOGEN
(95% UI) (95% UI) (95% UI) (95% UI) (95% UI) (95% UI) DALYs (95% UI)
1 912 159 038
715 196 55 139 959 0.29 548 285 159 199 892 15 780 400
Diarrhoeal Disease (1 413 002 730–
(603 325–846 397) (46 746 114–65 120 623) (0.22–0.36) (369 733 377–888 360 956) (136 903–286 616) (11 043 288–22 251 264)
2 849 323 016)
256 775 0 45 256 775 0 45
Bacillus cereus** 1.00
(43 875–807 547) (0–0) (7–171) (43 875–807 547) (0–0) (7–171)
2 000 626 631 581 902 722
1 092 584 78 730 084 0.29 350 686 25 175 035
TOTAL (1 494 986 030– (400 741 151–
(892 999–1 374 238) (64 963 913–97 740 062) (0.23–0.36) (240 030–524 042) (17 547 264–37 021 003)
2 942 534 533) 922 031 380)
Notes: * = Includes Guillain-Barré Syndrome cases and deaths; ** = 61 EUR and other subregion A (low mortality) countries only; *** 61 EUR and subregion A (low mortality) countries only, and excluding WPR
A countries
1
Note that non-typhoidal Salmonella enterica was split over diarrhoeal and invasive disease, whereas in Table 7 it was exclusively listed under diarrhoeal disease agents.
212
APPENDICES
Table A8.2 Median rates of foodborne illnesses, deaths and Disability Adjusted Life Years (DALYs) per 100 000 persons, by region, with 95% uncertainty
213
intervals, 2010.
108 0.7 53 10 0.07 5 73 0.5 37 1 0.007 0.5 250 2 128 77 0.5 33 110 0.8 54
Salmonella enterica Typhi
(24–317) (0.2–2) (12–155) (2–32) (0.01–0.2) (0.9–16) (9–240) (0.06–2) (5–122) (0.1–5) (0–0.03) (0.05–2) (50–725) (0.4–5) (29–361) (12–203) (0.07–1) (5–87) (34–272) (0.2–2) (17–133)
10 304 1 001 7 937 16 865 470 2 506 8 068 622 6 491 8 369 366
14 0.8 61 6 0.5 32 9 1 93 5
TOTAL (4 279– (562– (4 515– (8 051– (286– (1 455– (3 294– (306– (2 630– (5 723– (255–
(8–21) (0.5–1) (40–93) (4–9) (0.3–0.6) (24–45) (4–17) (0.6–2) (45–175) (3–8)
22 108) 1 543) 13 899) 34 712) 728) 4 168) 20 663) 1 145) 17 528) 13 318) 538)
Notes: * Table does not include four foodborne intoxications caused by Clostridium botulinum, Cl. perfringens, S. aureus, and Bacillus cereus due to a lack of data for global estimation.
.
214
APPENDICES
Table A8.3 Median number of Illnesses, Deaths, and Disability Adjusted Life Years (DALYs) by age group, with 95% uncertainty
215
intervals, 2010.
AGE GROUP: <5 YEARS OF AGE AGE GROUP: *5 YEARS OF AGE RATIO <5:*5
47 988 357 42 883 268
13 861 1 383 499 7 436 750 578 1.11 1.91 1.87
Campylobacter spp.** (22 436 891– (18 350 672–
(8 754–23 670) (911 878–2 279 897) (4 930–9 974) (540 003–956 663) (0.34–3.47) (1.21–3.08) (1.26–2.92)
102 663 926) 112 061 441)
5 986 213
1 989 185 057 2 253 036 1 673 104 794 2.61 1.23 1.83
Cryptosporidium spp. (2 569 532–
(678–5 683) (64 847–518 497) (774 628–8 639 265) (638–4 149) (40 408–256 055) (0.69–8.01) (0.42–2.72) (0.65–3.93)
12 738 924)
8 480 759 17 828 477
896 92 213 524 43 984 0.48 1.75 2.14
Entamoeba histolytica (1 593 697– (5 378 578–
(90–4 852) (15 997–444 002) (218–1 110) (20 149–85 551) (0.08–2.38) (0.18–8.71) (0.38–9.49)
30 849 576) 50 963 825)
17 312 780
22 156 2 004 543 5 458 601 14 647 911 012 3.20 1.52 2.21
Enteropathogenic E. coli (6 767 766–
(11 944–37 473) (1 084 856–3 389 584) (2 145 370–16 561 005) (7 305–25 447) (457 215–1 575 768) (0.85–11.78) (1.03–2.29) (1.52–3.29)
54 104 398)
38 352 806 46 811 878
14 056 1 303 490 11 933 767 975 0.82 1.21 1.74
Enterotoxigenic E. coli (21 144 875– (20 306 649–
(7 045–26 784) (668 837–2 446 758) (6 382–18 887) (419 834–1 204 273) (0.35–1.96) (0.63–2.10) (0.95–2.93)
64 795 160) 103 801 449)
18 773 028 8 693 968
0 20 677 0 5 016 2.11 4.04
Giardia spp. (8 075 497– (3 337 657– N/A
(0–0) (8 552–44 101) (0–0) (1 945–13 791) (0.84–5.22) (1.57–10.28)
38 649 748) 24 195 602)
34 582 700 89 056 582
8 992 844 376 25 807 1 638 925 0.38 0.35 0.52
Norovirus (19 595 826– (46 054 795–
(4 251–19 347) (406 822–1 776 252) (11 201–61 642) (730 924–3 844 771) (0.19–0.73) (0.22–0.54) (0.33–0.78)
59 592 939) 206 532 318)
15 274 234 60 293 254
Salmonella enterica, non- 12 531 1 149 675 15 807 1 016 047 0.26 0.84 1.19
(6 514 539– (18 488 275–
typhoidal (6 562–30 779) (609 216–2 792 992) (8 762–21 942) (576 408–1 405 079) (0.06–1.16) (0.44–1.83) (0.64–2.57)
41 696 874) 189 066 838)
34 049 173
15 516 627 8 863 819 280 6 060 404 144 0.45 1.49 2.06
Shigella spp. (10 186 959–
(5 416 319–38 620 351) (3 250–20 925) (309 576–1 909 450) (2 734–11 511) (188 009–749 866) (0.13–1.70) (0.60–3.26) (0.87–4.43)
95 312 884)
21 182 632 6 900 776
4 336 215 23 727 2 180 916 123 026 0.21 0.19 0.32
Invasive enteric diseases (8 375 340– (3 799 471–
(1 675 945–9 422 681) (11 866–45 950) (1 085 765–4 219 254) (69 306–230 318) (0.15–0.23) (0.14–0.22) (0.23–0.35)
49 059 198) 13 355 093)
(95% UI) (95% UI) (95% UI) (95% UI) (95% UI) (95% UI) (95% UI) (95% UI) (95% UI)
11 544 593
2 165 243 4 380 411 592 23 351 941 278 0.19 0.19 0.44
Hepatitis A (3 057 415–
(573 433–6 084 381) (1 132–12 211) (112 767–1 130 290) (6 036–65 109) (269 448–2 538 627) (0.19–0.19) (0.19–0.19) (0.39–0.46)
32 440 565)
Salmonella enterica, invasive non- 45 549 4 700 421 523 239 467 24 692 1 373 635 0.19 0.19 0.31
typhoidal (25 019–62 638) (2 268–8 188) (203 340–733 940) (142 115–321 539) (12 655–42 246) (684 718–2 373 326) (0.17–0.20) (0.17–0.20) (0.29–0.32)
APPENDICES
Table A8.4 Median rate per 100 000 of foodborne Illnesses, Deaths and Disability Adjusted Life Years (DALYs) by region, with 95% uncertainty intervals, 2010.
