Dissertations, Theses, and Capstone Projects

Date of Degree


Document Type


Degree Name



Public Health


Deborah Balk


Glen Johnson

Committee Members

Glen Johnson

Grace Sembajwe

Emmanuel d'Harcourt

Subject Categories

Environmental Public Health | International Public Health | Maternal and Child Health


diarrhea, malnutrition, climate variability, armed conflict, West Africa


Objectives: This dissertation aims to contribute to our understanding of how climate variability and armed conflict impacts diarrheal disease and malnutrition among young children in West Africa. Two studies examine the associations between climate and diarrheal disease across the whole study area – ten countries in West Africa during the period 2008-2013. The third study examines diarrheal disease and malnutrition in Northeast Nigeria before (2008) and after (2013) the start of the current armed conflict in the area.

Methods: Outcome variables and child, caregiver, and household characteristics for these studies are from the Demographic and Health Surveys. Additional datasets include specialized products for rainfall, temperature, climate class, population density, and urban population ranking. The first two studies use geo-spatial techniques to 1) determine if there are areas of elevated risk after controlling for climatic and other covariates and 2) determine whether there is spatial variation in the associations between diarrheal disease and climatic and other covariates. The third study uses a double-difference methodology to study both diarrheal disease and malnutrition and their associations with conflict.

Results: The first study found statistically significant clusters of elevated risk in 10 urban areas and 13 largely rural areas after adjusting for household and climatic factors with a relative risk range of 1.5 to 7.2. Results from the second study indicate that the associations between diarrheal disease and key household, environmental, and climatic factors vary according to location – a phenomenon which is masked by global models - and that the associations are both positive and negative (increase and decrease risk). The third study found that if children exposed to the conflict in Northeast Nigeria had not been exposed, their mean weight-for-height z-scores would be nearly half a standard deviation higher, the proportion of moderately and severely wasted children would be 13 percentage points lower, and the prevalence of diarrhea would be 8 percentage points lower than they are.

Conclusions: The studies in this dissertation provide additional evidence for the relationships between climate variability, conflict, and child health in West Africa. Previous studies have provided mixed evidence for the role of factors such as rainfall on child health outcomes. Here, geo-spatial methods identify both areas of heightened risk and the role of specific risk factors in particular locations. For example, the first study found elevated risk of diarrheal disease in northern Cameroon and the second study indicates that two risk factors in that area are low coverage of improved sanitation facilities and low rainfall at the time of the survey. The effects of conflict on child health are conceptually inherent and have been quantified with different methods and outcomes. The research here furthers those efforts by quantifying the impact of a particular conflict – the Boko Haram insurgency - on the health and growth of young children.

The findings from this research support on-going development goals which aim to improve modifiable factors including increased coverage of improved water sources and sanitation facilities, increased educational attainment, the alleviation of poverty and food insecurity, and resolving on-going conflicts around the world. The results also support adaptation measures, which are aimed at factors that are difficult to change in themselves, such as living in areas where temperatures are increasing due to climate change. Such measures include promoting income diversification and access to goods and services for farmers and pastoralists and improving infrastructure and public services in urban areas. Limitations, including spatial, temporal, and methodological issues with the data sources are discussed, as are suggestions for future research.