The overall aim of drinking-water projects is to improve health status by providing pathogen-free drinking-water. The impact of drinking-water quality on the incidence of diarrhoeal disease and health status depends on the importance of transmission of pathogens via drinking-water, relative to transmission via other routes. Feachem categorized the water-related infectious diseases as water-washed or water-borne (1). The classification was later expanded into a unitary environmental classification of water- and excreta-related communicable diseases with seven categories (2). Cairncross et al. suggest that a distinction should also be made between transmission in two different physical domains: the public domain (outside the household) and the domestic domain (inside the household) (3). A desired health benefit would only be obtained if transmission of pathogens in both the domains is prevented.
The drinking-water supply sector has mostly targeted the water-borne transmission of pathogens. The most common method employed is the chlorination of drinking-water at treatment plants and in the distribution systems. Chlorination is considered essential to make it safe for drinking, especially when water is obtained from surface-water sources. While there is universal acceptance and use of drinking-water chlorination in epidemic outbreak situations, there is an ongoing discussion on whether the interruption of water-borne transmission of pathogens reduces the occurrence of endemic diarrhoeal diseases in rural areas in developing countries (4-7). Most water-supply systems in Pakistan and other developing countries are not working according to design, and many are completely dysfunctional. This raises the question whether improvements in these schemes and provision of bacteriologically safe drinking-water would have an impact on the incidence of diarrhoeal disease. This study was done to examine the effect of chlorination of public-water sources on bacteriological water quality and endemic diarrhoea in a village in Pakistan.
MATERIALS AND METHODS
The intervention took place in a village (Village A) located in an irrigated area in southern Punjab, Pakistan. For comparison, water quality and incidence of diarrhoea were also monitored in a second village (Village B). The two villages were almost identical with respect to layout, ethnic composition, and socioeconomic status. They were situated 2 km apart and had separate water-supply schemes, both receiving raw water from the same irrigation canal. The irrigation canal was supplied with water from the Sutlej River, which received wastewater from the city of Lahore, located 250 km upstream from the two villages.
Both the water-supply schemes had a large sedimentation tank connected to slow sand-filters and a clear-water well, from which water was pumped to the village. The slow sand-filter in Village B was in a very poor condition due to lack of sand, and the filter could better be described as just a small retention tank. The slow sand-filter in Village A was in a somewhat better condition, although not functioning according to the original design because of non-optimal maintenance and growth of water plants on the filters. Accordingly, the initial water quality in the irrigation canal was expected to affect the water quality in both the water-supply systems. Both the systems provided water for about two hours early in the morning and for an hour in the afternoon. To overcome the non-supply hours, many households had installed a large concrete water-tank or a plastic drum for water storage. Nearly all households used traditional clay-pitchers inside the household perimeter for storage of drinking-water and cooling purposes. No families reported boiling or any other form of treatment of drinking-water before use. Not all households in the two villages used the water from the water-supply scheme. The alternative supply was shallow groundwater drawn by hand-pumps. …