Lost Lives and Malnutrition: Examining the Contribution of Anthropometric Malnutrition and Vitamin A Deficiency on Under-5-Mortality in Sub-Saharan Africa and South Asia Open Access

Das, Priya (Spring 2024)

Permanent URL: https://etd.library.emory.edu/concern/etds/bz60cx71b?locale=en
Published

Abstract

Our first publication, Impact of anthropometry training and feasibility of 3D imaging on anthropometry data quality among children under five years in a postmortem setting evaluated the impact of anthropometry training and the feasibility of 3D imaging on data quality, precision, and accuracy within the CHAMPS Kenya site. We found that manual anthropometry was feasible and reliable in the postmortem setting even in the presence of rigor mortis. 3D imaging software, currently, is not an accurate alternative to manual anthropometry. Further technology adjustments are needed to ensure future accuracy and usability.

Our second publication, currently under internal review, Degree of Postmortem Anthropometric Malnutrition and Malnutrition Attributed as a Cause of Death among Infants and Child Deaths in Sub-Saharan Africa and South Asia sought to examine the alignment between postmortem anthropometric malnutrition (PAM) and expert panel attribution, exploring variations by PAM type and severity. We found that nearly 3 in 4 of cases showed moderate or severe malnutrition, yet malnutrition was attributed by expert panels in only 4 in 10 cases. Performance metrics varied by anthropometric indices; underweight exhibited the highest sensitivity (89.7%), while wasting based on arm circumference showed the highest specificity (81.9%) and positive predictive value (76.8%). Discrepancies between classification methods varied by site, age, location, and preventability of death. Adjusted multivariate regression analysis revealed higher odds of expert panel attribution with increasing PAM severity, particularly for underweight and wasting as compared to stunting. Our findings suggest that 40% to 75% of under-5-mortality may be attributable to anthropometric malnutrition.

 

Our third publication, Hepatic Vitamin A Concentrations and Association with Infectious Causes of Child Death sought to examine one form of micronutrient malnutrition, Vitamin A. This final aim assesses postmortem vitamin A (VA) concentrations in children under 5 years of age and evaluate the association between vitamin A deficiency (VAD) and infectious causes of death (CoD). Liver biopsies collected within 72 hours of death were analyzed from children under 5 years in Kenya and South Africa. Pooled prevalence of VAD, adequacy, high VA, and hypervitaminosis was 34.2%, 51.1%, 6.0%, and 8.7%, respectively. VAD was more common among neonates compared with stillbirths, infants, or children, and among those with low birthweight, underweight or stunting (p<0.05). When controlling for site, age, and sex, logistic regression models revealed no significant association of VAD and increased infectious CoD (OR 1.9, 95%CI 0.9, 3.8, p=0.073). However, our stratified analyses revealed VA deficient boys, but not girls, had an increased risk of infectious CoD (OR 3.4, 95%CI 1.3, 10.3, p=0.013). Definitive  postmortem assessment of VA status identified both VAD and VA excess among children under 5 years of age in Kenya and South Africa. VAD in boys was associated with increased risk of infectious mortality, which may inform a transition from universal VA supplementation to targeted strategies in certain countries. 

 

The purpose of this dissertation project was to elucidate the proportion of under-5-mortality attributable to malnutrition, including deficits in anthropometric malnutrition and examination of one form of micronutrient malnutrition. We used data from a surveillance system that captures multiple causes of death with direct anthropometric and micronutrient measures. However, future estimations of the contribution of under-5- mortality attributable to malnutrition should encompass fetal growth restriction, stunting, wasting, and micronutrient deficiencies, alongside dietary data, to better quantify suboptimal breastfeeding and inform intervention strategies.

