[Experimentation on a mail questionnaire for a case-control study]

A mail questionnaire, collecting information mainly on occupational history, was sent to 50 subjects who had been interviewed at home with a longer questionnaire three years before, and to 26 subjects who refused at that time to be interviewed. The 50 subjects were sampled among controls enrolled in a population based case-control study on soft tissue sarcomas; the 26 were all non-respondent controls in the same study. The questionnaire was returned by 83% and 14% respectively. A comparison between home and mail interviews suggests that the quality of information is strictly comparable for 78% of the items. For discordant items, about 50% showed a better quality of information from home interviews; in the other 50% a better information was provided by the mail interview.