Case-control and qualitative study of attrition in a community epilepsy programme in rural India.

Dropout from epilepsy programmes is a serious problem in developing countries and has not been systematically studied before. We set up a community-based programme for children with epilepsy in rural India. The aim of this study was to assess reasons for dropout. We assessed medical and sociodemographic variables for their effect on dropout at 12 months using an unmatched case-control design on 32 cases and 62 controls. We also interviewed the parents of 32 children who dropped out of treatment, using a topic schedule. Two-thirds of the dropouts occurred within the first 6 months of treatment. Severely impaired children were more likely to drop out (odds ratio 4.60, 95% CI: 1.0-21.0) and families who had tried AEDs before were less likely to do so (odds ratio 0.12, 95% CI: 0.015-0.88). Denial of diagnosis, access problems and symptom resolution were the other main reasons underlying attrition. Active ascertainment methods should be reconsidered in community programmes. Very poor families without a male head or with long journey times are at high risk of dropout. People with severe impairments need appropriate integrated rehabilitation.