UNDERSTANDING LIVESTOCK OWNERSHIP, DIETARY DIVERSITY AND ASF CONSUMPTION, AND THE ENVIRONMENTAL AND SPATIAL DETERMINANTS OF NUTRITION STATUS IN CU5 IN RURAL SOUTHERN HAITI By LINDSEY A. LAYTNER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2018
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
UNDERSTANDING LIVESTOCK OWNERSHIP, DIETARY DIVERSITY AND ASF CONSUMPTION, AND THE ENVIRONMENTAL AND SPATIAL DETERMINANTS OF
NUTRITION STATUS IN CU5 IN RURAL SOUTHERN HAITI
By
LINDSEY A. LAYTNER
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
Environment and Climate ................................................................................. 28 Poverty ............................................................................................................. 29
Livelihood and Food Production ....................................................................... 29 Water, Hygiene, and Sanitation ........................................................................ 31 Undernutrition ................................................................................................... 32
2 I -LIVESTOCK OWNERSHIP AND DIETARY DIVERSITY OF CU5 IN RURAL HAITIAN HOUSEHOLDS ........................................................................................ 48
3 II - LIVESTOCK OWNERSHIP, WASH, AND CU5 NUTRITION STATUS IN RURAL HAITIAN HOUSEHOLDS .......................................................................... 71
Introduction ............................................................................................................. 71 Research Objective ................................................................................................ 73 Methods .................................................................................................................. 73
4 III - SPATIAL DETERMINANTS OF CU5 LINEAR GROWTH IN RURAL HAITIAN HOUSEHOLDS ........................................................................................ 89
Introduction ............................................................................................................. 89 Research Objective ................................................................................................ 91
definition, spatial resolution, and reference source) ......................................... 104
9
LIST OF FIGURES
Figure page 1-1 A) Map of Haiti with the Aquin and Côtes-de-Fer study communes. B) Map of
the Aquin (Flamands, Fonds des Blancs, Guirand, and Frangipane) and Côtes-de-Fer (Jamais Vu) study area sub-section. ............................................ 41
1-2 Percentage of CU5 that are stunted in Haiti compared to the DR according to the DHS66. .......................................................................................................... 42
1-3 One-health theoretical framework to understand linkages between livestock ownership and child under five nutrition in southern Haiti. Adapted from UNICEF 86. ......................................................................................................... 43
1-4 Malnutrition terminology. Definitions of the various forms of malnutrition and undernutrition from the Joint Child Malnutrition Estimates 2017 edition2. ........... 44
1-5 Five 5’s diagram (adapted from Penakapapti et al.52. ......................................... 44
1-6 Undernutrition pathway from pathogen exposure via livestock feces adapted from Penakapapti et al.52. ................................................................................... 45
1-7 Improved nutrition via livestock ownership and safe WASH practices (blocked exposure to livestock feces and improved access to safe and nutritious foods) adapted from Penakapapti et al. 52. .......................................... 46
1-8 Variable list and description for all chapters. For maternal knowledge scoring, see Appendix, Figure A-1. ..................................................................... 47
2-1 Table showing the breakdown and frequencies of each response per food item in HDDS and ASF consumption calculation. ............................................... 64
2-3 Descriptive statistics of survey respondents, overall. ......................................... 66
2-4 Descriptive statistics of livestock ownership, HDDS, and ASF consumption by sub-communal section. .................................................................................. 67
2-5 Bivariate regression results for study variables and HDDS. ............................... 67
2-6 Bivariate regression results for study variables ASF consumption. .................... 68
2-7 Multivariate binary backward-stepwise logistic regression results assessing the association of model 1: livestock ownership and HDDS status. ................... 69
2-8 Multivariate binary backward-stepwise logistic regression results assessing the association of model 2: livestock ownership and ASF consumption status. . 70
10
3-1 Definitions of undernutrition adapted from WHO, UNICEF, and the World Bank. .................................................................................................................. 83
3-2 Variable Descriptions used in chapter 3 analyses .............................................. 84
3-3 Descriptive statistics of surveyed households. ................................................... 85
3-4 Descriptive Statistics of WASH characteristics (Improved “I” and Unimproved “U”) broken down by sub communal section ...................................................... 85
3-5 Bivariate regression results for study variables and CU5 Stunting ..................... 86
3-6 Multivariate binary backward-stepwise logistic regression results for model 1 assessing the association of livestock ownership and CU5 Stunting status. ...... 87
3-7 Multivariate binary backward-stepwise logistic regression results for model 2 association the of livestock ownership, WASH Factors, and CU5 Stunting status. ................................................................................................................. 88
4-1 Conceptual diagram linking CU5 growth to Haiti-specific spatial and environmental drivers. Adapted from Grace et al.163. ...................................... 102
4-2 Map of Haiti and communes Aquin and Côtes-de-Fer surveyed (in red). ......... 103
4-3 Description/ distribution of spatial and environmental covariates considered in this analysis, across the country, as well as the Aquin and Cote de Fer study site communes. ....................................................................................... 105
4-4 Village coordinates geo-referenced using Google Earth Pro. ........................... 106
4-5 Village level CU5 HAZ score distribution across Aquin and Cote de Fer study site communes. ................................................................................................ 106
4-6 Livestock species distribution across Aquin and Cote de Fer study site communes. ....................................................................................................... 107
4-7 Results from the bivariate analysis of environmental and spatial covariates and village level CU5 HAZ. ............................................................................... 108
4-8 Final multivariate linear regression model results and overall model characteristics. .................................................................................................. 108
4-9 Map of the cluster and outlier analysis (Local Moran’s) in the surveyed villages. ............................................................................................................ 109
4-10 Model Residual vs. Predicted Plot indicating a properly specified model. ........ 109
A-1 Vaffriables included in Maternal Knowledge Score calculation calculations. Note, Iron and Vitamin A are included together in the combined score. ........... 114
11
B-1 Model Fit statistics for Chapter 2 Model 1: HDDS. ........................................... 115
B-2 Summary of backwards elimination procedure in multivariate backward stepwise logistic regression for Model 1: HDDS. .............................................. 115
B-3 Predictive power statistics of Model 1: HDDS................................................... 115
B-4 Receiver Operating Characteristic Curve (ROC) showing predictive power of final model, for Model 1: HDDS. ....................................................................... 116
B-5 Receiver Operating Characteristic Curves (ROC) showing predictive power of each model step until final model for Model 1: HDDS. ................................. 116
B-6 Model Fit statistics for Chapter 2 Model 2: ASF................................................ 117
B-7 Summary of backwards elimination procedure in multivariate backward stepwise logistic regression for Model 2: ASF. ................................................. 117
B-8 Predictive power statistics of Model 2: ASF. ..................................................... 117
B-9 Receiver Operating Characteristic Curve (ROC) showing predictive power of final model, for Model 2: ASF. .......................................................................... 118
B-10 Receiver Operating Characteristic Curves (ROC) showing predictive power of each model step until final model for Model 2: ASF. ..................................... 118
C-1 Model Fit statistics for Chapter 3 Model 1: Livestock and Stunting. .................. 119
C-2 Summary of backwards elimination procedure in multivariate backward stepwise logistic regression for Chapter 3 Model 1: Livestock and Stunting. ... 119
C-3 Predictive power statistics of Chapter 3, Model 1: Livestock and Stunting. ...... 119
C-4 Receiver Operating Characteristic Curve (ROC) showing predictive power of final model, for Chapter 3, Model 1: Livestock and Stunting. ............................ 120
C-5 Receiver Operating Characteristic Curves (ROC) showing predictive power of each model step until final model for chapter 3, Model 1: Livestock and Stunting. ........................................................................................................... 120
C-6 Model Fit statistics for Chapter 3 Model 2: Livestock, WASH, and Stunting. .... 121
C-7 Summary of backwards elimination procedure in multivariate backward stepwise logistic regression for Chapter 3 Model 2: Livestock, WASH, and Stunting. ........................................................................................................... 121
C-8 Predictive power statistics of Chapter 3, Model 2: Livestock, WASH, and Stunting. ........................................................................................................... 121
12
C-9 Receiver Operating Characteristic Curve (ROC) showing predictive power of final model, for Chapter 3, Model 2: Livestock, WASH, and Stunting. .............. 122
C-10 Receiver Operating Characteristic Curves (ROC) showing predictive power of each model step until final model for chapter 3, Model 2: Livestock, WASH, and Stunting. ........................................................................................ 122
13
LIST OF ABBREVIATIONS
ASF Animal Source Food
CI Confidence Interval
CU5 Child under five years old
DDS Dietary Diversity Score
DHS Demographic and Health Surveys
EBK Empirical Bayesian Kriging
EED Environmental Enteric Dysfunction
GDP Gross Domestic Product
GIS Geographic Information System
HAZ Height for age z score
HDDS Household Dietary Diversity Score
HDI Human Development Index
IRB Institutional Review Board
JMP WHO Joint Monitoring Program
KAP Knowledge, attitudes and practices
LMIC Low- and middle-income countries
MAL-ED The Etiology, Risk Factors and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development (MAL-ED) study
MAR Missing at Random
MCMC Multi Chain Monte Carlo
MI Multiple Imputation
MODIS Moderate Resolution Imaging Spectroradiometer
14
MUAC Middle and upper arm circumference
NDVI Normalized Difference Vegetation Index
NGO Non-Governmental Organization
OLS Ordinary Least Squares
OR Odds Ratio
PCA Principle components analysis
SD Standard Deviation
SES Socio-economic status
SRTM-DEM Shuttle Radar Topography Mission Digital Elevation Model
UNICEF United Nations Children’s Fund
VIF Variance Inflation Factor
WASH Water, Hygiene and Sanitation
WAZ Weight for age z score
WHO World Health Organization
WHZ Weight for height z score
15
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
UNDERSTANDING LIVESTOCK OWNERSHIP, DIETARY DIVERSITY AND ASF CONSUMPTION, AND THE ENVIRONMENTAL AND SPATIAL DETERMINANTS OF
NUTRITION STATUS IN CU5 IN RURAL SOUTHERN HAITI
By
Lindsey A. Laytner December 2018
Chair: Sarah L. McKune Major: Public Health
Livestock are ubiquitous in many parts of the developing world, with both humans
and domestic animals sharing close environments. Livestock have the potential to
provide nutrient-dense animal source foods (ASF) such as meat, dairy, and eggs,
providing vital micro and macronutrients to children to support their development and
growth. This is especially critical within their first 1000 days of life. However, this
potential benefit may be offset by the possibility that livestock may have the potential to
hinder growth benefits in children via child exposure to disease-causing pathogens in
their excreta. Thus, understanding the context of water, hygiene and sanitation, as well
as livestock ownership is crucial to designing positively impactful nutrition and hygiene
interventions.
Moreover, spatial and environmental factors on the landscape can influence child
growth indirectly. Understanding which environmental and spatial drivers are the most
influential on child growth is crucial to designing targeted interventions. Ultimately,
these potential associations between livestock ownership, dietary diversity and ASF
consumption, WASH, and the spatial and environmental covariates remain important
aspects to consider, yet are understudied in relation to undernutrition in Haiti. This
16
research will add to the growing body of literature to assess these associations in two
rural communes in southern Haiti.
17
CHAPTER 1 BACKGROUND
Undernutrition is a worldwide concern—one or more forms of malnutrition affect
populations within nearly every country. According to the World Health Organization
(WHO), malnutrition/undernutrition refers to “deficiencies, excesses, or imbalances in a
person’s intake of energy and/or nutrients” 1. Undernutrition is more common in low
and middle-income countries (LMICs) and disproportionately impacts children under five
(CU5). In LMIC, close to half of child mortality globally is linked to undernutrition. In
2016, the WHO estimated that 155 million CU5 in developing countries were stunted (a
sign of chronic undernutrition). Of these, 66% lived in LMIC1,2.
Combating undernutrition in all its forms is one of the greatest global health
challenges3. However, optimizing nutrition early—including the 1,000 days from
conception to a child’s second birthday—ensures the best possible start in life and
many associated long-term benefits4,5. For children living in LMICs, undernutrition is
associated with the chronic exposure to infectious disease-causing enteric and
respiratory pathogens. These pathogens, present in the environment through multiple
exposure pathways, may alter gut integrity and function, impairing absorption of
nutrients and resulting in Environmental Enteric Dysfunction (EED) 6–8. EED can
further undernutrition and likewise an increased susceptibility to and incidence of both
asymptomatic infection and symptomatic disease6–8. There is growing evidence from
the Etiology, Risk Factors and Interactions of Enteric Infections and Malnutrition and the
Consequences for Child Health and Development (MAL-ED) study that reducing
enteropathogen burden can improve child growth outcomes, especially if energy intake
is improved9. Other evidence suggest that these pathogens may also inhibit immune
18
responses to childhood vaccines, diminishing their effectiveness and impacting broader
child health outcomes10,11. Moreover, repeated infections by these pathogens can also
lead to cognitive and additional related developmental deficits. Therefore, the
cumulative effects of continual infection and asymptomatic colonization, undernutrition,
and impaired child growth and development have great social and economic
consequences for a child’s entire life. Unfortunately, these pathogens and diseases
place a disproportional burden on poor families and the communities where they
reside12–14.