217
APPENDICES
TableA8.5 Median number of total and foodborne Illnesses, Deaths, and Disability Adjusted Life Years (DALYs), with 95% uncertainty intervals, 2010
219
PROPORTION
PROPORTION FOODBORNE
ILLNESSES DEATHS DALYS FOODBORNE – ILLNES- FOODBORNE DEATHS FOODBORNE DALYS
PATHOGEN FOODBORNE – DALYS ILLNESSES
(95% UI) (95% UI) (95% UI) SES (95% UI) (95% UI)
(95% UI) (95% UI)
(95% UI)
356 688 438 67 182 645
33 925 2 940 604 0.19 0.17 5 558 492 354
Enteric protozoa (251 929 267– (35 794 977–
(22 933–53 614) (1 978 442–4 686 855) (0.12–0.28) (0.09–0.29) (2 593–11 958) (239 400–1 034 790)
517 872 930) 120 556 797)
64 003 709
27 553 2 159 331 0.13 0.14 8 584 805 3 759 296 156
Cryptosporidium spp.# (43 049 455–
(18 532–44 654) (1 392 438–3 686 925) (0.07–0.24) (0.06–0.28) (3 897 252–18 531 196) (1 520–9 115) (119 456–724 660)
104 679 951)
103 943 952
5 450 515 904 0.28 0.28 28 023 571 1 470 138 863
Entamoeba histolytica# (47 018 659–
(2 194–17 127) (222 446–1 552 466) (0.14–0.44) (0.13–0.47) (10 261 254–68 567 590) (453–5 554) (47 339–503 775)
210 632 459)
183 842 615 28 236 123
0 171 100 0.15 0.15 0 26 270
Giardia spp.# (130 018 020– (12 945 655–
(0–0) (115 777–257 315) (0.08–0.27) (0.07–0.27) (0–0) (11 462–53 577)
262 838 002) 56 996 454)
Invasive infectious 20 817 916 1 409 1 684 414 0.49 0.49 10 280 089 684 829 071
disease (16 337 908–29 091 701) (701–2 620) (1 236 005–2 452 060) (0.40–0.59) (0.40–0.59) (7 403 516–14 904 324) (333–1 300) (561 297–1 264 567)
Toxoplasma gondii, 98 900 1 409 526 515 0.49 0.49 48 823 684 259 618
congenital (67 858–188 748) (701–2 620) (359 756–835 537) (0.40–0.58) (0.40–0.58) (31 893–93 213) (333–1 300) (168 510–422 935)
48 405 537
TOTAL (excluding 59 724 8 777 198 0.48 0.76 23 220 595 45 927 6 639 989
(43 376 746–
enteric protozoa) (48 017–83 616) (7 620 016–12 511 566) (0.38–0.56) (0.65–0.81) (18 215 499–38 081 817) (34 763–59 933) (5 611 955–8 414 684)
79 049 913)
407 149 528 91 148 998
94 620 11 794 391 0.22 0.61 51 909 7 161 689
TOTAL (301 670 420– (58 576 614–
(77 140–125 670) (10 164 903–15 675 990) (0.16–0.31) (0.52–0.67) (40 020–66 992) (6 078 248–9 074 283)
585 323 226) 153 950 104)
220
APPENDICES
WHO Estimates of the global burden of foodborne diseases
221
Table A8.6 Median number of foodborne Illnesses, Deaths, and Disability Adjusted Life Years
(DALYs), with 95% Uncertainty Intervals, 2010
Notes: * = Only the burdens for AMR A, EUR A and WPR A were assessed.
Table A8.7 Median rate per 100 000 foodborne (FB) Illnesses, Deaths, and Disability Adjusted
Life Years (DALYs) by region, with 95% uncertainty intervals, 2010.
CHEMICAL
REGION CYANIDE IN
AFLATOXIN DIOXIN TOTAL
CASSAVA
FB Illnesses (95% UI) 0.4 (0.1–1) 0.1 (0.01–0.4) 0.2 (0.07–7) 0.7 (0.3–8)
AFR FB Deaths (95% UI) 0.4 (0.1–1) 0.03 (0.003–0.08) 0 (0–0) 0.4 (0.1–1)
FB DALYs (95% UI) 15 (5–40) 2 (0.2–6) 0.2 (0.07–8) 18 (7–49)
FB Illnesses (95% UI) 0.08 (0.02–0.6) 0 (0–0) 0.2 (0.05–6) 0.2 (0.1–7)
AMR FB Deaths (95% UI) 0.08 (0.02–0.6) 0 (0–0) 0 (0–0) 0.08 (0.02–0.6)
FB DALYs (95% UI) 2 (0.4–15) 0 (0–0) 0.2 (0.07–9) 2 (0.6–24)
FB Illnesses (95% UI) 0.2 (0.04–0.5) 0 (0–0) 2 (1–35) 2 (1–35)
EMR FB Deaths (95% UI) 0.1 (0.04–0.4) 0 (0–0) 0 (0–0) 0.1 (0.04–0.4)
FB DALYs (95% UI) 4 (1–13) 0 (0–0) 2 (2–43) 7 (3–51)
FB Illnesses (95% UI) 0.02 (0.01–0.03) 0 (0–0) 1 (0.7–13) 1 (0.7–13)
EUR FB Deaths (95% UI) 0.02 (0.01–0.03) 0 (0–0) 0 (0–0) 0.02 (0.01–0.03)
FB DALYs (95% UI) 0.5 (0.3–0.8) 0 (0–0) 1 (0.9–19) 2 (1–19)
FB Illnesses (95% UI) 0.2 (0.08–0.6) 0 (0–0) 9 (8–32) 10 (8–32)
SEAR FB Deaths (95% UI) 0.2 (0.08–0.5) 0 (0–0) 0 (0–0) 0.2 (0.07–0.5)
FB DALYs (95% UI) 7 (2–17) 0 (0–0) 12 (10–41) 19 (13–54)
FB Illnesses (95% UI) 0.6 (0.1–2) 0 (0–0) 0.05 (0.005–4) 0.8 (0.1–5)
WPR FB Deaths (95% UI) 0.5 (0.09–2) 0 (0–0) 0 (0–0) 0.5 (0.09–2)
FB DALYs (95% UI) 16 (3–63) 0 (0–0) 0.07 (0.007–6) 16 (3–65)
FB Illnesses (95% UI) 0.3 (0.1–0.8) 0.02 (0.002–0.04) 3 (2–16) 3 (3–17)
GLOBAL FB Deaths (95% UI) 0.3 (0.1–0.7) 0.003 (0–0.01) 0 (0–0) 0.3 (0.1–0.8)
FB DALYs (95% UI) 9 (4–24) 0.3 (0.03–0.8) 3 (3–20) 13 (7–39)
Appendices
222
APPENDICES
223
REFERENCES
24. Painter, J.A., Hoekstra, R.M., 31. Tyshenko, M.G., El Saadany, S.,
Ayers, T., Tauxe, R.V., Braden, C.R., Oraby, T., Darshan, S., Aspinall, W.,
Angulo, F.J., Griffin, P.M. 2013. Cooke, R., Catford, A. & Krewski,
Attribution of foodborne ILLNES- D. 2011. Expert elicitation for the
SES, hospitalizations, and Deaths judgment of prion disease risk
to food commodities by using uncertainties. Journal of Toxicology
outbreak data, United States, and Environmental Health-Part
1998–2008. Emerging Infectious A-Current Issues, 74(2-4; Special
Diseases,19(3): 407–415. issue): 261–285.
