PATTERNS OF ENVIRONMENTAL FECAL EXPOSURE AND ASSOCIATIONS WITH CHILDHOOD ILLNESS IN RURAL BANGLADESH: A LATENT CLASS ANALYSIS

Embargo until
2020-12-01
Date
2017-08-31
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
Inadequate sanitation, contaminated water and poor hygiene contribute to childhood disease morbidity and mortality through enteric and respiratory illness, trachoma, soil-transmitted helminths, parasitic diseases, environmental enteropathy and malnutrition. Multiple fecal exposures present different combinations of risk factors in low-income settings with limited infrastructure and poor hygiene behaviors. Large scale programs aimed to improve sanitation have an inconclusive health impact. Implementers still lean towards combining water, sanitation and hygiene interventions, despite limited evidence of additional health benefits from combined approaches. The WASH Benefits Bangladesh study is a community-based cluster randomized trial in rural Bangladesh designed to assess the impact of single and combined water, sanitation, hygiene and nutrition interventions in single and combined interventions on child health. This dissertation aims to 1) assess the impact on respiratory illness on children under 3 years of age from single water, sanitation, hygiene and nutrition interventions when delivered alone or in combination; 2) identify sub groups of rural households that vary in risk for environmental fecal exposures using latent class analysis, 3) to examine whether the latent classes are associated with higher risks of childhood diarrheal and respiratory illness and 4) if water, sanitation, hygiene and nutrition interventions have differential impact in reducing disease prevalence across latent classes. We found that water, sanitation and hygiene interventions reduce respiratory illness in young children. The same benefit was observed when water, sanitation and hygiene interventions were successfully integrated with nutrition interventions. Latent class analysis identified four subgroups (1-4) with increasing environmental risk profiles based on household characteristics in rural Bangladesh. Groups with unfavorable environmental conditions were associated with lower socioeconomic status, income and education. We found an increased risk of diarrheal disease in all latent subgroups compared to the ‘1-most favorable’ class characterized by water sealed improved latrines, notably a 5-fold increase risk of diarrhea in the ‘4 most unfavorable’ group who did not have access to any latrines. For diarrheal diseases, we found reductions in reported diarrheal disease prevalence in index children following sanitation (S), handwashing (H), nutrition (N) and WSHN interventions compared to control households in the ‘3- unfavorable’ latent subgroup. This indicates that households with less sanitary conditions are more likely to benefit from interventions that reduce the transmission of pathogens. Single WASH interventions may be effective in reducing respiratory illness and should be prioritized with limited resources. We highlighted the use of understanding the clusters of exposures to ensure interventions are adequately aligned to be effective. In low-income countries, where competing fecal pathways exist, improved health impact might be more practically achieved using approaches such as latent class analysis that incorporate interactions between environmental and socio-economic factors to inform holistic intervention strategies.
Description
Keywords
latent class analysis, diarrhea, respiratory illness, childhood illness, Bangladesh, water, sanitation, hygiene
Citation