A child is considered to be stunted if their height-for-age Z score (HAZ) is -2
(stunted) to -3 or more (severely stunted) standard deviations below the HAZ of the
WHO reference median of children worldwide 15,16. Dietary diversity has been
associated with better nutritional status of children in developing countries15,17–21, and
has an especially strong relationship to childhood stunting15. In the field of nutrition,
“dietary diversity” is a measure associated with (1) overall quality and (2) nutrient
adequacy in an individual’s dietary practices and is usually assessed through dietary
diversity scores (DDS). These measures compare the number of food groups an
individual or household consumes over a previously determined reference period15,18.
Several studies have shown that DDS is positively associated with overall dietary
quality, particularly improved micronutrient consumption in children15,18,22.
Consumption of livestock and livestock products, such as dairy, meat and fish, as
well as egg proteins provide bioavailable vitamins, such as vitamins B12, riboflavin, iron,
calcium, zinc that are essential to child nutrition23. Dietary diversity involves adequate
intake of macronutrients and micronutrients. The inclusion of ASF in the diet helps
19
prevent multiple nutrient deficiencies and any resultant, linear growth retardation24.
Children living in dietary-diverse households with quality diets are more likely to
consume animal sourced foods (ASFs)25. Previous studies looking at large datasets
have shown that livestock ownership can increase consumption of ASF in the
household through increased access, availability, and income generation14,26–31.
Health and dietary practices, including supplementation (e.g. vitamins A, iron,
etc.), are influenced by a wide array of complex interactions, including individual
knowledge, attitudes, and practices (KAP), social-cultural beliefs and psychological
factors (i.e. motivations), environmental contexts, resources, and other factors32,33.
There is growing recognition among scholars regarding the important role of structural,
environmental, cultural, social, and psychological factors that can influence a person’s
diet and dietary behaviors33,34. Decisions regarding diet and food choices are often
shaped by socio-cultural factors and cultural context beyond the individual’s personal
experience. However, careful integration of dietary KAP into education programs can
support and improve dietary practices in LMIC. Evidence from in-depth qualitative
ethnographic research in Tanzania shows that careful integration of dietary diversity into
local knowledge, attitudes beliefs and practices helped local people believe that dietary
diversity was important and felt that it could be achieved in their villages because the
nutrition messaging could easily be integrated into existing nutrition programs, local
concepts, and knowledge frameworks33.
There is little research on the complexities surrounding livestock ownership,
livestock husbandry, WASH (especially with regards to livestock husbandry), ASF
consumption, and child nutrition. While owning livestock can provide food and income-
20
livelihood security for nearly one billion poor people in developing countries35, there may
be an increased zoonotic infection risk for children in livestock-owning households
because of children’s proximity and continuous exposure to livestock and their excreta.
Studies have shown children can be exposed to (and can directly or indirectly) ingest
livestock fecal matter in Peru, Zimbabwe, and Bangladesh36–38. In a recent secondary
analysis of child stunting in Ethiopia, Bangladesh, and Vietnam, researchers found that
livestock, in particular poultry in the home overnight is associated with feces exposure.
Moreover, the presence of livestock feces is significantly and negatively associated with
child HAZ, in Ethiopia (β = −0.22), and Bangladesh (β = −0.13). This study also
suggests that livestock feces may be positively associated with diarrheal disease
symptoms in Bangladesh as well39. This potential for an increased risk of infection in
children in livestock-owning households warrants careful attention to WASH in and
around the household, especially with regards to livestock ownership and husbandry.
The research presented in this dissertation is a contribution to the small but
growing body of literature devoted to understanding the benefits and risks of livestock
ownership on CU5 health. This work serves as a baseline for understanding the
relationship between livestock ownership, dietary diversity (specifically ASF
consumption), and child stunting in the southern region of Haiti. Haiti, and the regions
presented in this dissertation are understudied, especially in regard to livestock
ownership, diet, WASH, and undernutrition.
The main research areas and hypotheses explored are as follows:
Chapter 2 focuses on whether there is a relationship between livestock ownership and dietary diversity or ASF consumption in rural Haitian households, as these are factors that may influence CU5 nutrition status and ultimately, childhood stunting. The two hypotheses are: (1) Livestock ownership is
21
associated with increased dietary diversity, and (2) Livestock ownership is associated with increased ASF consumption.
Chapter 3 focuses on whether livestock ownership has a relationship to CU5 nutrition status in rural Haitian households, and whether WASH may influence it. Hypotheses: (1) Livestock ownership is associated with decreased CU5 stunting. (2) When unimproved WASH factors are included, livestock ownership is associated with increased stunting.
Chapter 4 explores the environmental and spatial variables that may be contributing to CU5 nutrition status in rural Haitian villages. The hypotheses for this chapter are exploratory. The hypotheses for this chapter are that environmental factors are associated with CU5 growth patterns.
Livestock Ownership and Child Nutrition
Few studies have examined the direct effect of livestock ownership on child
nutrition14,28,29,31. Only one of these studies has assessed the association between
livestock ownership, DDS, ASF consumption, height-for-age z-score, and childhood
stunting. This cross-sectional study of children from Luangwa Valley, Zambia used
multilevel mixed-effects linear and logistic regression models to examine the association
between livestock types and four nutrition-related outcomes of interest40. They did not
find any statistically significant relationships between any of their livestock ownership
measures and a child’s odds of ASF consumption, height-for-age z-score, or stunting.
However, their linear models showed that while having fewer poultry was associated
with decreased child dietary diversity (β = -0.477; p<0.01) relative to owning no
livestock, as the number of chickens owned increased, a positive, significant association
with DDS (β = 0.022; p<0.01) was observed. However, livestock production can also
increase ASF intake indirectly, as seen in Kenya and Ethiopia—households that
produce livestock can have increased purchasing power for higher quality food
items28,41.
22
Livestock ownership can affect dietary intake and thus affect human growth
outcomes14. Some studies have suggested that livestock, especially chickens, can
contribute to child stunting patterns both positively and negatively39,42 . When children
have increased access to safely prepared eggs and poultry meat, they were shown to
have better nutritional outcomes43, which in turn can lead to better linear growth
outcomes. However, if increased access and availability of chickens is coupled with
poor husbandry and WASH practices, there may be an increased exposure to
pathogens, such as Campylobacter—a known cause of diarrheal diseases, and impact
growth faltering through the EED pathway 44–46. EED is a condition of chronic gut
inflammation from microbial (e.g. fecal bacteria) colonization in the gut, that have shown
to impact child nutrient absorption, growth patterns, among other adverse
developmental outcomes7,47,48.
Improved water, hygiene and sanitation (WASH) have been linked to
improvements in child health outcomes49–52, especially with regards to handwashing
and safe feces disposal47,53. However, there is limited empirical evidence about the
benefits of improved livestock WASH interventions to child nutrition status. This may be
a result of sanitation efforts focusing on human, rather than livestock excrement
containment52,54. Studies, such as the WASH Benefits Study in Bangladesh and Kenya
aimed to provide rigorous evidence on both health and developmental benefits of
WASH and nutritional interventions during a child’s first 1000 days of life55,56. However,
these studies did not find a relationship between WASH improvements and linear
growth outcomes. Despite the sanitation improvements made with these studies, the
results highlight the potential for targeting environmental exposure to feces57.
23
Given these recent studies, there are few empirical research studies that have
investigatived livestock WASH interventions that could potentially improve child nutrition
status. There may be a relationship between livestock husbandry and WASH practices,
since optimal WASH practices (e.g. hand washing, corralling livestock away from the
home) serve as a potential barrier between animals and young children39,54. Hazardous
livestock practices in low-income countries, such as corralling poultry close to children
at night38, and not separating poultry and other livestock from areas where children may
sit, crawl, play, and eat36,37 may be associated with pathogen exposure, colonization,
repeated infections, and eventually an increased risk for EED42,58.
Dietary Diversity and Child Nutrition
Several studies have found associations between DDS and child consumption
patterns or nutrition status within and across several countries in Africa and
Asia15,17,19,24,25,59–62. Each one is reviewed below.
Arimond et al. assessed dietary diversity in 11 countries across Africa and Asia.
Using Demographic and Health Surveys (DHS), these authors examined the
association between dietary diversity and HAZ for children 6 to 23 months old, while
controlling for confounding factors 17. Their bivariate and multivariate results found
significant positive associations between dietary diversity and CU5 HAZ. In the
multivariate models, 7 of the 11 countries had signficiant associations between DDS,
independent of socioeconomic factors17.
There were two studies in Kenya exploring ASF consumption and child growth.
Neuman et al., assessed the effects of ASF consumption and dietary diversity on child
growth63. This randomized, controlled feeding intervention study had three interventions
of meat, milk, or vegetable stew, and a control group who received no snack. The
24
outcome data were food intake (within 24 hours) recall surveys and anthropometric
measurements (e.g., height, weight, arm circumference, fat to muscle). The meat-
consuming group showed the greatest gain in arm muscle growth, followed by the milk-
consuming group as compared with the vegetable stew consumers (p< 0·05). The meat
group showed the least increase in fat area of all groups63. The longitudinal study by
Iannotti and Lesorogol explored the relationship between milk consumption and child
growth patterns in pastoralist communities in Samburu, Kenya. They found that milk
availability at the household level affected CU5 milk intake and anthropometry.
Specifically, that milk consumption was significantly associated with higher body mass
index z scores among youth21.
In Ethiopia, predictors of household dietary diversity and ASF consumption
patterns were assessed in the 2011 Ethiopian Welfare Monitoring Survey (WMS)59.
Dietary data were collected from 27,995 households using a questionnaire measuring
dietary diversity over the past 1 week. Household DDS (HDDS) was constructed
according to the Food and Agricultural Organization guidelines. The medianHDDS of
the surveyed households was 5 food groups, with cereals being the most commonly
(96%) consumed food group. Fish, egg, and fruits, on the other hand, were the least
consumed food groups. The ASFs were consumed in greater proportions in households
with higher HDDS. Additional factors that were identified as predictors and were
positively associated with higher HDDS included: being in the higher and middle socio-
well as moderate amounts of healthy fats87. For proper diet for infants and young
children under 2 years old, the WHO and UNICEF recommends introduction of
complimentary feeding (minimum of 4 food groups) for children 6 to 23 months of age,
in addition to continued breastfeeding. An additional recommendations for this age
group is to include are iron-fortified or iron-rich foods designed for infants and young
children in their diet88.
Disease. Diarrheal diseases are a leading cause of undernutrition in children
under 2 years old and are caused by exposure to waterborne and foodborne
pathogens54. Despite established UNICEF framework of malnutrition, there is growing
evidence that reflect enteric pathogen infection in the absence of diarrheal disease is
even more common9. Prolonged exposure to these diseases and any subsequent
asymptomatic colonization can impact nutrition outcomes in CU5 by impacting the gut
35
flora, intestinal permeability, and ability to absorb nutrients from food89; ultimately,
limiting the benefits of a nutritious diet. The MAL-ED study has identified across all
studied sites that the highest episodes of infection occurring in humans come from
human bacteria such as Shigella spp, norovirus GII, rotavirus, and Campylobacter in the
children under two years of age. Much of the prominent human diarrhea-causing
pathogens identified are of zoonotic origin such as Campylobacter and Salmonella,
where the primary reservoir is poultry.
Humans are usually infected by diarrheal pathogens through a fecal-oral
pathway. Some of the critical pathways that diarrheal pathogens are transmitted can be
summarized by the 5-Fs (i.e. food, flies, fomites, fingers, fluids, and fields)90. Figures 1-
5 illustrates the 5-Fs and incorporates primary exposure (i.e. direct feces contamination)
and secondary exposure (i.e. indirect contamination of food) to diarrheal pathogens by a
child. However, with livestock generating at least 85% of the world’s animal fecal
waste52, this environmental fecal contamination can increase the potential transmission
of zoonotic and foodborne diseases14,30,52,54,90. According to the 2015 Global Burden of
Disease study by Wang et al., at least one third of CU5 mortality was attributable to
microbes that can be found in animal feces96. Moreover, livestock and domestic animal
waste can contaminate soil, public and private water, and as a consequence, can lead
to human diarrhea91,92. Increased production of livestock, which is essential for
increased access to ASF, can also create new opportunities for infectious agents to
contaminate the environments via improper livestock waste management93–95.