25. Boysen, L., Rosenquist, H., Larsson, 32. Hoffmann, S., Fischbeck, P.,
J.T., Nielsen, E.M., Sorensen, G., Krupnick, A. & McWilliams, M. 2007.
Nordentoft, S. & Hald, T. 2014. Using expert elicitation to link
Source attribution of human foodborne illnesses in the United
Campylobacteriosis in Denmark. States to foods. Journal of Food
Epidemiology and Infection, Protection, 70(5): 1220–1229.
142(8): 1599–1608.
33. Havelaar, A.H., Galindo, A.V.,
26. De Knegt, L.V., Pires, S.M. & Hald, Kurowicka, D. & Cooke, R.M. 2008.
T. 2015. Attributing foodborne Attribution of foodborne pathogens
salmonellosis in humans to animal using structured expert elicitation.
reservoirs in the European Union Foodborne Pathogens and Disease,
using a multi-country stochastic 5(5): 649–659.
model. Epidemiology and infection,
34. Rice, J. 2007. Mathematical statistics
143(6): 1175–1186.
and data analysis. 3rd ed. Thomson
27. Cairncross, S., Hunt, C., Boisson, S., Higher Education, Belmond,
Bostoen, K., Curtis, V., Fung, I.C.H. & CA, USA.
Schmidt, W.P. 2010. Water, sanitation
35. Ravel, A., Davidson, V.J., Ruzante,
and hygiene for the prevention of
J.M., Fazil, A. 2010. Foodborne
diarrhoea. International Journal of
proportion of gastrointestinal
Epidemiology, 39(Suppl.1): 193–205.
illness: estimates from a Canadian
28. Hunter, P.R. 2009. Household water Expert Elicitation survey. Foodborne
treatment in developing countries: Pathogens and Disease, 7(12): 1463–
Comparing different intervention 1472.
types using meta-regression.
36. Davidson, V.J., et al., Food-
Environmental Science and
specific attribution of selected
Technology, 43(23): 8991–8997.
gastrointestinal ILLNESSES:
29. Aspinall, W. 2010. A route to more estimates from a canadian expert
tractable expert advice. Nature, elicitation survey. Foodborne
463(7279): 294–295. Pathogens and Disease, 2011. 8(9):
p. 983–995.
30. Tyshenko, M.G., El Saadany, S.,
Oraby, T., Darshan, S., Catford, A., 37. Vally, H., Glass, K., Ford, L., Hall, G.,
Aspinall, W., Cooke, R., & Krewski, Kirk, M.D., Shadbolt, C., Veitch, M.,
D. 2012. Expert judgement and Fullerton, K.E., Musto, J. & Becker, N.
re-elicitation for prion disease 2014. Proportion of illness acquired
risk uncertainties. International by foodborne transmission for nine
Journal of Risk Assessment and enteric pathogens in Australia:
Management, 16(1-3): 48–77. an expert elicitation. Foodborne
226
62. Mtove, G., Amos, B., von Seidlein, 68. Devleesschauwer, B., Haagsma, J.A.,
L., Hendriksen, I., Mwambuli, A., Bellinger, D., Cole, D., Döpfer, D.,
Kimera, J., Mallahiyo, R., Kim, Fazil, A., Fèvre, E., Lake, R., Maertens
D.R., Ochiai, R.L., Clemens, J.D., de Noordhout, C., McDonald,
Reyburn, H., Magesa, S. & Deen, S.A., Pires, S.M., Speybroeck,
J.L. 2010. Invasive salmonellosis N., Thomas, K., Torgerson, P.R.,
among children admitted to a Wu, F., Havelaar, A.H. & Praet, N.
rural Tanzanian hospital and a 2015. Methodological framework
comparison with previous studies. for World Health Organization
PLOS ONE, 5(2): Art e9244. estimates of the global burden of
foodborne disease. PLOS ONE.
63. Koch, K., Kristensen, B., Holt,
H.M., Ethelberg, S., Molbak, K. & 69. Dean, A.S., Crump, L., Greter,
Schonheyder, H.C. 2011. International H., Hattendorf, J., Schelling,
travel and the risk of hospitalization E. & Zinsstag, J. 2012. Clinical
with non-typhoidal Salmonella manifestations of human brucellosis:
bacteremia. A Danish population- a systematic review and meta-
based cohort study, 1999–2008. analysis. PLOS Neglected Tropical
BMC Infectious Diseases, 11: Diseases, 6(12): Art e1929.
Article 277.
70. Maertens de Noordhout, C.,
64. Muller, B., Durr, S., Alonso, S., Devleesschauwer, B., Angulo, F.J.,
Hattendorf, J., Laisse, C.J.M., Verbeke, G., Haagsma, J., Kirk, M.,
Parsons, S.D.C., van Helden, Havelaar, A. & Speybroeck, N. 2014.
P.D.& Zinsstag, J. 2013. Zoonotic The global burden of listeriosis:
Mycobacterium bovis-induced a systematic review and meta-
tuberculosis in humans. Emerging analysis. Lancet Infectious Diseases,
Infectious Diseases, 19(6): 899–908. 14(11): 1073–1082.
65. WHO. 2014. Global tuberculosis 71. Murrell, K.D. & Pozio, E. 2011.
report 2013, Available from: Worldwide occurrence and impact
http://apps.who.int/iris/bitstream/ of human trichinellosis, 1986–2009.
10665/91355/1/9789241564656_eng. Emerging Infectious Diseases,
pdf. WHO, Geneva. 17(12): 2194–2202.
66. Durr, S., Muller, B., Alonso, S., 72. Torgerson, P.R., Keller, K., Magnotta,
Hattendorf, J., Laisse, C.J.M., van M. & Ragland, N. 2010. The global
Helden, P.D. & Zinsstag, J. 2013. burden of alveolar echinococcosis.
Differences in primary sites of PLOS Neglected Tropical Diseases,
infection between zoonotic and 4: e722.
human tuberculosis: results from
73. Carabin, H., Ndimubanzi, P.C.,
a worldwide systematic review.
Budke, C.M., Nguyen, H., Qian, Y.,
PLOS Neglected Tropical Diseases,
Cowan, L.D., Stoner, J.A., Rainwater,
7(8): Art e2399.
