This population-based, cross-sectional study in the isolated Sahsa municipality of Nicaragua found a high prevalence of household diarrhea (41%), which is consistent with data from the Nicaragua Ministry of Health (MINSA)2. Importantly, the analysis suggests that latrine ownership with no reported overflow was associated with reduced household prevalence of diarrhea.
The use of prevalence differences in our analysis allows for a prediction of benefit from a change in exposure, with the number of needed interventions, or the number needed to treat (NNT) in order to reduce the household diarrheal burden by 1. Our data estimate that the NNT for water treatment with the use of filters is 17 households, and for chlorine treatment it is 20 homes. Similarly, the NNT is 10 for latrine use, and 6 for use of well-designed latrines which do not overflow. This suggests that an important reduction in diarrheal burden can be obtained with changes in the sanitation infrastructure.
Previous investigations addressing water source, water treatment, and latrine use have led to inconsistent conclusions on associations with diarrheal disease[11–17]. A recent review found that point-of-use chlorine treatment was associated with reduction in diarrhea in 9 of 10 reviewed studies. In only 5 of these, however, was the association statistically significant. Sand filtration systems have been shown to be effective at reducing diarrhea [16, 17]. Our results did identify an association of water source and treatment with reduced prevalence of diarrhea, but these findings were not statistically significant. For sanitation infrastructure, two studies have shown a reduction in diarrhea associated with latrine access[13, 18]. Our results are similar in that we found a reduction in diarrhea for households with access to a latrine. These results, however, were not statistically significant. A statistically significant association was found for those with access to a properly situated latrine that does not overflow.
The discordant results may stem from the specific emphasis of the studies. While water purification and latrine use are factors which protect against diarrheal disease, few studies have addressed these factors in a remote, impoverished, and high prevalence region. This study suggests that proper latrine function is effective in preventing diarrhea even in areas with many other risk factors for diarrhea, by highlighting the importance of proper placement and maintenance of latrines. Ownership of a latrine that overflowed was associated with little to no change in the prevalence of household diarrhea compared to not owning a latrine (PD: 0.02, 95%CI: -0.18, 0.23), whereas ownership of a latrine that did not overflow was associated with an appreciable difference (PD: -0.17 95% CI: -0.36, 0.01). The identification of the potential causal interaction between latrine ownership and overflow may have important implications for directing future interventions. Without the interaction term, the PD model treats all households reporting latrines with equal risk, regardless of the proper functioning of the latrine, which may lead to inaccurate effect estimates.
The main strengths of this study include the population-based sampling of the isolated Sahsa region and the high subject response rate. The incorporation of local health promoters was a key factor in the high response rate. The assessment has several limitations. The study region is isolated and the travel required to conduct interviews in selected communities was challenging. Three initially selected communities could not be reached due to travel conditions and safety concerns. As detailed maps of the region were not available, second stage sampling used the "compass" approach for household selection. By allowing interview teams to select interview locations in this way, selection bias may have been introduced. A further potential bias is that data are based on self-report. No inspections of the home were made to verify answers on water treatment or latrine set-up.