As mentioned previously, young children can be exposed to pathogens from
poorly managed animal feces, particularly in these communities (see Figure 1-6 and
36
Figure 1-7) which can impact CU5 growth. A recent systematic review found growing
evidence to support the importance of separating animal feces from human
environments, and limiting direct and indirect child contact with fecal-borne pathogens
54. When developing integrated WASH and/or child nutrition programs, the safe
containment and disposal of livestock excreta is often overlooked, but is likely a major
pathway of child enteric disease and growth faltering39,52.
Maternal care and knowledge practices are important in the prevention,
treatment, and management of child health. Increasing maternal knowledge of which
foods are vitamin-rich can be crucial for appropriate child care and feeding practices,
especially for young children71,97, as a child’s diet is contingent on the common feeding
practices of their mother and household members. Moreover, maternal knowledge of
disease risk factors, particularly the causal factors and methods of prevention or
treatment of diarrheal illness greatly influence child exposure to pathogens98. Maternal
hygiene behaviors, especially whether they safely prepare food and what type of
complimentary feeding practices they use can impact whether or not their child is
exposed to pathogens99,100. Additionally, maternal knowledge and practice of
preventative medical approaches to disease, such as ensuring that their child receives
vitamin A or zinc supplementation and is vaccinated, are other important factors
influencing disease risk in CU5101.
Basic Factors
As per the theoretical framework, the basic factors are the top-level drivers. Basic
factors affect immediate, underlying, and distal factors along a continuum. These
include the sociocultural, political, and large-scale economic drivers that permeate
society, as well as environmental drivers (e.g. rainfall, temperature, vegetation,
37
elevation, etc.). Societal practices, values, attitudes, and belief systems that influence
social norms and behaviors are all considered basic factors.
Data Overview
Survey Sites. This survey was conducted in the South and Southeast
departments of Haiti. The survey took place from October to November 2011 in regions
that have predominantly rural, socioeconomically disadvantaged populations located in
the commune municipalities of Aquin and Côtes-de-Fer. Both communes are remote
and isolated in mountainous areas of the country. The Saint Boniface foundation is a
longstanding non-profit in the community, offering a 60-bed primary care hospital and is
the only healthcare facility in the region. Saint Boniface sponsors several community
health interventions in the area102–104.The population of this region primarily practices
subsistence farming and also raise goats and pigs as their main source of income102. A
total of 800 households were selected for the survey using a two-stage sampling
method described elsewhere in two previously published studies using this
dataset102,104. In brief, children aged 6 to 59 months of age were randomly selected
using a census derived from the St. Boniface Hospital and their employed community
healthcare workers that serve the surrounding catchment. Only households that had a
CU5 were asked to participate in the survey. Overall, 828 women of child-bearing age
(15 to 49 years old) and their youngest child under the age of 5 years were recruited for
the study. For households with multiple children under 5, the youngest child-mother
dyad was selected for data collection.
This baseline, cross-sectional survey was conducted prior to the implementation
of pilot interventions to improve maternal and child health in the region. Prior to this
baseline survey, the overall study used a two-stage random sampling scheme. This
38
study was previously approved by the University of Florida Institutional Review Board
(IRB-02 clearance), and the St. Boniface Foundation hospital authorities in Haiti. The
survey consisted of household visits to conduct interviews and take serum (to measure
anemia) and anthropometric measurements from all eligible respondents from each
selected household. Written consent was obtained from each respondent prior to
assessment, and for CU5, consent was obtained from their parents or caregivers.
Interviews and assessments were conducted only after consent was obtained, and for
children with anemia, treatment was given free of charge102–104.
Data Cleaning and Manipulation. Survey data were cleaned and analyzed
using SAS version 9.4105. After translating the document to English, reorganizing and
reclassifying data into binary categories for analyses (see Figure 1-9), there were many
missing values. To adjust for this, skip patterns in the questionnaire were addressed.
Skip patterns are questions that were asked and depending on the respondent’s answer
will determine if the respondent will move onto the next question in that section
(sections are themes of the questionnaire such as food security or maternal health) or if
the respondent will end that section and move onto an entirely different set of questions.
Additionally, for some of the key explanatory and covariate variables in the analyses,
more than 5% were missing, even after controlling the skip patterns. Within SAS
software, any statistical procedure (e.g. regression analyses) will often exclude
observations with any missing variable values from analysis. Although analyzing only
respondents with complete data records has the advantage of simplicity (i.e. no
additional data cleaning/manipulation steps), the information contained in the
incomplete cases is lost. To adjust for this and to keep as many observations as
39
possible, while also minimizing bias, and obtain the appropriate estimates of
uncertainty, statistical imputation procedures were used in line with previous
research103.
The imputation procedure used is the Multiple Imputation (MI) procedure in SAS
9.4, which performs multiple imputation of missing data106. Similarly, Seraphin et. al.
also performed this specific MI procedure, but for an entirely different study objective
that looked at the determinants of institutional delivery among women aged 15-49
years103. We chose MI over single imputation, because single imputation does not
account for the uncertainty around the predictions of the unknown missing values, and
the resulting estimated variances of the parameter estimates will be biased toward
zero—whereas with MI, the model is unbiased by missing data because it replaces
each missing value with a set of plausible values that represent the uncertainty about
the best value to impute. After imputing the missing data, we analyzed the dataset in
SAS using customary procedures for complete data and combining the results from
these analyses into one (singular) estimate107. All missing patterns for each chapter
hypothesis was explored by sub-setting data relevant to each chapter’s specific
hypothesis, and then checking the missing patterns using means and frequency tables
of all variables in the analysis. Each missing dummy variable to run Little’s “Missing
Completely at Random” (MCAR) test110. Little’s test assesses if the missing data is
MCAR or missing at random (MAR) or not missing at random, on each variable in
question. To assess if the variable’s missingness is MCAR, the p value must be greater
than 0.05 (or not significant) and neither the variables in the dataset nor the unobserved
value of the variable itself predict whether a value will be missing111. Variables in this
40
research were considered to be MAR because of the survey design and skip patterns,
some variables in the dataset were predictive of missingness in another variable but this
wasn’t true for the variable in question (e.g. for questions to be answered more
frequently by women but not by men would indicate that the variable for “gender” would
predict missingness)112. To confirm this further, we visually inspected the data’s missing
pattern to determine whether the variables that had missing information exhibited a
monotonic trend or appeared to be MAR. Given the structure of this dataset, our
hypotheses, and previous research that used this raw data and had to impute103,104, a
Markov Chain Monte Carlo (MCMC) imputation method using Jeffreys Prior by using the
“PROC MI” procedure explored in SAS 9.4105.
Chapter Methods. We imputed the dataset separately for each chapter in line
with the hypotheses within each one. In Chapter 2, we focus on the association of
livestock ownership, dietary diversity and ASF consumption in CU5 and uses binary
multivariate logistic regression models to assess dietary diversity and ASF consumption
outcomes. In Chapter 3, we focus on the association of livestock ownership (with and
without inadequate WASH behaviors) and CU5 stunting, using multivariate logistic
regression models. In Chapter 4, we conducted an exploratory analysis to investigate
the relationship of environmental variables related to food production and spatial
associations with CU5 HAZ. To assess the relationship between CU5 HAZ and
environmental and spatial covariates, we employed a multivariate linear regression
using Ordinary Least Squares (OLS) and global and local spatial clustering and
autocorrelation detection.
41
Figures
B) Figure 1-1. A) Map of Haiti with the Aquin and Côtes-de-Fer study communes. B) Map of the Aquin (Flamands, Fonds
des Blancs, Guirand, and Frangipane) and Côtes-de-Fer (Jamais Vu) study area sub-section.
42
Figure 1-2. Percentage of CU5 that are stunted in Haiti compared to the DR according to the DHS66.
43
Figure 1-3. One-health theoretical framework to understand linkages between livestock ownership and child under five
nutrition in southern Haiti. Adapted from UNICEF 86.
44
Figure 1-4. Malnutrition terminology. Definitions of the various forms of malnutrition
and undernutrition from the Joint Child Malnutrition Estimates 2017 edition2.
Figure 1-5. Five 5’s diagram (adapted from Penakapapti et al.52.
Term Definition
Undernutrition Includes wasting or severe weight loss (low weight-for-height (WHZ)),
stunting or chronic growth retardation (low height-for-age (HAZ)), and
underweight (low weight-for-age (WAZ)), where an underweight child
may also be stunted, wasted or both
Micronutrient-related
undernutrition
Includes micronutrient deficiencies (a lack of important vitamins and
minerals-namely, Iodine, vitamin A, and iron) or micronutrient excess
Overweight Obesity and diet-related non-communicable diseases (such as heart
disease, stroke, diabetes and some cancers
45
Figure 1-6. Undernutrition pathway from pathogen exposure via livestock feces adapted from Penakapapti et al.52.
46
Figure 1-7. Improved nutrition via livestock ownership and safe WASH practices (blocked exposure to livestock feces and
improved access to safe and nutritious foods) adapted from Penakapapti et al. 52.
47
Figure 1-8. Variable list and description for all chapters. For maternal knowledge
scoring, see Appendix, Figure A-1.
Variables Description Chapter
Nutrition status and anthropometrics
CU5 Stunting Stunting was defined as: HAZ < −2.0 standard deviations (SD) below the 2006 WHO reference mean. Stunting status is a
binary score with children below -2 SD considered "stunted" while all other children considered "not stunted". Outlier
children greater than 5 SD or less than -5 SD were removed.
3
Dietary Diversity
Household Total Dietary Diversity Score
(HDDS)
Household dietary diversity was assessed using a qualitative 24-hour 18 food group consumption questionnaire, adapted to
Haiti. The binary HDDS score was calculated using the median DDS of the sample (i.e. 7). Scores that fell above the
median were considered to have a more diverse diet than the average and those that fell below the median were considered
to have a less diverse diet than the average for this sample.
2,3
ASF consumpion score Animal source food consumption as a sub-score of HDDS. Includes any consumption of livestock flesh or byproducts,
including: milk, meat, fish, and eggs. This binary scoring is based off the median ASF consumption score of the sample (i.e.
1) , those that fell above the median score were considered to have a more ASF consumption than the average. Those that
fell below the median score were considered to have a less ASF consumption than the average for this sample.
2,3
Livestock ownership
Any Species Includes any ownership of small or large ruminants, poultry or swine. This variable is binary. 2,3,4
Small Ruminants Households that owned goats and/or sheep. This variable is binary. 2,3,4
Large Ruminants Households that owned cattle, milk cows, donkeys, mules, and/or horses. This variable is binary. 2,3,4
Poultry Households that own chickens or other types of poultry. This variable is binary. 2,3,4
Swine Houesholds that own Pigs. This variable is binary. 2,3,4
Impoverishment
Impoverishment is a proxy for wealth. Wealth was assessed using the standard protocol and procedures developed
previously by Seraphin et al., The principle components were used to create a relative poverty index that captures the wealth
of the region, where everyone is considered "poor". This Impoverishment is a binary indicator ranges from 0 to 1,
representing least poor and poorest.
2,3
Transportation This variable is binary. Another proxy of impoverishment and also access. Indicates the household owned bike, car or
motorcycle.2,3
Number of CU5 living in the home This variable is binary. Indicates if the household has 1 child or more than 1 child under five years old living in the
household.2,3
Land ownership This variable is binary. Another proxy of impoverishment. Indicates if the household owns land, cultivates land or has tenure
over land.2,3
Child 2,3
Age This variable is binary. Age categories were dichotomized into 2 categories: Children 6 months to 2 years old and children
25 months to 5 years old.2,3
Vitamin A Supplementation Categories based off field data, with three categories (e.g. Vitamin A received, vitamin A not received, and do not know). 2,3
Maternal
Age Mothers ages ranged from 15 to 49 in the sample. Breakdowns of the maternal age are based off child-bearing, and
consistent with Seraphin et al., groups are: age 15-24, 25-34, and 35-49. 2,3
Employment Categories based off mothers responses, and include "farming", "steady work", "No job", and "Other". 2,3
Relationship Represents if mothers are married or not married (as a binary category). All categories outside of marriage or formal union
were considered not in a relationship.2,3
Education Represents if mothers received any formal education (e.g. primary, secondary or above) versus no formal education. 2,3
Mother and caregiver knowledge
Vitamin A and Iron Rich Food Knoweldge
Nutrition and malnutrition
Diarrhea risk*
Diarrhea prevention*
To assess maternal/caregiver knowledge surrounding nutrition, WASH and potential disease risk, four scores (each
measuring a particular knowledge construct: Nutrition (vitamin A and Iron), Nutrition and Malnutrition signs, diarrheal disease
risks, and diarrheal disease prevention). These constructs were created from questions listed in table 1-6. Each construct
was a summation of the questions in table 1-6 that were answered correctly. Mean scores were then taken for each
construct across all survey participants. To establish a knowledge score, the scores were dichotomized around these mean
scores for all study participants, per knowledge constuct, following Seraphin et al. method. The participants that fell below
the mean score were considered to be less knowledgeable while those participatns that fell above the mean score were
considered to be more knowledgeable.