E. & Dickey, M. 2011. Clinical
67. Dean, A.S., Crump, L., Greter, H., manifestations associated with
Schelling, E. & Zinsstag, J. 2012. neurocysticercosis: a systematic
Global burden of human brucellosis: review. PLOS Neglected Tropical
a systematic review of disease Diseases, 5(5): Art e1152.
frequency. PLOS Neglected Tropical
Diseases, 6(10): Art. e1865.
229
74. Ndimubanzi, P.C., Carabin, H., 80. Mas-Coma, S., Bargues, M.D.
Budke, C.M., Nguyen, H., Qian, &Valero, M.A. 2005. Fascioliasis
Y.J., Rainwater, E., Dickey, M., and other plant-borne trematode
Reynolds, S. & Stoner, J.A. 2010. A zoonoses. International Journal of
systematic review of the frequency Parasitology, 35(11-12): 1255–1278.
of neurocysticercosis with a focus
81. Murray, C.J.L., Vos, T., Lozano, R.,
on people with epilepsy. PLOS
Naghavi, M., Flaxman, A.D., Michaud,
Neglected Tropical Diseases,
C., Ezzati, M., Shibuya, K., Salomon,
4(11): Art e870.
J.A. and 350 others. 2012. Disability-
75. Budke, C.M., Carabin, H., adjusted life years (DALYs) for 291
Ndimubanzi, P.C., Nguyen, H., diseases and injuries in 21 regions,
Rainwater, E., Dickey, M., Bhattarai, 1990–2010: a systematic analysis for
R., Zeziulin, O. & Qian, M.B. 2013. A the Global Burden of Disease Study
systematic review of the literature 2010. Lancet, 380(9859): 2197–2223.
on cystic echinococcosis frequency
82. Salomon, J.A., Vos, T., Hogan, D.R.
worldwide and its associated clinical
and 124 others. 2012. Common
manifestations. American Journal
values in assessing health outcomes
of Tropical Medicine and Hygiene,
from disease and injury: disability
88(6): 1011–1027.
weights measurement study for
76. Torgerson, P.R. & Mastroiacovo, the Global Burden of Disease Study
P. 2013. The global burden of 2010. Lancet, 380(9859): 2129–2143.
congenital toxoplasmosis: a
83. Vos, T., Flaxman, A.D., Naghavi,
systematic review. Bulletin of
M. and 352 others. 2012. Years
the World Health Organization,
lived with disability (YLDs) for
91(7): 501–508.
1160 sequelae of 289 diseases and
77. Furst, T., Keiser, J. & Utzinger, J. injuries 1990–2010: a systematic
2012. Global burden of human food- analysis for the Global Burden
borne trematodiasis: a systematic of Disease Study 2010. Lancet,
review and meta-analysis. Lancet 380(9859): 2163–2196.
Infectious Diseases, 12(3): 210–221.
84. Devleesschauwer, B., Praet, N.,
78. Furst, T., Sayasone, S., Odermatt, Speybroeck, N., Torgerson, P.R.,
P., Keiser, J. & Utzinger, J. 2012. Haagsma, J.A., De Smet, K.,
Manifestation, diagnosis, and Murrell, K.D., Pozio, E. & Dorny,
management of foodborne P. 2015. The low global burden
trematodiasis. British Medical of trichinellosis: evidence and
Journal, 344: Art e4093. implications. International Journal
for Parasitology, 45(2-3): 95–99.
79. McDonald, S.A., Devleesschauwer,
B., Speybroeck, N., Hens, N., Praet, 85. Cardoso, A.P., Mirione, E., Ernesto,
N., Torgerson, P.R., Havelaar, A.H., M., Massaza, F., Cliff, J., Haque, M.R.
Wu, F., Tremblay, M., Amene, E.W. & Bradbury, J.H. 2005. Processing of
& Dopfer, D. 2015. Data-driven cassava roots to remove cyanogens.
methods for imputing national- Journal of Food Composition and
level incidence in global burden Analysis, 18(5): 451–460.
of disease studies. Bulletin of
the World Health Organization,
93(4): 228–236.
230
124. Nakamura, T., Nakai, K., Matsumura, 129. UNEP [United Nations Environment
T., Suzuki, S., Saito, Y., Satoh, H. Programme]. 2013. Human exposure
2008. Determination of dioxins and to POPs across the globe: POPs
polychlorinated biphenyls in breast levels and human health implication:
milk, maternal blood and cord blood Results of the WHO/UNEP human
from residents of Tohoku, Japan. milk survey. In: [Proceedings of the]
Science of the Total Environment, Conference of the Parties to the
394(1): 39–51. Stockholm Convention on Persistent
Organic Pollutants, Sixth Meeting.
125. Wittsiepe, J., Furst, P., Schrey, P.,
Geneva, 28 April–10 May 2013. Doc.
Lemm, F., Kraft, M., Eberwein, G.,
No. UNEP/POPS/COP.6/INF/33.
Winneke, G., Wilhelm, M. 2007.
Available from: http://www.unep.
PCDD/F and dioxin-like PCB in
org/chemicalsandwaste/portals/9/
human blood and milk from German
POPs/docs/UNEP-POPS-COP.6-
mothers. Chemosphere, 67(9;
INF-33.English.pdf Accessed
Special Issue): S286–S294.
2015-10-21.
126. UNEP [United Nations Environment
130. Crump, K.S. 1984. A new method
Programme]. 2013. Human exposure
for determining allowable daily
to POPs across the globe: POPs
intakes. Fundamental and Applied
levels and human health implication:
Toxicology, 4(5): 854–871.
Results of the WHO/UNEP human
milk survey. In: [Proceedings of the] 131. EFSA, 2009. Scientific opinion. Use
Conference of the Parties to the of the benchmark dose approach
Stockholm Convention on Persistent in risk assessment. Guidance of the
Organic Pollutants, Sixth Meeting. Scientific Committee (Question No
Geneva, 28 April–10 May 2013. Doc. EFSA-Q-2005–232). EFSA Journal,
No. UNEP/POPS/COP.6/INF/33. 1150: 1–72.
Available from: http://www.unep.
132. Slob, W. 2002. Dose-response
org/chemicalsandwaste/portals/9/
modeling of continuous endpoints.
POPs/docs/UNEP-POPS-COP.6-
Toxicological Sciences, 66(2):
INF-33.English.pdf Accessed
298–312.
2015-10-21.
133. Zeilmaker, M.J.,Devleesschauwer,
127. Slob, W., Bakker, M.I., te Biesebeek,
B., Mengelers, M.J.B., Hoekstra, J.,
J.D. & Bokkers, B.G.H. 2014.
Brandon, E.F.A., Bokkers, B.G.H1.
Exploring the uncertainties in
2015. The disease burden of dioxins:
cancer risk assessment using
A global perspective. RIVM Report.
the Integrated Probabilistic Risk
Assessment (IPRA) approach. Risk 134. Baccarelli, A., Giacomini, S.M.,
Analysis, 34(8): 1401–1422. Corbetta, C., Landi, M.T., Bonzini, M.,
Consonni, D., Grillo, P., Patterson,
128. van der Voet, H. & Slob, W. 2007.