Table of Contents

TABLE OF CONTENTS

Acknowledgements

Abbreviations

List of Tables

Table 3.5.1

Table 3.5.2

Table 3.5.3

Table 3.5.4

Table 4.6.1

Table 4.6.2

Table 4.6.3

Table 4.6.4

List of Figures

Figure 3.6.1

Figure 4.6.1

Figure 5.6.1

Figure 5.6.2

RESEARCH IN CONTEXT

Evidence Before This Research

Added Value Of This Research

Implications Of All Available Research

CHAPTER 1: INTRODUCTION

References Chapter 1

CHAPTER 2: OVERVIEW OF DATA SOURCE AND METHODS

2.1: Overview of Study Design and Sampling

2.2: Consent for Enrollment and Participation

2.3: Eligibility Criteria

2.5: Postmortem External examination

2.6: Minimally Invasive Tissue Samples (MITS)

2.7: Maternal and Child Clinical Abstraction

2.8: Determination of Cause of Death (DeCoDe)

2.9: Data Analyses

References Chapter 2

CHAPTER 3: FEASIBILITY AND ACCURACY OF MANUAL ANTHROPOMETRY IN THE POSTMORTEM SETTING

3.1: Introduction

3.2: Methods

3.3: Results

3.3.1: Quality- Digit Preference

3.3.2: Quality- Means and Standard Deviations of Anthropometric Indices

3.3.3: Precision-Technical Errors of Measurement

3.3.4: Accuracy- Spearman Correlation and Bland Altman Plots

3.3.5: Qualitative Findings: Use of 3D Imaging in Morgue Setting

3.4: Discussion

3.5: Tables

Table 3.5.1.

Table 3.5.2

Table 3.5.3

Table 3.5.4

3.6: Figures

Figure 3.6.1

Chapter 3 References

CHAPTER 4: EXAMINING DIFFERENCES BETWEEN POSTMORTEM ANTHROPOMETRIC MALNUTRITION AND MALNUTRITION ATTRIBUTED AS A CAUSE OF INFANT AND CHILD DEATHS IN SUB-SAHARAN AFRICA AND SOUTH ASIA

4.1: Introduction

4.2: Methods

4.2.1: Overview of Data Source

4.2.2: Exposures: Postmortem Anthropometric Malnutrition

4.2.3: Outcome: Determination of Cause of Death Panel Attributing Malnutrition as a Causal or Other Significant Condition to Death

4.2.4: Inclusion Criteria

4.2.5: Sensitivity Analyses

4.2.6: Statistical analyses

4.2.7 Contingency Tables and Evaluating Performance Metrics

4.2.8: Covariates and Modeling

4.3: Results

4.3.1: Demographic characteristics of analytic sample

4.3.2: Data Missingness

4.3.3: Contingency Tables and Evaluating Performance Metrics

4.3.4: Factors associated with Positive Concordance between Any Postmortem Malnutrition and DeCoDe attributing Malnutrition as a Causal or Significant Condition to Death

4.3.5: Ordinal Logistic Regression

4.4: Discussion

4.5: Conclusions

4.6: Tables

Table 4.6.1

Table 4.6.2

Table 4.6.3

Supplemental Table 4.6.4

4.6: Figures

Figure 4.6.1:

Chapter 4 References

CHAPTER 5: HEPATIC VITAMIN A CONCENTRATIONS AND ASSOCIATION WITH INFECTIOUS CAUSES OF CHILD DEATH

5.1: Introduction

5.2: Methods

5.2.1: Data Source

5.2.2: Sample Selection and Exclusions

5.2.3: Exposure: Liver Vitamin A Concentrations

5.2.4: Outcome: Infectious Causes of Death

5.2.5: Statistical Analyses

5.2.5: Covariates Assessed for Confounding and Effect Modification

5.2.5: Ordinal Logistic Regression

5.3: Results

5.3.1: Demographic characteristics of analytic sample

5.3.2: Vitamin A Status

5.3.2: Vitamin A Deficiency and the association with Infectious Mortality

5.4: Discussion

5.5: Tables

Table 5.5.1

Table 5.5.2

Table 5.5.3

5.6: Figures

Figure 5.6.1.

Figure 5.6.2.

Chapter 5 References

CHAPTER 6: DISCUSSION SUMMARY OF FINDINGS

Aim 1

Aim 2

Aim 3

COMPARISON OF FINDINGS TO SIMILAR STUDIES

STUDY STRENGTHS AND LIMITATIONS

Remaining Research Gaps and Directions for Future Research

Recommendations for CHAMPS

Recommendations for Future Research

Recommendations for Programs

Recommendations for Ministries of Health

Modifying the CHAMPS study design to move beyond “case series” and to better assess causality of malnutrition and child mortality?

CONCLUSIONS

Chapter 6 References

About this Dissertation

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
School
Department
Degree
Submission
Language
  • English
Research Field
Keyword
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files