Child stunting, or linear growth faltering, affects 1 in 5 children under the age of five years. These 151 million stunted children predominantly live in low- and middle-income countries. For decades, clinicians and researchers considered stunting a form of chronic malnutrition but modest gains were made in eliminating stunting through nutritional interventions. This dissertation explored emerging areas in child stunting. Chapters 1 and 2 introduced stunting and provided backgrounds, respectively. The research was guided by the Social Ecological Model and Life Course Perspective in approaching stunting as embedded in macro structures at multiple levels over the life course of women and children. Chapter 3 described theoretical frameworks and an integrated model. The analyses focused on Madagascar and Zambia, which were ranked among countries with the highest proportions of child stunting. We used health data from the Demographic and Health Surveys (DHS) and armed conflict data from the Armed Conflict Location and Event Data Project (ACLED) database in Madagascar and Zambia. Chapter 4 detailed the methodology, including multivariate logistic and linear regression models assessing child stunting and height-for-age z-score (HAZ) outcomes. Chapter 5 examined maternal anthropometry and child gender factors. Results indicated that short stature or underweight in mothers were associated with increased stunting odds while higher maternal height and BMI scores were associated with higher child HAZ. Chapter 6 investigated household water and sanitation measures. We found that households without piped water and finished flooring were associated with increased stunting odds. Stunting was associated with lack of an advanced flush toilet in Zambia but not in Madagascar. Chapter 7 explored proximity of armed conflict events during critical developmental periods with child stunting and height. Conflict exposure during pregnancy was associated with increased stunting odds and lower HAZ in Madagascar but decreased odds and higher HAZ in Zambia. This dissertation framed stunting as occurring over the life course and embedded in multiple external structures. These studies were among the first to examine population-level environmental enteric dysfunction risk factors and apply disaggregated conflict data to stunting. We also contributed stunting context in Madagascar and Zambia, which have been understudied.

Population conflicts and tropical cyclones represent significant public health threats, particularly in low and middle-income countries, disrupting societies and profoundly impacting child health outcomes. Birthweight and stunting are key measures for assessing impacts of environmental, nutritional, and social stressors during early development. Although existing research suggests that conflicts and cyclones adversely affect birthweight and childhood stunting, most studies focus on single countries or events. Large-scale studies across multiple countries and extended periods are scarce. This dissertation combines geo-referenced data from various independent sources to examine the impacts of conflicts and cyclones on birthweight and stunting across multiple countries over time. Causal inference methods were used to evaluate potential interventions to mitigate these adversities.First, the impact of population conflicts on birthweight and stunting was investigated in Sub-Saharan African countries, using data from the 1990-2020 DHS and UCDP GED. Fixed effect regression models compared children in conflict-affected regions to those in unaffected areas, considering conflict intensity and duration. Results showed reduced birthweight and increased stunting odds for children exposed to conflicts, particularly girls in rural areas with less educated mothers and no healthcare autonomy. Second, the impact of tropical cyclones on birthweight and stunting was examined in Southern and Southeastern Asia using data from the 2000-2018 DHS and EM-DAT. Similar regression models were used. In-utero exposure to cyclones in the Philippines reduced birthweight, especially for children in rural areas. In Bangladesh and India, cyclones increased stunting odds, with more pronounced effects in girls and rural children with less educated mothers. No interactions between cyclones and conflicts were found. Lastly, hypothetical interventions were estimated using causal inference methods. Results indicated that preventing exposure to conflicts and cyclones could significantly increase birthweight and reduce stunting prevalence. In summary, these studies provide evidence of the adverse impacts of population conflicts and tropical cyclones on child growth and development, highlighting the need for effective prevention and adaptation strategies in low and middle-income countries. Addressing these issues aligns with the SDGs and offers insights for global public health interventions to mitigate environmental and social disruptions for vulnerable children.

Background

Methods

Results

Conclusions

Objectives

Methods

Results

Conclusions

Objectives

Methods

Results

Conclusions

Purpose

Design

Methods

Results

Conclusions

Objective

Methods

Results

Conclusion