D.G. Jr., Pesatori, A.C. & Bertazzi,
Integration of probabilistic exposure
P.A. 2008. Neonatal thyroid
assessment and probabilistic hazard
function in Seveso 25 years after
characterization. Risk Analysis,
maternal exposure to dioxin. Plos
27(2): 351–371.
Medicine, 5(7): 1133–1142 (Article
Number: e161).
234
135. Aoki, Y., Belin, R.M., Clickner, R., Toxicology Program, Research
Jeffries, R., Phillips, L., Mahaffey, K.R. Triangle Park, North Carolina, USA.
2007. Serum TSH and total T-4 in
140. Van den Berg, M., Birnbaum,
the United States population and
L.S., Denison, M. and 16 others.
their association with participant
2006. The 2005 World Health
characteristics: National Health
Organization re-evaluation of
and Nutrition Examination Survey
human and mammalian toxic
(NHANES 1999-2002). Thyroid,
equivalency factors for dioxins and
17(12): 1211–1223.
dioxin-like compounds. Toxicological
136. Gray, L.E., Ostby, J.S. & Kelce, W.R. Sciences, 93(2): 223–241.
1997. A dose-response analysis
141. Haagsma, J.A., Polinder, S., Cassini,
of the reproductive effects of
A., Colzani, E., Havelaar, A.H. 2014.
a single gestational dose of
Review of disability weight studies:
2,3,7,8-tetrachlorodibenzo-p-dioxin
comparison of methodological
in male Long Evans Hooded rat
choices and values. Population
offspring. Toxicology and Applied
Health Metrics, 12: Article No. 20.
Pharmacology, 146(1): 11–20.
142. Salomon, J.A., Haagsma, J.A., Davis,
137. Cooper, T.G., Noonan, E., von
A., de Noordhout, C.M., Polinder,
Eckardstein, S., Auger, J., Baker,
S., Havelaar, A.H., Cassini, A.,
H.W.G., Behre, H.M., Haugen, T.B.,
Devleesschauwer, B., Kretzschmar,
Kruger, T., Wang, C., Mbizvo, M.T. &
M., Speybroeck, N., Murray, C.J.L. &
Vogelsong, K.M. 2010. World Health
Vos, T. 2015. Disability weights for
Organization reference values
the Global Burden of Disease 2013
for human semen characteristics.
study.The Lancet Global health,
Human Reproduction Update,
3(11): Art. e712-23.
16(3): 231–245.
143. Nithiuthai, S., Anantaphruti, M.T.,
138. NTP [National Toxicology
Waikagul, J. & Gajadhar, A. 2004.
Program]. 2006. Toxicology and
Waterborne zoonotic helminthiases.
Carcinogenesis Studies of 2,3,7,8–
Veterinary Parasitology, 126(1-2;
Tetrachlorodibenzo-p-dioxin
Special Issue): SI167–193.
(TCDD) (CAS No. 1746–01–6) in
Female Harlan Sprague-Dawley 144. Blakely, T., Hales, S., Kieft, C., Wilson,
Rats (Gavage Studies). National N. & Woodward, A. 2005. The
Toxicology Program technical global distribution of risk factors by
report series 526. 180 p. National poverty level. Bulletin of the World
Toxicology Program, Research Health Organization, 83(2):118–126.
Triangle Park, North Carolina, USA. 145. Cooke, R.M. & Goossens, L.H.J.
139. NTP [National Toxicology 2000. Procedures Guide for
Program]. 2006. Toxicology Structural Expert Judgement in
and Carcinogenesis Studies of Accident Consequence Modelling.
2,3,4,7,8–pentachlorodibenzofuran Radiation Protection Dosimetry,
(PeCDF) (CAS No. 57117–31–4) in 90(3): 303–309.
Female Harlan Sprague-Dawley 146. Cooke, R.M. 1991. Experts in
Rats (Gavage Studies). National Uncertainty: Opinion and Subjective
Toxicology Program technical Probability in Science. Oxford
report series Issue: 525. National University Press.
235
159. Murray, C.J. 2007. Towards good Cawthorne, A., Mathers, C., Stein,
practice for health statistics: lessons C., Devleesschauwer, B. on behalf
from the Millennium Development of the World Health Organization
Goal health indicators. Lancet, Foodborne Disease Burden
369(9564): 862–873. Epidemiology Reference Group.
2015. World Health Organization
160. Gelman, A. 2006. Prior distributions
global estimates and regional
for variance parameters in
comparisons of the burden of
hierarchical models (comment on
foodborne disease, 2010.
article by Browne and Draper).
PLOS Medicine, DOI: 10.1371/journal.
Bayesian Analysis, 1(3): 515–534.
pmed.1001923
161. Plummer, M. 2003. JAGS: A program
168. Kirk, M.D., Pires, S.M., Black,
for analysis of Bayesian graphical
R.E., Caipo, M., Crump, J.A.,
models using Gibbs sampling. In: K.
Devleesschauwer, B., Döpfer, D.,
Hornik, F. Leisch and A. Zeileis (eds).
Fazil, A., Fischer-Walker, C.L.,
Proceedings of the 3rd International
Hald, T., Hall, A.J., Keddy, K.H.,
Workshop on Distributed Statistical
Lake, R., Lanata, C.F., Torgerson,
Computing (DSC 2003). Vienna,
P.R., Havelaar, A.H. & Angulo, F.J.
20–22 March 2003. 10 p.
2015. World Health Organization
162. Plummer, M. rjags: Bayesian estimates of the global and regional
graphical models using MCMC. disease burden of 22 foodborne
R package version 3–12. 2014; bacterial, protozoal and viral
Available from: http://CRAN.R- diseases, 2010: a data synthesis
project.org/package=rjags. PLOS Medicine, DOI:10.1371/journal.
163. Rubin, D. 1976. Inference and pmed.1001921
missing data. Biometrika, 63(3): 169. Mead, P.S., Slutsker, L., Dietz, V.,
581–592. McCaig, L.F., Bresee, J.S., Shapiro,
164. Bhaskaran, K. & Smeeth, L. 2014. C., Griffin, P.M. & Tauxe, R.V. 1999.
What is the difference between Food-related illness and death in the
missing completely at random and United States. Emerging Infectious
missing at random? International Diseases, 5(5): 607–625.
Journal of Epidemiology, 170. Scallan, E., Hoekstra, R.M., Angulo,
43(4): 1336–1339. F.J., Tauxe, R.V., Widdowson, M.A.,
165. WHO. 2013. Global tuberculosis Roy, S.L., Jones, J.L., & Griffin, P.M.
report 2013. Doc. WHO/HTM/ 2011. Foodborne illness acquired in
TB/201311. the United States– major pathogens.
Emerging Infectious Diseases,
166. United Nations Population Division. 17(1): 7–15.
2013. World population prospects–
the 2012 revision. United Nations, 171. Scallan, E., Griffin, P.M., Angulo, F.J.,
New York, USA. Tauxe, R.V. & Hoekstra, R.M. 2011.
Foodborne illness acquired in the
167. Havelaar, A.H., Kirk, M.D., Torgerson, United States– unspecified agents.
P.R., Gibb, H.J., Hald, T., Lake, Emerging Infectious Diseases,
R.J., Praet, N., Angulo, F.J., 17(1): 16–22.
