Health impact evaluation of improved water supplies and hygiene practices in Sri Lanka: background and methodology.

Between January 1987 and March 1988 a Health Impact Evaluation (HIE) of the Rural Water Supply and Sanitation Project (RWSSP) was conducted in Kurunegala District. Four related activities were undertaken: 6598 children were recruited into a case-control study of diarrhoea morbidity conducted in five hospitals; an additional 1649 children from the catchment areas of three of the five hospitals were recruited as a community comparison group; environmental microbiology was performed on water samples collected during 3092 visits to the homes of children recruited into the study; diagnostic stool microbiology was performed for 371 children with diarrhoea and 121 controls. In this paper the seasonality of reported diarrhoea and the socio-demographic characteristics and health-related behaviours of each study group are examined. Cases, clinic controls and the community comparison group were all drawn from the essentially rural settlements typical of the southern part of the dry zone of Sri Lanka. The majority of households in these settlements are Sinhalese buddhist, and the main economic activity is subsitence farming. There was some evidence of differences between the three groups, community recruits tending to come from larger and wealthier households than the children recruited at the hospitals. Issues arising in the design and analysis of this Health Impact Evaluation are discussed. In particular, potential sources of bias are examined and the question of validity investigated.