2,3*
WASH
Latrine This variable is binary. This is the type of toilet facility; an improved latrine is considered to be a toilet that isn’t shared and
follows the WHO JMP standards.3
Child Stool Dispoal This variable is binary. This is based off how child stool is disposed of in the household. The variable reflect Improved child
stool practices (e.g. "Child used toilet", "Put/rinsed into toilet", "Put/rinsed into drain", versus unimproved (e.g. "threw in the
trash", "left it in the open", and "other").
3
Household Waste Disposal This variable is binary. it reflects the Household waste disposal practices are considered improved versus unimproved.
Unimproved waste disposal include answers such as: "Dump in Street/Open Space", 'burn it", and "other". Improved waste
disposal incude: "Bury it", and "Dispose of on farm/compost".
3
Water Source This variable is binary. An improved water source includes water that is piped water supply, and stored water from protected
wells, springs, public standpipes or stored rainwater.3
Distance to Water This variable is binary. Distance to water as considered improved if the water was less than 30 minutes away, round trip.
Distance to water was considered to be umimptoved if the time it took to retrieve water was over 30 minutes, round trip. 3
Water treatment This variable is binary. Water treatment is considered to be improved if the water treatment falls into categories such as
boiling the water, treating the water with chlorine or disinfectant tablets, Treatment with chlorine or equivalent, boiling of
water, solar disinfecting, etc.
3
Handwashing Practices This variable is binary. This variable is based off response from mothers and caregivers who practice safe handwashing
practices before cooking, eating or using the latrine.3
Disease Status and Prevention
Diarrheal Disease This variable is binary. It reflects the Cu5 passing of 2-3 loose or watery stools per day, within the 2 weeks of the survey. 3
Febrile Illness This variable is binary. It reflects the Cu5 having a fever within 2 weeks of the survey. 3
Breast Feeding Status This variable is binary. It reflects children that are currently breastfeeding. 3
Deworming This variable is binary. It reflects that the cu5 has recived deworming medication. 3
Environmental and Spatial
Elevation Measure of the height above seal level in meters. 4
Vegetation (NDVI) Index of vegetation conditions from NASA MODIS. Ranges from -1 (no vegetation) to 1 (complete vegetated) per 250
meters.4
Land Surface Temperature (Day and
Night)
Temperature from NASA MODIS, calculated Kelvin and converted to celsius degrees.4
Precipitation Long-term cumulative (i.e. over 3 months) rainfall data based on average monthly rainfall in milimeters from 1970 - 2000 4
Population Density measurement of the number of people per 100 meters squared. 4
Accessibility Travel-time measure of the distance to the nearest urban center. 4
Distancee to Health Facility Euclidean distance from St. Bonifcace Hospital. 4
Distancee to Roads Euclidean distance from established road network. 4
Slope Percentage rise in elevation, calculated in ArcGIS software. 4
*assessed only in chapter 3
48
CHAPTER 2 I -LIVESTOCK OWNERSHIP AND DIETARY DIVERSITY OF CU5 IN RURAL HAITIAN
HOUSEHOLDS
Introduction
According to the World Health Organization (WHO), malnutrition/undernutrition
refers to “deficiencies, excesses, or imbalances in a person’s intake of energy and/or
nutrients”1. Undernutrition is a serious problem plaguing many low- and middle-income
countries (LMICs). Haiti is one of the poorest countries in the western hemisphere, and
suffers the highest rates of undernutrition in Latin America and the Caribbean65.
Children under 5 years old (CU5) are particularly vulnerable to undernutrition. These
nutritional deficits, if chronic, can affect the development status of CU5, including their
linear growth patterns and cognitive functioning. Thus, CU5 micro- and macronutrient
deficiencies, as well as immune function and disease statuses (both symptomatic and
asymptomatic), can lead to recurring undernutrition, which can have immediate and
lasting effects on their health and well-being13. Therefore, it is essential that CU5 have
an adequate dietary intake pattern consisting of safe, nutritious, and diverse food
groups to promote and foster proper growth and development patterns.
Dietary diversity is the universal term and measure associated with (1) the overall
quality, and (2) the nutrient adequacy of a person’s dietary practices. Dietary diversity
has been shown to be a strong predictor of CU5 nutrition and has been found to show
an association with CU5 stunting (-2 to -3 standard deviations below normal Height for
Age Z scores [HAZ])15,16. Dietary diversity considers an individual or household’s
consumption of a higher number of food groups compared with a set standard amount
of food groups considered to be adequate 15,18. Usually, medium or moderate is termed
adequate [compared to low and high-dietary diversity 15,18. To assess an individual’s
49
dietary diversity, a simple tool has been developed and tested called the dietary
diversity score (DDS). The DDS counts the number of food groups consumed by an
individual or household over a given reference period.18 Several studies show a high
DDS is positively associated with overall dietary quality, particularly with improved
micronutrient consumption in CU515,22. Studies have shown that dietary diversity and
infant feeding practices vary by department (i.e. region) in Haiti, in general
demonstrating low dietary diversity and poor infant feeding practices associated with
underweight, wasting, and stunting across Haitian CU516.
Moreover, livestock has the potential to provide food and nutritional security, as
well as income and livelihood, to nearly one billion poor people in LMICs35. With the
worldwide increase in demand for livestock, owning and rearing livestock have the
potential to provide many benefits--increasing ASF access, availability, income (to
purchase ASF or other diverse foods in markets) 14,16,26–31,67,85 . In Haiti, livestock too
can potentially increase animal source food (ASF) access and consumption (especially
for CU5), reduce vulnerability and improve livelihoods with food and income14,16,26–
31,67,85. These livestock benefits may provide better nutritional statuses and overall
health outcomes for CU5 in Haiti.
Safe ASF, if available and accessible to families in need, have the potential to
improve CU5 nutrition by impacting dietary quality113. ASF such as milk, meat, fish, and
eggs are rich in bioavailable vitamin B12, riboflavin, iron, calcium, zinc, and a variety of
essential amino acids23. These are necessary for positive CU5 growth and nutrition
outcomes. Ultimately, for many vulnerable groups (i.e. CU5), ASF consumption may be
50
the only means to absorb these critical vitamins and micronutrients (especially vitamin
B12) in their diet28.
Research Objective
There is limited empirical evidence to support a relationship between small-scale
livestock production, ASF consumption, and nutritional status in children under 5 years
old in southern Haiti. Thus, the goal of this research chapter is to assess if a
household’s (with a CU5) ownership of livestock is associated with the overall
household’s dietary diversity and consumption of ASF. The hypotheses for this chapter
are: (1) household livestock ownership is associated with greater dietary diversity
scores (HDDS) in households of CU5 in rural southern Haiti; and (2) household
livestock ownership is associated with greater ASF consumption in households of CU5
in rural southern Haiti. To assess these hypotheses, this analysis investigated whether
certain livestock species or groups (i.e., small ruminant, large ruminant, poultry or
swine) were associated with either HDDS or ASF consumption in the CU5 surveyed,
when accounting for covariates.
Methods
Data for this study came from a cross-sectional, household-based survey
conducted from October to November 2011 in a predominantly rural region of about
65,000 inhabitants in the Aquin and Côtes-de-Fer communes of southern Haiti102–104.
The survey selected 828 households from the Institut Haïtien de Statistique et
d’Informatique (Haiti’s census) to participate in the survey using a random, two-staged
sampling design. The first stage included a selection of 30 out of 69 villages. In the
second stage, households within each of the village cluster were selected randomly.
51
Within each household selected to participate, a mother (or caretaker) and their CU5
was then selected to participate in the survey (n=828)102–104.
For this secondary data analysis, only observations containing a mother and CU5
(between the ages of 6 months and 59 months) pair were selected for inclusion and
further analyses. All other observations were excluded. Moreover, detailed descriptions
of variable construction (e.g. recoding, statistics and data manipulation) for variables are
referenced elsewhere in Chapter 1, subsection: Data Overview as well as Figure 1-8.
The outcome variables of interest were the (1) HDDS and (2) ASF consumption
score. In brief, the HDDS is a composite score of all food groups (Figure 2-1) consumed
by the entire household (including any CU5) within a previously defined dietary recall
period. For this study, the recall period was 24 hours, and the total raw scoring was out
of 18 food groups. In comparison to other studies that usually construct dietary diversity
score out of 9 to 12 groups (to measure low, medium and high DDS), this study assess
whether a factor is contributing to either higher HDDS or low HDDS. HDDS is recoded
into a binary outcome variable following a similar approach used by Mukherjee et al. 114,
using the median HDDS in the sample (i.e. median =7). A score of “1” was assigned to
a household that fell above this median, indicating the household members consumed a
more diverse diet than the average for this sample. A score of “0” reflected the opposite.
ASF consumption score was also assessed. This was a subset of a household’s total
DDS and was the sum of any meat product, fish, dairy or egg consumption. This raw
score was out of 6 groups. Like the binary HDDS score, the ASF consumption score
was also binarized, based off the median ASF consumption score for the sample (i.e.
52
median = 1). A score of “1” indicated more ASF consumption than the median for this
sample; a score of “0” indicated less ASF consumption than the median for this sample.
The complete list of variables (including covariates) and their description are in
Figure 2-2. The main independent variable(s) for these logistic regression models was
livestock ownership status (i.e., “household owns livestock”), particularly whether a
Household Total Dietary Diversity Score (HDDS) Household dietary diversity was assessed using a qualitative 24-hour 18 food group consumption questionnaire,
adapted to Haiti. The binary HDDS score was calculated using the median DDS of the sample (i.e. 7). Scores
that fell above the median were considered to have a more diverse diet than the average and those that fell below
the median were considered to have a less diverse diet than the average for this sample.ASF consumpion score Animal source food consumption as a sub-score of HDDS. Includes any consumption of livestock flesh or
byproducts, including: milk, meat, fish, and eggs. This binary scoring is based off the median ASF consumption
score of the sample (i.e. 1) , those that fell above the median score were considered to have a more ASF
consumption than the average. Those that fell below the median score were considered to have a less ASF
consumption than the average for this sample.Livestock ownership
Any Species Includes any ownership of small or large ruminants, poultry or swine. This variable is binary.
Small Ruminants Households that owned goats and/or sheep. This variable is binary.
Large Ruminants Households that owned cattle, milk cows, donkeys, mules, and/or horses. This variable is binary.
Poultry Households that own chickens or other types of poultry. This variable is binary.
Swine Houesholds that own Pigs. This variable is binary.
Impoverishment
Impoverishment score Impoverishment is a proxy for wealth. Wealth was assessed using the standard protocol and procedures
developed previously by Seraphin et al., The principle components were used to create a relative poverty index
that captures the wealth of the region, where everyone is considered "poor". This Impoverishment is a binary
indicator ranges from 0 to 1, representing least poor and poorest. Transportation This variable is binary. Another proxy of impoverishment and also access. Indicates the household owned bike,
car or motorcycle.Number of CU5 living in the home This variable is binary. Indicates if the household has 1 child or more than 1 child under five years old living in the
household.Land ownership This variable is binary. Another proxy of impoverishment. Indicates if the household owns land, cultivates land or
has tenure over land.Child
Age This variable is binary. Age categories were dichotomized into 2 categories: Children 6 months to 2 years old and
children 25 months to 5 years old.Vitamin A Supplementation Categories based off field data, with three categories (e.g. Vitamin A received, vitamin A not received, and do not
know).Maternal
Age Mothers ages ranged from 15 to 49 in the sample. Breakdowns of the maternal age are based off child-bearing,
and consistent with Seraphin et al., groups are: age 15-24, 25-34, and 35-49.
Employment Categories based off mothers responses, and include "farming", "steady work", "No job", and "Other".
Relationship Represents if mothers are married or not married (as a binary category). All categories outside of marriage or
formal union were considered not in a relationship.
Education Represents if mothers received any formal education (e.g. primary, secondary or above) versus no formal
education. Mother and caregiver knowledge
Vitamin A and Iron Rich Food Knoweldge
Nutrition and malnutrition
To assess maternal/caregiver knowledge surrounding nutrition were created from questions listed in Chapter 1.
The scores are each dichotomized around the mean score following Seraphin et al. methods previously
developed. Scores were given to all study participants, per knowledge construct. The participants that fell below
the mean score were considered to be less knowledgeable regarding the construct compared to the average for
the sample on that construct. In contrast, participatns that fell above the mean score were considered to be
more knowledgeable than the average for the construct, for the sample.
66
Figure 2-3. Descriptive statistics of survey respondents, overall.