Bellinger, D.C., de Silva, N.R.,
Gargouri, N., Speybroeck, N.,
237
172. Adak, G.K., Long, S.M. &O’Brien, the Netherlands, 2009. International
S.J. 2002. Trends in indigenous Journal of Food Microbiology,
foodborne disease and Deaths, 156(3): 231–238.
England and Wales: 1992 to 2000.
179. Scallan, E., Hoekstra, R.M., Mahon,
Gut, 51(6): 832–841.
B.E., Jones, T.F. & Griffin, P.M. 2015.
173. Hall, G., Kirk, M.D., Becker, N., An assessment of the human
Gregory, J.E., Unicomb, L., health impact of seven leading
Millard, G., Stafford, R. & Lalor, foodborne pathogens in the United
K. 2005. Estimating foodborne States using disability adjusted
gastroenteritis, Australia. Emerging life years. Epidemiology and
Infectious Diseases, 11(8): 1257–1264. Infection,143(13): 2795–2804.
174. Vaillant, V., de Valk, H., Baron, E., 180. Ao, T.T., Feasey, N.A., Gordon, M.A.,
Ancelle, T., Colin, P., Delmas, M.C., Keddy, K.H., Angulo, F.J. & Crump,
Dufour, B., Pouillot, R., Le Strat, Y., J.A. 2015. Global Burden of Invasive
Weinbreck, P., Jougla, E., Desenclos, Nontyphoidal Salmonella Disease,
J.C., 2005. Foodborne infections in 2010. Emerging Infectious Diseases,
France. Foodborne Pathogens and 21(6): 941–949.
Disease, 2(3): 221–232.
181. Aspinall, WP., Cooke, RM., Havelaar,
175. Thomas, M.K., Murray, R., Flockhart, AH., Hoffman, S., Hald, T. 2015.
L., Pintar, K., Pollari, F., Fazil, Science-based global attribution
A., Nesbitt, A. & Marshall, B. of foodborne diseases: Findings of
2013. Estimates of the burden WHO expert elicitation.
of foodborne illness in Canada PLOS ONE. In press.
for 30 specified pathogens and
182. WHO. 2008. WHO initiative to
unspecified agents, circa 2006.
estimate the global burden of
Foodborne Pathogens and Disease,
foodborne diseases: First formal
10(7): 639–648.
meeting of the Foodborne Disease
176. Lake, R.J., Cressey, P.J., Campbell, Burden Epidemiology Reference
D.M., Oakley, E. 2010. Risk ranking Group (FERG): implementing
for foodborne microbial hazards in strategy, setting priorities and
New Zealand: Burden of Disease assigning the tasks. 2008. WHO,
estimates. Risk Analysis, 30(5): Geneva, Switzerland. Available at
743–752. http://www.who.int/foodsafety/
publications/burden_nov07/en/
177. Gkogka, E., Reij, M.W., Havelaar, A.H.,
Accessed 2015-10-22.
Zwietering, M.H., & Gorris, L.G.M.
2011. Risk-based estimate of effect 183. Porter, C.K., Gormley, R.,
of foodborne diseases on public Tribble, D.R., Cash, B.D., Riddle,
health, Greece. Emerging Infectious M.S. 2011. The incidence and
Diseases, 17(9): 1581–1590. gastrointestinal infectious risk
of functional gastrointestinal
178. Havelaar, A.H., Haagsma, J.A.,
disorders in a healthy US adult
Mangen, M.J.J., Kemmeren, J.M.,
population. American Journal of
Verhoef, L.P.B., Vijgen, S.M.C.,
Gastroenterology, 106(1): 130–138.
Wilson, M., Friesema, I.H.M.,
Kortbeek, L.M., van Duynhoven, 184. Haagsma, J.A., Siersema, P.D., De
Y.T.H.P. & van Pelt, W. 2012. Disease Wit, N.J. & Havelaar, A.H. 2010.
burden of foodborne pathogens in Disease burden of post-infectious
238
221. Hall, R.L., Jones, J.L., Hurd, S., viverrini. Parasitology International,
Smith, G., Mahon, B.E. & Herwaldt, 61(1; Special Issue): 38–45.
B.L. 2012. Population-based
228. WHO, 2015. Global burden of
active surveillance for Cyclospora
disease regions used for WHO-
infection– United States, Foodborne
CHOICE analyses. Cost effectiveness
Diseases Active Surveillance
and strategic planning. World
Network (FoodNet), 1997–2009.
Health Organization.
Clinical Infectious Diseases,
54(Suppl.5): S411–S417 229. Wild, C.P. & Hall, A.J. 2000. Primary
prevention of hepatocellular
222. Adams E, Yoder J, Gould LH,
carcinoma in developing countries.
Hlavsa MC, Gargano J. (2015)
Mutation Research– Reviews in
Giardiasis outbreaks in the United
Mutation Research, 462(2-3):
States, 1971-2011. In: Proceedings
381–393.
of the International Conference
on Emerging Infectious Diseases, 230. Haagsma, J.A., Polinder, S.,
August 24-26, 2015, Atlanta Georgia. Stein, C.E. & Havelaar, A.H. 2013.
P105. Available from: Systematic review of foodborne
www.iceid.org/images/2015/ burden of disease studies:
ICEID2015_web.pdf Quality assessment of data and
methodology. International Journal
223. Robertson, L.J., van der Giessen,
of Food Microbiology, 166(1): 34–47.
J.W.B., Batz, M.B., Kojima, M. & Cahill,
S. 2013. Have foodborne parasites 231. Polinder, S., Haagsma, J.A., Stein, C.
finally become a global concern? & Havelaar, A.H. 2012. Systematic
Trends in Parasitology, 29(3): review of general Burden of disease
101–103. studies using disability-adjusted life
years. Population Health Metrics,
224. FAO/WHO. 2014. Multicriteria-
10: Art 21.
based ranking for risk management
of food-borne parasites. . Food 232. Mathers, C.D., Vos, T., Lopez, A.D.,
and Agriculture Organization of Salomon, J., Ezzati, M. (ed.) 2001.
the United Nations/World Health National Burden of Disease Studies:
Organization. Rome, Italy. 302 p. A Practical Guide. Edition 2.0. Global
Programme on Evidence for Health
225. Craig, P.S. and C. Echinococcosis
Policy. WHO, Geneva. Available
Working Group. 2006. Epidemiology
at http://www.who.int/healthinfo/
of human alveolar echinococcosis
nationalburdenofdiseasemanual.pdf
in China. Parasitology International,
Accessed 2015-10-22.
55(Suppl.): S221–S225.
233. Etienne, C.F. 2013. Supplement
226. Al Shahrani, D., Frayha, H.H.,
on Caribbean burden of illness
Dabbagh, O., Al Shail, E. 2003.
study. Foreword. Journal Of Health,
First case of neurocysticercosis in
Population and Nutrition,31(4;
Saudi Arabia. Journal of Tropical
Suppl.1): 1–2.
Pediatrics, 49(1): 58–60.
