Voluntary health agencies as target populations for epidemiologic research.

The ability of two voluntary health agencies to provide suitable target populations for epidemiologic research was explored in a pilot study of epilepsy. The results suggest that, properly approached, voluntary agencies offer advantage for this purpose. In the first agency (Group A), subjects were recruited by mail, producing a response rate of 15%. In the second agency (Group B), subjects were recruited by telephone, producing a response rate of 87%. A structured, precoded telephone interview about personal and family history of seizure disorders was administered to both groups of subjects. Subjects in Group A gave permission to contact a higher proportion of their eligible relatives than did those in Group B (73 vs 57%). Permission was obtained more often for relatives reported to have had seizures (Group A 86%, Group B 78%) than for other relatives. 89% of relatives contacted directly agreed to be interviewed. Consent forms for medical record review were signed and returned by 95% of Group A and 77% of Group B subjects. Diagnoses of etiology and seizure type of epilepsy based on the interview data agreed with diagnosis based on the medical records in most cases. In first-degree relatives of subjects with epilepsy, reported rates of epilepsy did not appear to be seriously biased.