%
Name of Variable N (Y/N)
Land ownership 435 (24% / 76%)
Any Livestock Ownership 421 (32% / 68%)
Food Security
In the last four weeks the house had enough to eat 435 (41% / 59%)
In the last 4 weeks the household could eat the food they wanted because there was enough food 430 (47% / 53%)
In the last 4 weeks, the household did not have to eat less food because the household had enough to eat 432 (51% / 49%)
Impoverishment 439 (49% / 51%)
Maternal Characteristics
Maternal Education Status 286 (68% / 32%)
Maternal Relationship Status 399 (31% / 69%)
Maternal Knowledge Score
Vitamin A and Iron Rich Food Sources 438 (50% / 50%)
Overall Nutrition and Signs of Malnutrition 433 (76% / 24%)
Maternal Age Categories
15-24 131 31%
25-34 193 46%
35-49 92 22%
Maternal Employment Status
Farming 32 10%
Steady Work 75 24%
No Income 74 23%
Other 138 43%
Child Characteristics
Age categories 6 to 24 months (vs. 2 to 5 years old) 439 (41% / 59%)
CU5 breastfeeding 410 (37% / 63%)
CU5 Gender (Males to Females)
Female 232 53%
Male 207 47%
Vitamin A Supplementation Status
Yes 101 26%
No 243 63%
Don't Know 44 11%
67
Figure 2-4. Descriptive statistics of livestock ownership, HDDS, and ASF consumption
by sub-communal section.
Figure 2-5. Bivariate regression results for study variables and HDDS.
Section Guirand Frangipane Flamands Fond des Blancs Jamais Vu All Sections
Variable Total (N) 73 128 71 90 77 439Any Livestock Ownership 67% 73% 70% 52% 62% 65%Livestock species specific count
Large Ruminant Animals
Own 49% 41% 55% 29% 40% 42%
Do Not Own 48% 48% 37% 70% 56% 52%
Did Not Answer 3% 10% 8% 1% 4% 6%
Small Ruminant Animals*
Own 51% 58% 51% 32% 47% 46%
Do Not Own 47% 30% 42% 67% 51% 48%
Did Not Answer 3% 12% 7% 1% 3% 6%
Chickens*
Own 53% 62% 61% 46% 43% 54%
Do Not Own 45% 28% 32% 51% 52% 41%
Did Not Answer 1% 10% 7% 3% 5% 6%
Pigs*
Own 33% 43% 30% 19% 32% 32%
Do Not Own 64% 45% 62% 78% 65% 61%
Did Not Answer 3% 12% 8% 3% 3% 6%
HDDS
Household DDS above average 40% 25% 42% 36% 42% 38%
Household DDS below average 55% 30% 48% 56% 54% 49%
cognitively” as a result of recurrent poor nutrition.
Stunting can have devastating
lifelong impacts on a child
affected
A child that is too thin for his or her height
It is referred to as acute
malnutrition, rapid weight loss
or the failure to gain weight. Wasting, without treatment,
puts a child at increased risk of
death
A child that is too heavy for his or her weight
It is referred to as obesity that
results from an imbalance of
calorie expenditure and intake from food and drinks. Children
suffering from obesity have
long-term risks of
noncommunicable diseases
A child suffering from both stunting and overweight
undernutrition
Research is ongoing to
determine the joint estimates and long-term effects from
these combined conditions
A child suffering from both stunting and wasting
undernutrition
Research is ongoing to
determine the joint estimates and long-term effects from
these combined conditions
84
Figure 3-2. Variable Descriptions used in chapter 3 analyses
Variables Description
Nutrition status and anthropometrics
CU5 Stunting Stunting was defined as: HAZ < −2.0 standard deviations (SD) below the 2006 WHO reference mean. Stunting
status is a binary score with children below -2 SD considered "stunted" while all other children considered "not
stunted". Outlier children greater than 5 SD or less than -5 SD were removed.
Livestock ownership
Any Species Includes any ownership of small or large ruminants, poultry or swine. This variable is binary.
Small Ruminants Households that owned goats and/or sheep. This variable is binary.
Large Ruminants Households that owned cattle, milk cows, donkeys, mules, and/or horses. This variable is binary.
Poultry Households that own chickens or other types of poultry. This variable is binary.
Swine Houesholds that own Pigs. This variable is binary.
Impoverishment
Impoverishment score Impoverishment is a proxy for wealth. Wealth was assessed using the standard protocol and procedures
developed previously by Seraphin et al., The principle components were used to create a relative poverty index
that captures the wealth of the region, where everyone is considered "poor". This Impoverishment is a binary
indicator ranges from 0 to 1, representing least poor and poorest.
Transportation This variable is binary. Another proxy of impoverishment and also access. Indicates the household owned bike,
car or motorcycle.
Number of CU5 living in the home This variable is binary. Indicates if the household has 1 child or more than 1 child under five years old living in the
household.
Land ownership This variable is binary. Another proxy of impoverishment. Indicates if the household owns land, cultivates land or
has tenure over land.
Child
Age This variable is binary. Age categories were dichotomized into 2 categories: Children 6 months to 2 years old and
children 25 months to 5 years old.
Vitamin A Supplementation Categories based off field data, with three categories (e.g. Vitamin A received, vitamin A not received, and do not
know).
Maternal
Age Mothers ages ranged from 15 to 49 in the sample. Breakdowns of the maternal age are based off child-bearing,
and consistent with Seraphin et al., groups are: age 15-24, 25-34, and 35-49.
Employment Categories based off mothers responses, and include "farming", "steady work", "No job", and "Other".
Relationship Represents if mothers are married or not married (as a binary category). All categories outside of marriage or
formal union were considered not in a relationship.
Education Represents if mothers received any formal education (e.g. primary, secondary or above) versus no formal
education.
Mother and caregiver knowledge
Vitamin A and Iron Rich Food
Knoweldge
Nutrition and malnutrition
Diarrhea risk
Diarrhea prevention
To assess maternal/caregiver knowledge surrounding nutrition, WASH and potential disease risk, four scores
(each measuring a particular knowledge construct: Nutrition (vitamin A and Iron), Nutrition and Malnutrition signs,
diarrheal disease risks, and diarrheal disease prevention). These constructs were created from questions listed
in table 1-6. Each construct was a summation of the questions in table 1-6 that were answered correctly. Mean
scores were then taken for each construct across all survey participants. To establish a knowledge score, the
scores were dichotomized around these mean scores for all study participants, per knowledge constuct, following
Seraphin et al. method. The participants that fell below the mean score were considered to be less
knowledgeable while those participatns that fell above the mean score were considered to be more
knowledgeable.
WASH
Latrine This variable is binary. This is the type of toilet facility; an improved latrine is considered to be a toilet that isn’t
shared and follows the WHO JMP standards.
Child Stool Dispoal This variable is binary. This is based off how child stool is disposed of in the household. The variable reflect
Improved child stool practices (e.g. "Child used toilet", "Put/rinsed into toilet", "Put/rinsed into drain", versus
unimproved (e.g. "threw in the trash", "left it in the open", and "other").
Household Waste Disposal This variable is binary. it reflects the Household waste disposal practices are considered improved versus
unimproved. Unimproved waste disposal include answers such as: "Dump in Street/Open Space", 'burn it", and
"other". Improved waste disposal incude: "Bury it", and "Dispose of on farm/compost".
Water Source This variable is binary. An improved water source includes water that is piped water supply, and stored water
from protected wells, springs, public standpipes or stored rainwater.
Distance to Water This variable is binary. Distance to water as considered improved if the water was less than 30 minutes away,
round trip. Distance to water was considered to be umimptoved if the time it took to retrieve water was over 30
minutes, round trip.
Water treatment This variable is binary. Water treatment is considered to be improved if the water treatment falls into categories
such as boiling the water, treating the water with chlorine or disinfectant tablets, Treatment with chlorine or
equivalent, boiling of water, solar disinfecting, etc.
Handwashing Practices This variable is binary. This variable is based off response from mothers and caregivers who practice safe
handwashing practices before cooking, eating or using the latrine.
Disease Status and Prevention
Diarrheal Disease This variable is binary. It reflects the Cu5 passing of 2-3 loose or watery stools per day, within the 2 weeks of the
survey.
Febrile Illness This variable is binary. It reflects the Cu5 having a fever within 2 weeks of the survey.
Breast Feeding Status This variable is binary. It reflects children that are currently breastfeeding.
Deworming This variable is binary. It reflects that the cu5 has recived deworming medication.
85
Figure 3-3. Descriptive statistics of surveyed households.
Figure 3-4. Descriptive Statistics of WASH characteristics (Improved “I” and
Unimproved “U”) broken down by sub communal section
%
Name of Variable N (Y/N)
Land ownership 435 (24% / 76%)
Any Livestock Ownership 421 (32% / 68%)
Food Security
In the last four weeks the house had enough to eat 435 (41% / 59%)
In the last 4 weeks the household could eat the food they wanted 430 (47% / 53%)
In the last 4 weeks, the household did not have to eat less food because the household had
enough to eat 432 (51% / 49%)
WASH Characteristics
Time to fetch water (under vs. over 30 min) 437 (32% / 68%)
Figure 3-5. Bivariate regression results for study variables and CU5 Stunting
Name of Variable
Beta
Estimate SE* p**
Land ownership 0.13 0.04 0.00 Livestock Species Specific Information
Any Livestock Ownership 0.04 0.03 0.18 Owns Large Ruminant Animals 0.12 0.03 0.00 Owns Small Ruminant Animals -0.11 0.03 0.00 Owns Pigs -0.12 0.03 0.00
Food Security
In the last four weeks the house did not have enough to eat 0.24 0.03 0.00 In the last 4 weeks the household could eat the food they wanted -0.18 0.03 0.00 In the last 4 weeks, the household did not have to eat less food because
the household had enough to eat
-0.22 0.03 0.00
WASH Characteristics
Time to fetch water (under vs. over 30 min) -0.40 0.03 0.00 Water Source Status -0.82 0.06 0.00 Improved toilet 0.38 0.03 0.00 Child Stool Disposal Status: Unimproved 0.31 0.03 0.00 Household Waste Disposal Status: Unimproved to Improved -0.10 0.03 0.00 Handwashing Before Feeding CU5 -0.82 0.22 0.00
Dietary Diveristy
Household has diverse diet -0.22 0.03 0.00 Impoverishment
Impoverishment -0.33 0.03 0.00 Access to Transportation (no access vs. access) 0.45 0.03 0.00 More than one CU5 living in household 0.36 0.03 0.00
Maternal Characteristics
Mother is not educated -0.06 0.03 0.06 Maternal Formal Relationship Status -0.20 0.03 0.00 Maternal Knowledge Score
Vitamin A and Iron Rich Food Sources 0.21 0.03 0.00 Diarrheal Disease Prevention 0.24 0.04 0.00 Diarrheal Disease Risk -0.10 0.03 0.00
Child Characteristics
Fever Episode in Last 2 Weeks -0.37 0.03 0.00 Diarrheal Disease Episode in Last 2 Weeks -0.21 0.04 0.00 Deworming Supplementation -0.12 0.03 0.00 Vitamin A Supplementation Status 0.08 0.02 0.00
*SE=Standard Error
**P-value (p<0.20)
Stunting
87
Figure 3-6. Multivariate binary backward-stepwise logistic regression results for
model 1 assessing the association of livestock ownership and CU5 Stunting status.
Name of Variable OR* CI** p***
Livestock Species Specific Information
Any Livestock Ownership 0.26 (0.18, 0.39) <.0001
Reference: No Livestock Ownership
Owns Large Ruminant Animals 0.79 (0.64, 0.99) 0.04
Reference: No Large Ruminant Ownership
Owns Small Ruminant Animals 0.23 (0.16, 0.32) <.0001
Reference: No Small Ruminant Ownership
Food Security
In the last four weeks the house did not have enough to eat 2.04 (1.50, 2.79) <.0001
Reference: In the last four weeks the house had enough to eat
In the last 4 weeks the household could eat the food they wanted 0.28 (0.20, 0.39) <.0001
Reference: In the last 4 weeks the household could not eat the food they wanted
In the last 4 weeks, the household had to eat less food because the household did not
have enough to eat 5.71 (4.27, 7.63) <.0001
Reference: In the last 4 weeks, the household did not have to eat less food because the
household had enough to eat
Dietary Diversity
Household has diverse diet 0.62 (0.53, 0.74) <.0001
Reference: Household does not have a diverse diet
Impoverishment
No access to Transportation 2.27 (1.81, 2.86) <.0001
Reference: Access to Transportation
More than one CU5 living in household 1.66 (1.37, 2.01) <.0001
Reference: Only one CU5 living in household
Maternal Characteristics
Maternal Education
Mother is not educated 1.23 (1.04, 1.46) 0.02
Reference: Mother is educated
Maternal Knowledge Score
Mother is less knowledgeable of vitamin A and Iron Rich foods 2.18 (1.84, 2.59) <.0001
Reference: Mother is more knowledgeable of vitamin A and Iron rich foods
Child Characteristics
Under 2 years old 0.59 (0.50, 0.70) <.0001
Reference: 2 to 5 years old
*OR = Odds Ratio
**CI = 95% Confidence Limits
*** = P-value (p<0.05)
Model 1: Livestock
88
Figure 3-7. Multivariate binary backward-stepwise logistic regression results for
model 2 association the of livestock ownership, WASH Factors, and CU5 Stunting status.