234. Valenciano, M., Bergeri, I., Jankovic,
227. Petney, T., Sithithaworn, P., Andrews,
D., Milic, N., Parlic, M. & Coulombier,
R., Kiatsopit, N., Tesana, S., Grundy-
D. 2004. Strengthening early
Warr, C. & Ziegler, A. 2012. The
warning function of surveillance
ecology of the Bithynia– first
in the Republic of Serbia:
intermediate hosts of Opisthorchis
242
295. Yoder, J.S., Gargano, J.W., Wallace, R. and the Toxoplasmosis Study
R.M. & Beach, M.J. 2012. Giardiasis Group. 2010. Chorioretinal lesions in
surveillance– United States, 2009– mothers of children with congenital
2010. Morbidity and mortality toxoplasmosis in the National
weekly report. Surveillance Collaborative Chicago-based,
summaries, 61(5): 13–23. Congenital Toxoplasmosis Study.
Scientia Medica (Porto Alegre),
296. Havelaar, A.H., Kemmeren, J.M. &
20(1): 20-26.
Kortbeek, L.M. 2007. Disease burden
of congenital toxoplasmosis. Clinical 302. Usubalieva, J., Minbaeva, G.,
Infectious Diseases, 44(11): Ziadinov, I., Deplazes, P. & Torgerson,
1467–1474. P.R. 2013. Human Alveolar
Echinococcosis in Kyrgyzstan.
297. Kortbeek, L.M., Hofhuis, A., Nijhuis,
Emerging Infectious Diseases,
C.D.M. & Havelaar, A.H. 2009.
19(7): 1095–1097.
Congenital toxoplasmosis and
DALYs in the Netherlands. Memorias 303. Nahorski, W.L., Knap, J.P., Pawlowski,
do Instituto Oswaldo Cruz, Z.S., and 18 others. 2013. Human
104(2): 370–373. Alveolar Echinococcosis in Poland:
1990-2011. PLOS Neglected Tropical
298. Hens, N., Aerts, M., Faes, C., Shkedy,
Diseases, 7(1): Art. No. e1986.
Z., Lejeune, O., van Damme, P.
DOI: 10.1371/journal.pntd.0001986
& Beutels, P. 2010. Seventy-five
years of estimating the force 304. IHME [Institute for Health Metrics
of infection from current status and Evaluation]. DisMod 3. Available
data. Epidemiology and Infection, from: http://winthrop.ihme.
138(6): 802–812. washington.edu
299. Balasundaram, M.B., Andavar, R., 305. IARC [International Agency for
Palaniswamy, M. & Venkatapathy, Research on Cancer]. No date.
N. 2010. Outbreak of Acquired GLOBOCAN 2012: All Cancers
Ocular Toxoplasmosis involving (excluding non-melanoma skin
248 patients. Archives of cancer): Estimated cancer
Ophthalmology, 128(1): 28–32. incidence, mortality and prevalence
worldwide in 2012 [Online]. [16
300. Neves, E.S., Bicudo, L.N., Curi,
March 2012]. Available from: http://
A.L., Carregal, E., Bueno, W.F.,
globocan.iarc.fr/Pages/fact_sheets_
Ferreira, R.G., Amendoeira, M.R.,
cancer.aspx
Benchimol, E. & Fernandes, O. 2009.
Acute acquired toxoplasmosis: 306. Yeh, F.S., Yu, M.C., Mo, C.C., Luo,
clinical-laboratorial aspects and S., Tong, M.J. & Henderson, B,E.
ophthalmologic evaluation in a 1989. Hepatitis B virus, aflatoxins,
cohort of immunocompetent and hepatocellular carcinoma in
patients. Memorias do Instituto southern Guangxi, China. Cancer
Oswaldo Cruz, 104(2): 393–396 Research, 49(9): 2506–2509.
301. Noble, A.G., Latkany, P., 307. Wu-Williams, A.H., Zeise, L. &
Kusmierczyk, J., Mets, M., Rabiah, Thomas, D. 1992. Risk assessment
P., Boyer, K., Jalbrzikowski, J., for aflatoxin B1: a modeling
Wroblewski, K., Karrison, T., Swisher, approach. Risk Analysis, 12(4):
C.N., Mieler, W.F., Meier, P., McLeod, 559–567.
248
308. WHO. No date [online]. Global 310. WHO. 2013. WHO methods and
Health Estimates by Cause, Age, Sex data sources for global burden
and Country, 2000–2012. Available of disease estimates 2000–2011.
from: http://www.who.int/entity/ Global Health Estimates Technical
healthinfo/global_burden_disease/ Paper, 2013-4. Doc. ID: WHO/HIS/
HSI/GHE/2013.4. Available at http://
309. Tshala-Katumbay, D., Banea-
www.who.int/healthinfo/statistics/
Mayambu, J.-P., Kazadi-Kayembe,
GlobalDALYmethods_2000_2011.pdf
T., Nunga-Matadi, R., Bikangi-
Accessed 2015-10-23.
Nkiabungu, F., Edebol Eeg-Olofsson,
K. & Tylleskär, T. 2001. African 311. Yu, S.C., Tan, F., Zhou, M.G., Liu,
Journal of Neurological Science, S.W., Zhu, X.J. & Zhu, Y.L. 2014.
20(1): 8–12. Global Burden of Disease, Injury and
Risk Factor Study 2010: Its Policy
Implications for China. Biomedical
and Environmental Sciences,
27(1): 45–48.
249
LIST OF TABLES
Table 1. FERG hazards, causally related health states and corresponding disability weights (DWs).
Details on the derivation of the DWs are provided in Appendix 4..........................................................................................................43
Table 2. Foodborne hazards, and structure of the expert panels............................................................................................................46
Table 3. Examples of calibration seed questions .............................................................................................................................................48
Table 4. Exposure routes included in the expert elicitation, per hazard ............................................................................................... 51
Table 5. Modelling strategies for the hazards included in the WHO global burden of foodborne disease estimates.................53
Table 6. The number of experts enrolled, interviewed and finally included in the elicitation across panels ...................... 63
Table 7. Median global number of foodborne illnesses, deaths, Years Lived with Disability (YLDs),
Years of Life Lost (YLLs) and Disability Adjusted Life Years (DALYs), with 95% uncertainty intervals, 2010. ................... 73
Table 8. Median rates of foodborne Disability Adjusted Life Years (DALYs) per 100 000 population,
by subregion, with 95% uncertainty intervals, 2010........................................................................................................................................ 78
Table 9. Comparisons of the total burden of parasitic diseases (foodborne and non-foodborne) with 95%
uncertainty intervals, estimated by FERG and by GBD2010 [9] ........................................................................................................... 108
Table A7.1 Subregional estimates (median and 95% uncertainty interval) of the proportion of illnesses caused
by Campylobacter spp., non-typhoidal Salmonella spp., Shiga-toxin producing Escherichia coli (STEC),
Brucella spp. and Shigella spp. through each exposure pathway. ......................................................................................................... 199
Table A7.2 Subregional estimates (median and 95% uncertainty interval) of the proportion of Diarrhoeal
Disease illnesses caused by four hazards: enteropathogenic E. coli (EPEC), enterotoxigenic E. coli (ETEC),
Cryptosporidium spp. and Giardia spp. through each exposure pathway. ...................................................................................... 202
Table A7.3 Subregional estimates (median and 95% uncertainty interval) of the proportion of Diarrhoeal
Disease illnesses caused by Salmonella Typhi, Vibrio cholerae, Entamoeba histolytica, norovirus, and
hepatitis A virus through each exposure pathway. ......................................................................................................................................204
Table A7.4 Subregional estimates (median and 95% uncertainty interval) of the proportion of illnesses
caused by Toxoplasma gondii, Echinococcus multilocularis, Echinococcus granulosus and Ascaris spp.
