The contribution of unimproved water and toilet facilities to pregnancy-related mortality in Afghanistan: analysis of the Afghan Mortality Survey.

OBJECTIVE: To estimate the effect of unimproved household water and toilet facilities on pregnancy-related mortality in Afghanistan. METHODS: The data source was a population-based cross-sectional study, the Afghan Mortality Survey 2010. Descriptive, univariate and multivariate logistic regression analyses were carried out, comparing 69 pregnancy-related deaths (cases) and 15386 surviving women (non-cases) who had a live birth or stillbirth between 2007 and 2010. RESULTS: After adjusting for confounders, households with unimproved water access had 1.91 the odds of pregnancy-related mortality [95% confidence interval (CI) 1.11-3.30] compared to households with improved water access. We also found an association between unimproved toilet facilities and pregnancy-related mortality (OR = 2.25; 95% CI 0.71-7.19; P-value = 0.169), but it was not statistically significant. CONCLUSIONS: Unimproved household water access was an important risk factor for pregnancy-related mortality in Afghanistan. However, we were unable to discern whether unimproved water source is a marker of unhygienic environments or socio-economic position. There was weak evidence for the association between unimproved toilet facilities and pregnancy-related mortality; this association requires confirmation from larger studies.