Sewerage may lower incidence of diarrhea

Issue: August 2010

Sewerage interventions appeared to cut diarrhea incidence in developing countries by approximately 30%, although onsite solutions, such as latrines, septic tanks or dry-composting toilets, may be just as beneficial and more cost-effective, recent study data indicate.

“Worldwide deaths from diarrhea in children younger than 5 years are estimated at about 1.9 million every year, about 20% of total child deaths,” wrote researchers from sites in the U.K. and Spain. “In developing countries, 10% of the total burden of disease is estimated to be attributable to poor water, sanitation, and hygiene, and about 60% of this is due to diarrheal disease.”

To determine whether sewerage reduces prevalence of diarrhea and related outcomes, the researchers conducted a meta-analysis by searching various databases, including Medline and the Institute for Scientific Information Proceedings database, for relevant studies. After reviewing the articles, they compared results of sewerage with those pertaining to other interventional strategies.

Of the 9,274 studies found, 25 met inclusion criteria. Six cohort studies, four case-control studies and one non-randomized trial were among the incidence-based investigations evaluated, while the other 14 were cross-sectional. All reviewed studies were conducted between 1980 and 2006, the researchers noted, with 14 coming from Brazil and others from Mexico, Nicaragua, Honduras, Peru, Iran, Syria, Saudi Arabia, Australia and the United States. Twenty-one focused on children, with target ages ranging from 0 to 12 months and 3 months to 14 years. Although the primary outcome among most studies was diarrhea, the prevalence of Giardia, intestinal parasites and cryptosporidum were also included in several analyses.

The researchers employed a modified five-point binary scale to evaluate study quality. One point was awarded based on the following criteria:

Clearly defined sanitations levels;
Whether outcome measurement was acceptable;
Whether selection of a control group was acceptable;
Whether participation or follow-up was acceptable;
Whether adjustment, matching and restriction were acceptable.

The pooled RR estimate from all 25 studies was 0.70 (95% CI, 0.61-0.79), according to study results. Heterogeneity was high, and the researchers reported no publication bias. Heterogeneity also remained high in most subgroup analyses but dropped to moderate among cross-sectional studies, those in which sewerage was compared with septic-tank only and studies in which diarrhea was not the primary outcome. The researchers noted that heterogeneity was low among studies conducted outside Latin America and those excluding children.

Results also revealed that the pooled relative risk estimate of effect for studies with diarrhea as the outcome was 0.70 (95% CI, 0.58-0.84). This calculation, however, did not differ significantly for those studies in which diarrhea was not the outcome (0.71; 95% CI, 0.62-0.82). The pooled relative risk for incidence studies was 0.72 (95% CI, 0.59-0.88) and 0.67 (95% CI, 0.58-0.77) for studies with at least a three-point rating.

Data also showed that the pooled relative risk estimate for studies with sewerage as one level of sanitation factor was 0.67 (95% CI, 0.59-0.77) when compared with 0.70 (95% CI, 0.61-0.79) for all studies. Subgroup analyses also indicated that sewerage was more beneficial in studies where sanitation level was deemed “very poor” (RR=0.41; 95% CI, 0.27-0.61).

“Considering all studies included in this review, pooled estimates suggest a reduction of about 30% (RR=0.70; 95% CI, 0.61-0.79) in diarrhea incidence and in other indicators of enteric disease burden; a similar reduction (34%) is obtained if we included only studies rated as high quality,” the researchers wrote.

They noted, however, that their findings do not suggest that sewerage is superior to urban onsite sanitation and would likely show improvement in many developing countries as long as the system was functioning well. Other concerns, however, must be considered before implementation.

“Most studies included in our review have been from middle-income countries, such as Brazil and Mexico, as opposed to low-income countries in Africa and Asia where cost constraints are more severe and current institutional capacity to maintain sewerage systems is, in some countries, clearly inadequate,” they wrote.

Norman G. Lancet. 2010; doi: 10.106/S1473-3099(10)70123-7.