Name of Variable OR* CI** p***
Livestock Species Specific Information
Any Livestock Ownership 0.36 (0.24, 0.53) <.0001
Reference: No Livestock Ownership
Owns Large Ruminant Animals 0.76 (0.60, 0.97) 0.03
Reference: No Large Ruminant Ownership
Owns Small Ruminant Animals 0.41 (0.30, 0.56) <.0001
Reference: No Small Ruminant Ownership
Owns Pigs 0.77 (0.62, 0.97) 0.02
Reference: No Pig Ownership
Food Security
In the last four weeks the house did not have enough to eat 2.30 (1.67, 3.15) <.0001
Reference: In the last four weeks the house had enough to eat
In the last 4 weeks the household could eat the food they wanted 0.38 (0.28, 0.52) <.0001
Reference: In the last 4 weeks the household could not eat the food they wanted
In the last 4 weeks, the household had to eat less food because the household did not
have enough to eat 4.11 (3.04, 5.55) <.0001
Reference: In the last 4 weeks, the household did not have to eat less food because the
household had enough to eat
Dietary Diversity
Household has diverse diet 0.54 (0.44, 0.65) <.0001
Reference: Household does not have diverse diet
WASH Characteristics
Time to fetch water (under vs. over 30 min) 0.30 (0.25, 0.35) <.0001
Reference: Time to fetch water (over 30 min vs. under)
power of each model step until final model for chapter 3, Model 2: Livestock, WASH, and Stunting.
123
LIST OF REFERENCES
1. WHO. WHO | Malnutrition. WHO (2017).
2. UNICEF, WHO & The World Bank. Levels and Trends in Child Malnutrition. Joint Child Malnutrition Estimates edition (2017). doi:10.1016/S0266-6138(96)90067-4
3. WHO | WHO child growth standards and the identification of severe acute malnutrition in infants and children. WHO (2014).
4. Leroy, J. L., Ruel, M., Habicht, J.-P. & Frongillo, E. A. Linear Growth Deficit Continues to Accumulate beyond the First 1000 Days in Low- and Middle-Income Countries: Global Evidence from 51 National Surveys. J. Nutr. (2014). doi:10.3945/jn.114.191981
5. Martorell, R. Improved nutrition in the first 1000 days and adult human capital and health. Am. J. Hum. Biol. 29, e22952 (2017).
6. Syed, S., Ali, A. & Duggan, C. Environmental Enteric Dysfunction in Children: A Review. J. Pediatr. Gastroenterol. Nutr. 1 (2016). doi:10.1097/MPG.0b013e318228574e
7. Crane, R. J., Jones, K. D. J. & Berkley, J. A. Environmental enteric dysfunction: An overview. Food Nutr. Bull. (2015). doi:10.1177/15648265150361S113
8. Owino, V. et al. Environmental Enteric Dysfunction and Growth Failure/Stunting in Global Child Health. Pediatrics (2016). doi:10.1542/peds.2016-0641
9. Caulfield, L. E. Relationship between growth and illness, enteropathogens and dietary intakes in the first 2 years of life: findings from the MAL-ED birth cohort study. BMJ Glob. Heal. (2017). doi:10.1136/bmjgh-2017-000370
10. Valdez, Y., Brown, E. M. & Finlay, B. B. Influence of the microbiota on vaccine effectiveness. Trends in Immunology (2014). doi:10.1016/j.it.2014.07.003
11. Praharaj, I., John, S. M., Bandyopadhyay, R. & Kang, G. Probiotics, antibiotics and the immune responses to vaccines. Philosophical Transactions of the Royal Society B: Biological Sciences (2015). doi:10.1098/rstb.2014.0144
12. Dewey, K. G. & Begum, K. Long-term consequences of stunting in early life. Matern. Child Nutr. 7, 5–18 (2011).
13. Victora, C. G. et al. Maternal and child undernutrition: consequences for adult health and human capital. The Lancet 371, 340–357 (2008).
14. Mosites, E. M. et al. The relationship between livestock ownership and child stunting in three countries in eastern Africa using national survey data. PLoS One 10, (2015).
124
15. Rah, J. H. et al. Low dietary diversity is a predictor of child stunting in rural Bangladesh. Eur. J. Clin. Nutr. (2010). doi:10.1038/ejcn.2010.171
16. Heidkamp, R. A., Ayoya, M. A., Teta, I. N., Stoltzfus, R. J. & Marhone, J. P. Complementary feeding practices and child growth outcomes in Haiti: an analysis of data from Demographic and Health Surveys. Matern. Child Nutr. (2015). doi:10.1111/mcn.12090
17. Arimond, M. & Ruel, M. T. Dietary Diversity Is Associated with Child Nutritional Status: Evidence from 11 Demographic and Health Surveys. J. Nutr. (2004). doi:10.1093/jn/134.10.2579
18. Ruel, M. T. Is dietary diversity an indicator of food security or dietary quality? A review of measurement issues and research needs. Food Nutr. Bull. (2003). doi:10.1177/156482650302400210
19. Darapheak, C., Takano, T., Kizuki, M., Nakamura, K. & Seino, K. Consumption of animal source foods and dietary diversity reduce stunting in children in Cambodia. Int. Arch. Med. 6, (2013).
20. Neumann, C. G. et al. Animal source foods improve dietary quality, micronutrient status, growth and cognitive function in Kenyan school children: background, study design and baseline findings. J. Nutr. 133, 3941S–3949S (2003).
21. Iannotti, L. & Lesorogol, C. Animal milk sustains micronutrient nutrition and child anthropometry among pastoralists in Samburu, Kenya. Am. J. Phys. Anthropol. (2014). doi:10.1002/ajpa.22547
22. Steyn, N., Nel, J., Nantel, G., Kennedy, G. & Labadarios, D. Food variety and dietary diversity scores in children: are they good indicators of dietary adequacy? Public Health Nutr. (2006). doi:10.1079/PHN2005912
23. Murphy, S. P. & Allen, L. H. Animal Source Foods to Improve Micronutrient Nutrition and Human Function in Developing Countries. J. Nutr 133, 3965–3971 (2003).
24. Muslimatun, S. & Wiradnyani, L. A. A. Dietary diversity, animal source food consumption and linear growth among children aged 1–5 years in Bandung, Indonesia: a longitudinal observational study. British Journal of Nutrition (2016). doi:10.1017/S0007114515005395
25. Darrouzet-Nardi, A. F. et al. Child dietary quality in rural Nepal: Effectiveness of a community-level development intervention. Food Policy (2016). doi:10.1016/j.foodpol.2016.03.007
26. Randolph, T. F. et al. Role of livestock in human nutrition and health for poverty reduction in developing countries. J. Anim. Sci. 85, 2788–800 (2007).
125
27. Thornton, P. K. Livestock production: recent trends, future prospects. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 365, 2853–2867 (2010).
28. Jin, M. & Iannotti, L. L. Livestock production, animal source food intake, and young child growth: The role of gender for ensuring nutrition impacts. Soc. Sci. Med. 105, 16–21 (2014).
29. Azzarri, C., Zezza, A., Haile, B. & Cross, E. Does Livestock Ownership Affect Animal Source Foods Consumption and Child Nutritional Status? Evidence from Rural Uganda. J. Dev. Stud. 51, 1034–1059 (2015).
30. Thumbi, S. M. et al. Linking human health and livestock health: A ‘one-health’ platform for integrated analysis of human health, livestock health, and economic welfare in livestock dependent communities. PLoS One 10, (2015).
31. Hetherington, J. B., Wiethoelter, A. K., Negin, J. & Mor, S. M. Livestock ownership, animal source foods and child nutritional outcomes in seven rural village clusters in Sub-Saharan Africa. Agric. Food Secur. 6, 9 (2017).
32. Köster, E. P. Diversity in the determinants of food choice: A psychological perspective. Food Qual. Prefer. (2009). doi:10.1016/j.foodqual.2007.11.002
33. Powell, B., Bezner Kerr, R., Young, S. L. & Johns, T. The determinants of dietary diversity and nutrition: Ethnonutrition knowledge of local people in the East Usambara Mountains, Tanzania. J. Ethnobiol. Ethnomed. (2017). doi:10.1186/s13002-017-0150-2
34. Marteau, T. M., Hollands, G. J. & Fletcher, P. C. Changing human behavior to prevent disease: The importance of targeting automatic processes. Science (2012). doi:10.1126/science.1226918
35. Swanepoel, F., Stroebel, a. & Moyo, S. The role of livestock in developong communities: Enhancing multifunctionality. Proceedings of the satellite symposium (2008).
36. Marquis, G. S. et al. Fecal contamination of shanty town toddlers in households with non-corralled poultry, Lima, Peru. Am. J. Public Health (1990). doi:10.2105/AJPH.80.2.146
37. Mbuya, M. N. N. et al. Design of an intervention to minimize ingestion of fecal microbes by young children in Rural Zimbabwe. Clin. Infect. Dis. 61, S703–S709 (2015).
38. George, C. M. et al. Fecal markers of environmental enteropathy are associated with animal exposure and caregiver hygiene in Bangladesh. Am. J. Trop. Med. Hyg. (2015). doi:10.4269/ajtmh.14-0694
126
39. Headey, D. et al. Is exposure to animal feces harmful to child nutrition and health outcomes? A multicountry observational analysis. American Journal of Tropical Medicine and Hygiene 96, 961–969 (2017).
40. Dumas, S. E., Kassa, L., Young, S. L. & Travis, A. J. Examining the association between livestock ownership typologies and child nutrition in the Luangwa Valley, Zambia. PLoS One (2018). doi:10.1371/journal.pone.0191339
41. Ayele, Z. & Peacock, C. Improving access to and consumption of animal source foods in rural households: the experiences of a women-focused goat development program in the highlands of Ethiopia. J. Nutr. 133, 3981S–3986S (2003).
42. Headey, D. & Hirvonen, K. Is exposure to poultry harmful to child nutrition? An observational analysis for rural Ethiopia. PLoS One 11, (2016).
43. Iannotti, L. L. et al. Eggs in Early Complementary Feeding and Child Growth: A Randomized Controlled Trial. Pediatrics 140, e20163459 (2017).
44. Amour, C. et al. Epidemiology and Impact of Campylobacter Infection in Children in 8 Low-Resource Settings: Results from the MAL-ED Study. Clin. Infect. Dis. (2016). doi:10.1093/cid/ciw542
45. J.A., P.-M. et al. Prevalence of non-jejuni/coli campylobacter species detected by enzyme immunoassay and culture in the mal-ed cohort study. American Journal of Tropical Medicine and Hygiene (2013).
46. Platts-Mills, J. A. et al. Pathogen-specific burdens of community diarrhoea in developing countries: A multisite birth cohort study (MAL-ED). Lancet Glob. Heal. (2015). doi:10.1016/S2214-109X(15)00151-5
47. Humphrey, J. H. Child undernutrition, tropical enteropathy, toilets, and handwashing. Lancet (2009). doi:10.1016/S0140-6736(09)60950-8
48. Harper, K. M., Mutasa, M., Prendergast, A. J., Humphrey, J. & Manges, A. R. Environmental enteric dysfunction pathways and child stunting: A systematic review. PLoS Negl. Trop. Dis. 12, (2018).
49. Clasen, T. F. et al. Interventions to improve disposal of human excreta for preventing diarrhoea. in Cochrane Database of Systematic Reviews (2010). doi:10.1002/14651858.CD007180.pub2
50. Stocks, M. E. et al. Effect of Water, Sanitation, and Hygiene on the Prevention of Trachoma: A Systematic Review and Meta-Analysis. PLoS Med. 11, (2014).
51. Strunz, E. C. et al. Water, Sanitation, Hygiene, and Soil-Transmitted Helminth Infection: A Systematic Review and Meta-Analysis. PLoS Med. 11, (2014).
127
52. Penakalapati, G. et al. Exposure to Animal Feces and Human Health: A Systematic Review and Proposed Research Priorities. Environ. Sci. Technol. 51, 11537–11552 (2017).
53. Curtis, V. & Cairncross, S. Effect of washing hands with soap on diarrhoea risk in the community: A systematic review. Lancet Infectious Diseases (2003). doi:10.1016/S1473-3099(03)00606-6
54. Zambrano, L. D., Levy, K., Menezes, N. P. & Freeman, M. C. Human diarrhea infections associated with domestic animal husbandry: A systematic review and meta-analysis. Transactions of the Royal Society of Tropical Medicine and Hygiene 108, 313–325 (2014).
