Despite growing urbanization, our understanding of the
impacts
of water and sanitation on human health has largely come from studies
in rural sectors. To this end, we collected data at both regional
(water quality measures from water treatment systems) and community
(cross-sectional surveys) scales to examine determinants of enteric
pathogen infection and diarrheal disease among infants in Addis Ababa,
Ethiopia. Regionally, the Legedadi water treatment plant had significantly
lower heterotrophic plate counts, total coliform counts, and fecal
coliform counts compared with the Gefersa water treatment plant. The
number of pathogen types in infant stool also differed by plant. Decreases
in chlorine levels and increases in the relative abundance of
Gammaproteobacteria
with distance from treatment plants
suggest a compromised water distribution system. In communities, infants
in households that obtained water from yard pipes or public taps had
significantly lower odds of diarrhea compared to households that had
water piped into their dwellings (OR = 0.35, 95% CI 0.16, 0.76, and
OR = 0.39, 95% CI 0.15, 1.00, respectively). Similarly, infants in
households that boiled or filtered water had significantly lower odds
of diarrhea compared to households that did not treat water (OR =
0.40, 95% CI 0.19, 0.86 and OR = 0.23, 95% CI 0.06, 0.84, respectively).
Integrating multiscalar data better informs the health impacts of
water in urban settings.