through each exposure pathway. .......................................................................................................................................................................... 206
Table A7.5 Subregional estimates (median and 95% uncertainty interval) of the proportion of illnesses caused by
exposure to lead through each pathway. .......................................................................................................................................................... 209
Table A7.6 Percent of illness acquired through the foodborne transmission route for six national studies
and this studya. ............................................................................................................................................................................................................... 210
Table A8.1 Median number of foodborne Illnesses, Deaths, and Disability Adjusted Life Years (DALYs),
with 95% uncertainty intervals, 2010. ..................................................................................................................................................................... 211
Table A8.2 Median rates of foodborne illnesses, deaths and Disability Adjusted Life Years (DALYs)
per 100 000 persons, by region, with 95% uncertainty intervals, 2010. ..............................................................................................213
Table A8.3 Median number of Illnesses, Deaths, and Disability Adjusted Life Years (DALYs)
by age group, with 95% uncertainty intervals, 2010. .....................................................................................................................................215
Table A8.4 Median rate per 100 000 of foodborne Illnesses, Deaths and Disability Adjusted
Life Years (DALYs) by region, with 95% uncertainty intervals, 2010. .....................................................................................................217
TableA8.5 Median number of total and foodborne Illnesses, Deaths, and Disability Adjusted
Life Years (DALYs), with 95% uncertainty intervals, 2010 ...........................................................................................................................219
Table A8.6 Median number of foodborne Illnesses, Deaths, and Disability Adjusted
Life Years (DALYs), with 95% Uncertainty Intervals, 2010 ..........................................................................................................................221
Table A8.7 Median rate per 100 000 foodborne (FB) Illnesses, Deaths, and Disability Adjusted
Life Years (DALYs) by region, with 95% uncertainty intervals, 2010. .....................................................................................................221
250
LIST OF FIGURES
Figure 1. Structure of the initiative to estimate the global burden of foodborne diseases ...........................................................6
Figure 2. Hazards for which burden of foodborne disease estimates were prepared by FERG, grouped
according to TF. Hazards in grey boxes were addressed by individual TFs but were not included in the global
overview. Hazards in blue boxes are pending. ................................................................................................................................................... 31
Figure 3. Geographical distribution of experts according to working experience (>3 years) per subregion.
Several experts had experience in more than one subregion. ..................................................................................................................64
Figure 4. Statistical accuracy versus informativeness of the experts included, when using equal weight (blue)
or performance weight (red) combinations, respectively. ......................................................................................................................... 65
Figure 5. Subregional estimates of the proportion of foodborne illnesses caused by Campylobacter spp.,
non-typhoidal Salmonella spp., Shiga-toxin producing Escherichia coli (STEC), Brucella spp. and Shigella spp.
Indicated on the line plot are the 2.5th, 5th, 50th, 95th and 97.5th percentiles. .............................................................................. 67
Figure 6. Subregional estimates of the proportion of foodborne illnesses caused by enteropathogenic E. (EPEC),
enterotoxigenic E. coli (ETEC), Cryptosporidium spp. and Giardia spp. .............................................................................................68
Figure 7. Subregional estimates of the proportion of foodborne illnesses caused by typhoidal Salmonella, Vibrio
cholerae, Entamoeba histolytica, norovirus, and hepatitis A virus. ........................................................................................................69
Figure 8. Subregional estimates of the proportion of foodborne illnesses caused by Toxoplasma gondii,
Echinococcus multilocularis, Echinococcus granulosus and Ascaris spp. ..........................................................................................70
Figure 9. Subregional estimates of the proportion of disease caused by foodborne exposure to lead. ............................. 71
Figure 10. Subregional estimates (medians) of the proportion of disease caused by exposure to lead through
eight different exposure routes. ............................................................................................................................................................................... 71
Figure 11. Ranking of foodborne hazards, based on Disability-Adjusted Life Years at the global level,
with 95% uncertainty intervals, 2010. ..................................................................................................................................................................... 76
Figure 12. The global burden of foodborne disease (DALYS per 100 000 population) by hazard groups
and by subregion, 2010. ................................................................................................................................................................................................80
Figure 13. Relative contribution of Years of Life Lost due to premature mortality (YLL) and Years Lived with
Disability (YLD) to the global burden of 31 hazards in food, 2010. .......................................................................................................... 81
Figure 14. Age-distribution of disability adjusted life years for 31 hazards contributing to the global burden of
foodborne disease, 2010. ............................................................................................................................................................................................. 82
Figure 15. Scatterplot of the global burden of foodborne disease per 100 000 population and per incident case....................83
Figure 16. The global burden of foodborne disease by subregion (DALYS per 100 000 population) caused
by enteric hazards, 2010. .............................................................................................................................................................................................. 85
Figure 17. Disability Adjusted Life Years for each pathogen acquired from contaminated food ranked from
lowest to highest with 95% Uncertainty Intervals, 2010. .............................................................................................................................86
Figure 18A. The relative contribution to the DALY incidence by each agent for each of the subregions.
This includes enteric protozoa to complete the picture on foodborne parasitic diseases. However the detail is
reported in the accompanying manuscript on foodborne enteric pathogens [168]. ..................................................................... 87
Figure 18B. Disability Adjusted Life Years for each parasite acquired from contaminated food ranked from
lowest to highest with 95% Uncertainty Intervals, 2010. This includes enteric protozoa to complete the picture
on foodborne parasitic diseases. However the detail is reported in the accompanying manuscript on foodborne
enteric pathogens [168]. ............................................................................................................................................................................................... 88
Figure 19. The relative proportion of the burden of each of the foodborne parasitic diseases contributed
by YLLs and YLDs .......................................................................................................................................................................................................... 88
Figure 20. The relative contribution to the DALY incidence by each of four chemicals for each of
the WHO Regions. ..........................................................................................................................................................................................................90
Figure 21. The relative contributions from YLLs and YLDs for each of four chemicals................................................................90
Figure 22. Disability Adjusted Life Years for each of four chemicals from contaminated food ranked,
from lowest to highest, with 95% uncertainty intervals ................................................................................................................................. 91
Figure 23. Major transmission routes of human foodborne diseases indicate two points of attribution:
the reservoir level and the exposure level. ........................................................................................................................................................ 101
Figure A6.1 FERG hazards, causally related health states and corresponding disability weights (DWs). The fourth
column describes how the various DWs were derived from the Global Burden of Disease Studies (GBD) and the
World Health Organization Global Health Estimates (WHO/GHE). .......................................................................................................195
251
GLOSSARY
ABBREVIATIONS
ISBN 9789241565165