55. Luby, S. P. et al. Effects of water quality, sanitation, handwashing, and nutritional interventions on diarrhoea and child growth in rural Bangladesh: A cluster randomised controlled trial. The Lancet Global Health (2018). doi:10.1016/S2214-109X(17)30490-4
56. Null, C. et al. Effects of water quality, sanitation, handwashing, and nutritional interventions on diarrhoea and child growth in rural Kenya: A cluster-randomised controlled trial. The Lancet Global Health (2018). doi:10.1016/S2214-109X(18)30005-6
57. Arnold, B. F., Null, C., Luby, S. P. & Colford, J. M. Implications of WASH Benefits trials for water and sanitation – Authors’ reply. The Lancet Global Health (2018). doi:10.1016/S2214-109X(18)30229-8
58. Mbuya, M. N. N. & Humphrey, J. H. Preventing environmental enteric dysfunction through improved water, sanitation and hygiene: An opportunity for stunting reduction in developing countries. Maternal and Child Nutrition (2016). doi:10.1111/mcn.12220
59. Workicho, A. et al. Household dietary diversity and Animal Source Food consumption in Ethiopia: evidence from the 2011 Welfare Monitoring Survey. BMC Public Health 16, 1192 (2016).
60. Parappurathu, S., Kumar, A., Bantilan, M. C. S. & Joshi, P. K. Food consumption patterns and dietary diversity in eastern India: evidence from village level studies (VLS). Food Secur. (2015). doi:10.1007/s12571-015-0493-2
61. Motbainor, A., Worku, A. & Kumie, A. Stunting is associated with food diversity while wasting with food insecurity among underfive children in East and West Gojjam Zones of Amhara Region, Ethiopia. PLoS One (2015). doi:10.1371/journal.pone.0133542
62. Mahmudiono, T., Sumarmi, S. & Rosenkranz, R. R. Household dietary diversity and child stunting in East Java, Indonesia. Asia Pac. J. Clin. Nutr. (2017). doi:10.6133/apjcn.012016.01
128
63. Neumann, C. G. et al. Meat supplementation increases arm muscle area in Kenyan schoolchildren. Br. J. Nutr. (2013). doi:10.1017/S0007114512003121
64. The World Bank. Investing in people to fight poverty in Haiti : Reflections for evidence-based policy making (English). (2014).
65. Black, R. E. et al. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 382, 427–451 (2013).
66. ICF International. The DHS Program: Demographic and Health Surveys - Stata Compiler. DHS Model Questionnaires (2016).
67. Pauzé, E., Batal, M., Philizaire, Y., Blanchet, R. & Sanou, D. Determinants of diet quality among rural households in an intervention zone of Grande Anse , Haiti. Food Secur. 1123–1134 (2016). doi:10.1007/s12571-016-0615-5
68. Spray, A. L., Eddy, B., Hipp, J. A. & Iannotti, L. Spatial analysis of undernutrition of children in léogâne commune, haiti. Food Nutr. Bull. 34, 444–461 (2013).
69. Haggerty, R. A. Haiti: A Country Study. Washington: GPO for the Library of Congress (1989).
70. The World Bank. Haiti: Overview. (2018). Available at: http://www.worldbank.org/en/country/haiti/overview.
71. The World Bank. Promoting Nutrition Security in Haiti: An Assessment of Pre- and Post-Earthquake Conditions for the Way Forward. Nutrition 74 (2010).
72. UN radio & ReliefWeb. Poverty is behind Haiti’s vulnerability to natural disasters.
73. WFP. Haiti transitional interim country strategic plan (year 2018). (2018).
74. UNDP. Human Development Index (HDI). Human Development Reports (2018). Available at: http://hdr.undp.org/en/content/human-development-index-hdi.
75. Oxfam International. Planting Now: Agricultural challenges and opportunities for Haiti’s reconstruction. (2010).
76. Lawless, R. & Ferguson, J. A. Haiti. Encyclopædia Britannica (2018).
77. Baro, M. Food Insecurity and Livelihood Systems in Northwest Haiti. 9, 1–34 (2002).
78. Heifer International. Heifer Haiti. Available at: https://www.heifer.org/ending-hunger/our-work/countries/north-america/haiti.html. (Accessed: 31st July 2018)
79. Gelting, R., Bliss, K., Patrick, M., Lockhart, G. & Handzel, T. Water , Sanitation and Hygiene in Haiti : Past , Present , and Future. 89, 665–670 (2013).
129
80. USAID. Water, Sanitation, and Hygiene Sector Status and Trends Assessment in Haiti. USAID, Dec. 2014 (2014).
81. WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation. Progress on Drinking Water and Sanitation. Update (2012). doi:978-924-1503279
82. Ministry of public health and population: national directorate for water supply and sanitation. National Plan for the Elimination of Cholera in Haiti: 2013-2022.
83. Bargout, R. N. & Raizada, M. N. Soil nutrient management in Haiti, pre-Columbus to the present day: Lessons for future agricultural interventions. Agriculture and Food Security (2013). doi:10.1186/2048-7010-2-11
84. Cayemittes, M. et al. Survey on mortality, morbidity and use of services EMMUS-V: Haiti 2012. Enquete mortalite, morbidite et utilisation des services EMMUS-V: Haiti 2012. (2013).
85. Ayoya, M. A., Heidkamp, R., Ngnie-Teta, I., Pierre, J. M. & Stoltzfus, R. J. Child malnutrition in Haiti: progress despite disasters. Glob. Heal. Sci. Pract. (2013). doi:10.9745/GHSP-D-13-00069
86. UNICEF. Improving child nutrition: The achievable imperative for global progress. Div. Commun. UNICEF (2013). doi:978-92-806-4686-3
87. Schönfeldt, H. C., Pretorius, B. & Hall, N. The impact of animal source food products on human nutrition and health.: Find articles, e-books in one search. S. Afr. J. Anim. Sci. (2013). doi:10.4314/sajas.v43i3.11
88. UNICEF. Infant and young child feeding. UNICEF data (2018). Available at: https://data.unicef.org/topic/nutrition/infant-and-young-child-feeding/.
89. Crane, R. J., Jones, K. D. J. & Berkley, J. A. Environmental enteric dysfunction: An overview. Food Nutr. Bull. 36, S76–S87 (2015).
90. Ercumen, A. et al. Animal Feces Contribute to Domestic Fecal Contamination: Evidence from E. coli Measured in Water, Hands, Food, Flies, and Soil in Bangladesh. Environ. Sci. Technol. 51, 8725–8734 (2017).
91. Barnes, A. N., Davaasuren, A., Baasandagva, U. & Gray, G. C. A systematic review of zoonotic enteric parasitic diseases among nomadic and pastoral people. PLoS ONE (2017). doi:10.1371/journal.pone.0188809
92. Barnes, A. N. Domestic animals and drinking water contamination in peri-urban households of kisumu, kenya. (University of Florida, 2016).
93. Greger, M. The human/animal interface: Emergence and resurgence of zoonotic infectious diseases. Critical Reviews in Microbiology 33, 243–299 (2007).
130
94. Engering, A., Hogerwerf, L. & Slingenbergh, J. Pathogen–host–environment interplay and disease emergence. Emerg. Microbes Infect. 2, e5 (2013).
95. Lindahl, J. & Grace, D. The consequences of human actions on risks for infectious diseases: a review. Infect. Ecol. Epidemiol. 5, 11 (2015).
96. Wang, H. et al. Global, regional, and national life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death, 1980-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet 388, 1459–1544 (2016).
97. Block, S. A. Maternal nutrition knowledge versus schooling as determinants of child micronutrient status. Oxf. Econ. Pap. (2007). doi:10.1093/oep/gpm001
98. George, C. M. et al. Risk factors for diarrhea in children under five years of age residing in Peri-urban Communities in Cochabamba, Bolivia. Am. J. Trop. Med. Hyg. (2014). doi:10.4269/ajtmh.14-0057
99. Dikassa, L. et al. Maternal behavioural risk factors for severe childhood diarrhoeal disease in kinshasa, zaire. Int. J. Epidemiol. (1993). doi:10.1093/ije/22.2.327
100. Bilenko, N., Fraser, D. & Naggan, L. Maternal knowledge and environmental factors associated with risk of diarrhea in Israeli Bedouin children. Eur. J. Epidemiol. (1999). doi:10.1023/A:1007649820423
101. Vikram, K., Vanneman, R. & Desai, S. Linkages between maternal education and childhood immunization in India. Soc. Sci. Med. (2012). doi:10.1016/j.socscimed.2012.02.043
102. Ayoya, Mohamed Ag; Ngnie- Teta, Ismael; Seraphin Marie Nancy; Mamadoultaibou, Aissa; Boldon, Ellen; Saint- Fleur, Jean Ernest; Koo, Leslie; Bernard, S. Prevalence and Risk Factors of Anemia Among Children 6- 59 Mpnths Old in Haiti. Hindawi Publ. (2013). doi:10.1155/2013/502968
103. Seraphin, M. N. et al. Determinants of Institutional Delivery Among Women of Childbearing Age in Rural Haiti. Matern. Child Health J. (2015). doi:10.1007/s10995-014-1646-1
104. Seraphin, M. N. et al. Childhood anemia in Rural Haiti: the potential role of community health workers. Glob. Heal. Res. policy (2017). doi:10.1186/s41256-016-0022-7
105. SAS Institute. SAS version 9.4. SAS Institute Inc. (2012).
106. SAS Institute Inc. Base SAS® 9.4 Procedures Guide. Statistical Procedures (2013). doi:10.1016/B978-0-444-59425-9.00014-7
131
107. SAS. The MI procedure. in SAS/STAT 13.1 User’s Guide (2013).
108. Schunk, D. A Markov chain Monte Carlo algorithm for multiple imputation in large surveys. AStA Adv. Stat. Anal. (2008). doi:10.1007/s10182-008-0053-6
109. Zhang, P. Multiple imputation: Theory and method. Int. Stat. Rev. (2003). doi:10.1111/j.1751-5823.2003.tb00213.x
110. Little, R. J. A. A test of missing completely at random for multivariate data with missing values. J. Am. Stat. Assoc. (1988). doi:10.1080/01621459.1988.10478722
111. UCLA: Statistical Consulting Group. Introduction to SAS. Available at: https://stats.idre.ucla.edu/sas/modules/sas-learning-moduleintroduction-to-the-features-of-sas/. (Accessed: 29th October 2018)
112. UCLA: Statistical Consulting Group. Introduction to SAS.
113. Ruel, M. T. Animal Source Foods to Improve Micronutrient Nutrition and Human Function in Developing Countries Operationalizing Dietary Diversity : A Review of Measurement Issues. J. Nutr. (2003).
114. Mukherjee, A., Paul, S., Saha, I., Som, T. & Ghose, G. Dietary diversity and its determinants: A community-based study among adult population of Durgapur, West Bengal. Med. J. Dr. D.Y. Patil Vidyapeeth 11, 296–301 (2018).
