The intervention area was an urban slum in Vellore town. Demographic data of the inhabitants of this area, water supply, and health facilities were available from an earlier survey. Most of the houses in this area (>60%) do not have toilets. There is an open drain running in front of each house, which is used to dispose of solid and liquid household wastes, as well as for defecation by younger children. The major communities are Muslim (47%), Hindu (45%), and Christian (7%), and most belong to the lower socioeconomic strata. The single most common occupation is beedi‐work, which is undertaken by all family members. Most mothers have 2–3 children and live in extended or joint families.
The water supply to this area is from two sources, bore wells which provide 24 hour water supply and municipal taps which supply water for 1–2 hours at variable intervals of time ranging from alternate days to once in 3–4 weeks. The municipal supply is used for drinking water, while the bore well water is used for bathing and cleaning. In all seasons, there is a shortage of drinking water, and water is stored in households in a number of containers, mainly wide mouthed metal or plastic containers, with a capacity of 10–12 litres.
Study children and recruitment
A list of houses with children under 5 years of age was obtained from a previous census of the area. From 689 children, 100 study children and 100 age and sex matched controls were selected by simple random sampling of the listed children. No household had more than one child included in the study or as a control. The inclusion criteria were children resident in Ramanaickapalayam, aged between 6 and 59 months, with no history of chronic diarrhoea, from families not using water treatment methods.
Informed consent was obtained from the mothers, who were taught about solar disinfection of water and given water bottles for the use of the study child. They also received health education on the causes and complications of diarrhoea, home based oral rehydration therapy, and methods of preventing diarrhoea. Control children, who did not receive plastic bottles for solar disinfection of water, were also similarly recruited. A nutritional assessment of both study and control children was carried out using Indian Academy of Pediatrics standards.6
Solar disinfection method
Each household with a recruited study child was given twelve 1‐litre polyethylene terephthalate bottles, sold in the local market, which had one vertical half painted black. The mother was instructed to fill six bottles with water and place them with the clear side on top and the black painted surface below, in a pre‐selected position which was determined by identification of the location that received the maximum direct sunlight. The bottles remained in position through the day and were used for drinking the next day. Mothers were instructed not to transfer the water to other containers.
The period of follow up was from February to August 2002. Follow up visits were conducted by three field workers, under the supervision of a doctor. At each visit, compliance with the solar disinfection methods and incidence of diarrhoea in the study or control child were recorded. A diagnosis of diarrhoea was made if there was passage of three or more loose or watery stools in a 24 hour period. Agreement between the field workers and the supervising doctor was tested at the beginning and at the end of the study, by the doctor paying an unscheduled visit to 20 randomly selected houses one day after the field worker's visit.
Compliance was measured by observing the water bottles being disinfected and was recorded as the percentage of visits during which the water bottles were found in the correct position.
Determination of acceptability
Acceptability was determined by three qualitative methods, focus group discussions, in‐depth interviews, and administration of a questionnaire to record practices and opinions. Two focus group discussions were held among 18 mothers who had children enrolled in the study; these assessed ease of use, financial implications, mechanism of action, and limitations of solar disinfection. Ten mothers were selected for in‐depth interviews, and were interviewed repeatedly by the study doctor about their perspectives and experiences with drinking water safety, diarrhoeal disease, and solar disinfection. The practice and opinion questionnaire was administered to all study family and assessed their practice and knowledge of the solar disinfection method at the end of the study follow up period.
The data collected during six months of follow up was entered using EpiInfo and analysed using SPSS version 9.0.
The study protocol was approved by the institutional ethics committee of the Christian Medical College, Vellore.