Date of Award

Fall 8-10-2015

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Public Health

First Advisor

Dr. Christine Stauber

Second Advisor

Dr. Ciara O'Reilly

Third Advisor

Dr. Eric Mintz

Fourth Advisor

Dr. Richard Rothenberg


Diarrhea, malnutrition, and human immunodeficiency virus (HIV) are three leading causes of childhood morbidity and mortality. All of these conditions impact children living in sub-Saharan Africa disproportionally to the rest of the world. This dissertation outlines three papers in which these three childhood health problems are explored in greater detail in one low-income country in sub-Saharan Africa (Kenya) using data from the Global Enteric Multicenter Study (GEMS). In particular the three papers will explore further: 1) risk factors for the duration of diarrhea, 2) water, sanitation, and hygiene (WASH) practices among people living with HIV (PLHIV), and 3) a methodological paper exploring distribution and outlier analysis from anthropometric data documenting stunting. As evidenced here the burden of diarrhea, malnutrition, and HIV/AIDS is significant within this population in rural western Kenya. It’s likely that in many countries in sub-Saharan Africa similar health problems and challenges are being experienced in young children. These studies contribute to the epidemiology of diarrheal duration. They add to the knowledge about the water, sanitation, and hygiene practices (WASH) practices and behaviors among HIV infected individuals. And, they contribute to the methodological processes for cleaning anthropometric data. Based on findings from our studies, stunting, a signal of long-term nutritional deficiencies and malnutrition is a serious health problem within this population. We found that a high proportion of both case and control children were stunted. In addition, children who were stunted at enrollment tended to have longer duration diarrhea. Improving overall nutritional status among young children in rural western Kenya is imperative. Furthermore, a more standardized approach to dealing with extreme values within anthropometric data is important so that these measures can be better understood and interpreted. Another key finding was the importance of the main drinking water source. We found using an unimproved water source to be a risk factor for diarrheal duration. We found no differences in drinking water sources among HIV-positive households as compared to HIV-negative households. However, a large proportion of HIV-positive and HIV-negative households were using unimproved water sources. Due to the increased adverse consequences among PLHIV and the increased risk for longer duration diarrhea improving access to adequate water sources is especially important among children living in this region of rural western Kenya. Our studies also highlighted differences in WASH practices among HIV-positive households who were aware of their status for at least one month prior to enrollment into GEMS as compared to households who found out their status more recently. Further investigation should identify what is contributing to these results so they can be applied at a larger scale in similar populations. Last, the etiologic agent with which the child is infected seemed to be an important factor in diarrheal duration. Prevention and treatment methods for pathogens causing severe illness, such as Cryptosporidium are warranted.