An investigation made by local government agency into the social conditions and state of health of old people had an answer frequency of 89 %. This investigation has been followed up with inquiries into the non-response group. The main result is a striking difference between the non-response group and the response group as to the state of health. Far more people in the non-response group said that they were in good health. Fewer people took medicine in the non-response group. One way of explaining this difference is that there are more relatively young old-aged pensioners in the non-response group. Finally, the possibilities are discussed of using mail questionnaires as an alternative to expensive interview investigations when old peopleʼs needs of care are investigated and also the possibilities of using the questionnaire interview investigation made by a local government agency as an instrument for checking the health of old people.