115. RStudio Team, -. RStudio: Integrated Development for R. [Online] RStudio, Inc., Boston, MA URL http//www. rstudio. com (2015). doi:10.1126/science.aad6351
116. Mosites, E. M. The influence of livestock ownership and health on the nutritional status of children in Eastern Africa. ProQuest Diss. Theses 69 (2015). doi:10.1017/CBO9781107415324.004
117. US Department of Agriculture. The Caribbean Import Market for Poultry.
118. Pachón, H. et al. Constraints on the delivery of animal-source foods to infants and young children: Case studies from five countries. Food Nutr. Bull. (2007). doi:10.1177/156482650702800211
119. Murray, G. The Haitian Pig Slaughter.
120. Bukania, Z. N. et al. Food insecurity and not dietary diversity is a predictor of nutrition status in children within semiarid agro-ecological zones in Eastern Kenya. J. Nutr. Metab. (2014). doi:10.1155/2014/907153
121. Swindale, A. & Bilinsky, P. Household Dietary Diversity Score (HDDS) for measurement of household food access: Indicator guide. Food Nutr. Tech. Assist. … (2006). doi:10.1017/CBO9781107415324.004
132
122. Ruel, M. T. & Alderman, H. Nutrition-sensitive interventions and programmes: How can they help to accelerate progress in improving maternal and child nutrition? Lancet (2013). doi:10.1016/S0140-6736(13)60843-0
123. McDonald, C. M. et al. Household food insecurity and dietary diversity as correlates of maternal and child undernutrition in rural Cambodia. Eur. J. Clin. Nutr. (2014). doi:10.1038/ejcn.2014.161
124. Njuki, J. & Sanginga, P. C. Women, livestock ownership and markets: Bridging the gender gap in eastern and Southern Africa. Women, Livestock Ownership and Markets: Bridging the gender gap in Eastern and Southern Africa (2013). doi:10.4324/9780203083604
125. Carlson, G. J., Kordas, K. & Murray-Kolb, L. E. Associations between women’s autonomy and child nutritional status: A review of the literature. Maternal and Child Nutrition (2015). doi:10.1111/mcn.12113
126. Cunningham, K. et al. Women’s empowerment in agriculture and child nutritional status in rural Nepal. Public Health Nutr. (2015). doi:10.1017/S1368980015000683
127. Cunningham, K., Ruel, M., Ferguson, E. & Uauy, R. Women’s empowerment and child nutritional status in South Asia: a synthesis of the literature. Matern. Child Nutr. (2014). doi:10.1111/mcn.12125
128. Gardella, A. Gender Assessment for USAID/Haiti Country Strategy Statement. (2006).
129. Savy, M., Martin-Prével, Y., Sawadogo, P., Kameli, Y. & Delpeuch, F. Use of variety/diversity scores for diet quality measurement: Relation with nutritional status of women in a rural area in Burkina Faso. Eur. J. Clin. Nutr. (2005). doi:10.1038/sj.ejcn.1602135
130. Appoh, L. Y. & Krekling, S. Maternal nutritional knowledge and child nutritional status in the Volta Region of Ghana. Matern. Child Nutr. (2005). doi:10.1111/j.1740-8709.2005.00016.x
131. Smith, L. C. & Haddad, L. Reducing Child Undernutrition: Past Drivers and Priorities for the Post-MDG Era. World Dev. (2015). doi:10.1016/j.worlddev.2014.11.014
132. Oduor, F. O., Boedecker, J., Kennedy, G., Mituki-Mungiria, D. & Termote, C. Caregivers’ nutritional knowledge and attitudes mediate seasonal shifts in children’s diets. Maternal and Child Nutrition (2018). doi:10.1111/mcn.12633
133. Frongillo, E. A., de Onis, M. & Hanson, K. M. Socioeconomic and demographic factors are associated with worldwide patterns of stunting and wasting of children. J. Nutr. (1997). doi:10.1093/jn/127.12.2302
133
134. Daumerie, B. & Madsen, E. The effects of a very young age structure in Uganda. Country case study. The Shape of things to Come Series (2010).
135. Aurino, E. Do boys eat better than girls in India? Longitudinal evidence on dietary diversity and food consumption disparities among children and adolescents. Econ. Hum. Biol. (2017). doi:10.1016/j.ehb.2016.10.007
136. Herrador, Z. et al. Cross-Sectional Study of Malnutrition and Associated Factors among School Aged Children in Rural and Urban Settings of Fogera and Libo Kemkem Districts, Ethiopia. PLoS One 9, e105880 (2014).
137. FAO. Guidelines for measuring household and individual dietary diversity. Fao (2010). doi:613.2KEN
138. WHO and UNICEF. WHO child growth standards and the identification of severe acute malnutrition in infants and children: A Joint Statement by the World Health Organization and the United Nations Children’s Fund. 1–12 (2009).
139. de Onis, M. & Branca, F. Childhood stunting: A global perspective. Maternal and Child Nutrition (2016). doi:10.1111/mcn.12231
140. Unicef/ & WHO/Wolrd Bank. Levels and Trends in Child malnutrition: Key findings of the 2016 edition. Midwifery (2016). doi:10.1016/S0266-6138(96)90067-4
141. Ayoya, M. A. et al. Précis of nutrition of children and women in Haiti: Analyses of data from 1995 to 2012. Ann. N. Y. Acad. Sci. (2014). doi:10.1111/nyas.12373
142. Stewart, C. P., Iannotti, L., Dewey, K. G., Michaelsen, K. F. & Onyango, A. W. Contextualising complementary feeding in a broader framework for stunting prevention. Matern. Child Nutr. (2013). doi:10.1111/mcn.12088
143. Caulfield, L. E., Richard, S. a, Rivera, J. a, Musgrove, P. & Black, R. E. Stunting , Wasting , and Micronutrient Deficiency Disorders. in Disease Control Priorities in Developing Countries 551–567 (2006). doi:10.1093/ije/dyq085
144. Ngure, F. M. et al. Formative research on hygiene behaviors and geophagy among infants and young children and implications of exposure to fecal bacteria. Am. J. Trop. Med. Hyg. 89, 709–716 (2013).
145. George, C. M. et al. Unsafe Child Feces Disposal is Associated with Environmental Enteropathy and Impaired Growth. J. Pediatr. (2016). doi:10.1016/j.jpeds.2016.05.035
146. Cumming, O. & Cairncross, S. Can water, sanitation and hygiene help eliminate stunting? Current evidence and policy implications. Maternal and Child Nutrition (2016). doi:10.1111/mcn.12258
134
147. Baker, J., Sanghvi, T., Hajeebhoy, N., Martin, L. & Lapping, K. Using an evidence-based approach to design large- scale programs to improve infant and young child feeding. Food Nutr. Bull. (2013). doi:10.1177/15648265130343S202
148. Iannotti, L. L., Lutter, C. K., Bunn, D. A. & Stewart, C. P. Eggs: The uncracked potential for improving maternal and young child nutrition among the world’s poor. Nutr. Rev. 72, 355–368 (2014).
149. WHO and UNICEF. Progress on sanitation and drinking-water - 2014 update. … Monitoring Programme for water supply and sanitation. (2014). doi:http://apps.who.int/iris/bitstream/10665/81245/1/9789241505390_eng.pdf?ua=1
150. WHO. Progress on Drinking Water, Sanitation and Hygiene. World Health Organization (2017). doi:10.1111/1463-1652.00063
151. Black, R. E., Morris, S. S. & Bryce, J. Where and why are 10 million children dying every year? Lancet 361, 2226–2234 (2003).
152. Vella, V. et al. Determinants of child nutrition and mortality in north-west Uganda. Bull. World Health Organ. (1992).
153. Arthur, S. S. et al. Tackling malnutrition: A systematic review of 15-year research evidence from INDEPTH health and demographic surveillance systems. Glob. Health Action (2015). doi:10.3402/gha.v8.28298
154. Bawankule, R., Singh, A., Kumar, K. & Pedgaonkar, S. Disposal of children’s stools and its association with childhood diarrhea in India. BMC Public Health (2017). doi:10.1186/s12889-016-3948-2
155. Nizame, F. A. et al. Handwashing before food preparation and child feeding: A missed opportunity for hygiene promotion. Am. J. Trop. Med. Hyg. (2013). doi:10.4269/ajtmh.13-0434
156. Black, R. E. et al. Maternal and child undernutrition: global and regional exposures and health consequences. The Lancet 371, 243–260 (2008).
157. L’Enfance, I. H. de, Petion-Ville, DHS Haiti & MEASURE & ICF International. Haiti Enquete Mortalite, Morbidite, et Utilization de Services (EMMUS-V). (2012).
158. Piper, J. D. et al. Water , sanitation and hygiene ( WASH ) interventions : effects on child development in low- and middle-income countries ( Protocol ). Cochrane Database Syst Rev (2017). doi:10.1002/14651858.CD012613.www.cochranelibrary.com
159. Hagos, S., Lunde, T., Mariam, D. H., Woldehanna, T. & Lindtjørn, B. Climate change, crop production and child under nutrition in Ethiopia; a longitudinal panel study. BMC Public Health 14, 884 (2014).
135
160. Gebreyesus, S. H., Mariam, D. H., Woldehanna, T. & Lindtjørn, B. Local spatial clustering of stunting and wasting among children under the age of 5 years: Implications for intervention strategies. Public Health Nutr. (2016). doi:10.1017/S1368980015003377
161. Haile, D., Azage, M., Mola, T. & Rainey, R. Exploring spatial variations and factors associated with childhood stunting in Ethiopia: Spatial and multilevel analysis. BMC Pediatr. (2016). doi:10.1186/s12887-016-0587-9
162. Kinyoki, D. K. et al. Environmental predictors of stunting among children under-five in Somalia: Cross-sectional studies from 2007 to 2010. BMC Public Health (2016). doi:10.1186/s12889-016-3320-6
163. Grace, K., Davenport, F., Funk, C. & Lerner, A. M. Child malnutrition and climate in Sub-Saharan Africa: An analysis of recent trends in Kenya. Appl. Geogr. (2012). doi:10.1016/j.apgeog.2012.06.017
164. Martínez, J. I., Román, E. M., Alfaro, E. L., Grandi, C. & Dipierri, J. E. Geographic altitude and prevalence of underweight, stunting and wasting in newborns with the INTERGROWTH-21st standard. Jornal de Pediatria (2018). doi:10.1016/j.jped.2018.03.007
165. Shively, G., Sununtnasuk, C. & Brown, M. Environmental variability and child growth in Nepal. Heal. Place (2015). doi:10.1016/j.healthplace.2015.06.008
166. Fick, S. E. & Hijmans, R. J. WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. (2017). doi:10.1002/joc.5086
167. Funk, C. et al. A global satellite-assisted precipitation climatology. Earth Syst. Sci. Data (2015). doi:10.5194/essd-7-275-2015
168. Didan, K. MOD13A1 MODIS/Terra Vegetation Indices 16-Day L3 Global 500m SIN Grid V006 [Data set]. NASA EOSDIS LP DAAC (2015).
169. Jarvis, A., H.I., Reuter, A., Nelson, A. & Guevara, E. Hole-filled SRTM for the globe Version 4, available from the CGIAR-CSI SRTM 90m Database. CGIAR CSI Consort. Spat. Inf. (2016). doi:10.1016/S0550-3213(96)00589-5
170. Weiss, D. J. et al. A global map of travel time to cities to assess inequalities in accessibility in 2015. Nature (2018). doi:10.1038/nature25181
171. Stevens, F. R. et al. WorldPop-RF, Version 2b.1.1. figshare (2015). doi:doi:10.6084/m9.figshare.1491490
172. Bartholomé, E., Belward, A. S. & F. Achard. Global Land Cover 2000 database. European Commission,Joint Research Centre (2003).
173. GADM. Database of Global Administrative Areas, version 3.6. Univ. Calif. (2018).
136
174. Noor, A. M. et al. The changing risk of Plasmodium falciparum malaria infection in Africa: 2000-10: A spatial and temporal analysis of transmission intensity. Lancet (2014). doi:10.1016/S0140-6736(13)62566-0
175. Sullivan, D. Google Earth Pro. Geospatial Solutions (2009). doi:10.1056/NEJMoa1509342
176. Gorelick, N. et al. Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sens. Environ. (2017). doi:10.1016/j.rse.2017.06.031
177. ESRI Environmental Systems Research Institute. ArcGIS Desktop: Release 10.2.2. Redlands CA (2014).
178. Rinaldo, A. et al. Reassessment of the 2010–2011 Haiti cholera outbreak and rainfall-driven multiseason projections. Proc. Natl. Acad. Sci. 109, 6602 LP-6607 (2012).
179. Eisenberg, M. C., Kujbida, G., Tuite, A. R., Fisman, D. N. & Tien, J. H. Examining rainfall and cholera dynamics in Haiti using statistical and dynamic modeling approaches. Epidemics (2013). doi:10.1016/j.epidem.2013.09.004
180. Wan, Z., Hook, S. & Hulley, G. MOD11A2 MODIS/Terra Land Surface Temperature/Emissivity 8-Day L3 Global 1km SIN Grid V006 [Data set]. NASA EOSDIS LP DAAC (2015).
181. Center for International Earth Science Information Network - CIESIN. Global Roads Open Access Data Set, Version 1 (gROADSv1). NASA Socioecon. Data Appl. Cent. (2013). doi:http://dx.doi.org/10.7927/H4VD6WCT
137
BIOGRAPHICAL SKETCH
Lindsey Amanda Laytner was born and raised in Fort Lauderdale, FL. She went
on to pursue her Bachelor of Art (BA) in Anthropology, a Master in Public Health (MPH)
in Social and Behavioral Sciences, and a Doctor of Philosophy in Environmental and
Global Public Health from the University of Florida. Throughout her academic career,
Lindsey has been fortunate to work in East Africa and the Caribbean. She spent 3
months in the highlands of Ethiopia as an archeologist, and on a multi-sectoral WASH
campaign in Kisumu, Kenya where she trained enumerators, field assistants, and
oversaw the administration of an 800-household survey that involved the University of
Florida, Great Lakes University in Kisumu, CDC-Kemri, and the London School of
Hygiene and Tropical medicine based in Kisumu, Kenya. During her doctoral studies,
she has worked on a variety of projects concerning WASH, livestock husbandry and
animal source food consumption, child health outcomes (i.e. particularly, stunting). Her
work has included consulting for PATH, the Bill and Melinda Gates Foundation, USAID,
UNICEF, and the World Bank. Her ongoing desire is to continue her work in the WASH-
One Health research and programming arena, using community outreach and
engagement principles to design impactful WASH interventions and communication
tools that focus on human, animal, and environmental health through the WASH