To the Editor: The paper by Norton et al. in the December 1994 issue of the Journal raises the issue of older nonresponders.1 It has been suggested that nonresponders tend to be more unhealthy than responders.2, 3 However, the characteristics of nonresponders may differ according to the nature of the invitation issued. Response to research surveys may not be the same as response to invitations from personal medical practitioners to have a health check. Since 1990, general practitioners in the United Kingdom are contractually obliged to invite all their patients who are more than 75 years old to have an annual health check.4 We report the reasons for lack of response among persons more than 75 years who did not respond to an invitation from their general practitioner to have a health check. The period of study was October 1991 to December 1992. Three general practices within Cambridge and Huntingdon Health Commission, United Kingdom, were recruited to take part in the study. The combined patient population was 21,477 people, and of these, 1342 were older than 75 years. All persons aged 75 years and older were sent a postal invitation to have a health check. An attempt was made to contact by telephone all those who did not respond to the letter of invitation. Those who then agreed to a visit were seen. The checks, performed by a nurse who had been specifically recruited for this study, took the form of a questionnaire, which was completed by the nurse in the presence of each patient. Those who did not initially respond were specifically questioned about their reason for nonresponse. Data from the questionnaires were analyzed using the statistical package for social sciences on the mainframe computer of the University of Cambridge (SPSS Inc., Chicago, IL). Invitations to have a health check were mailed to 1342 individuals, 841 female and 501 male, mean age 82.1 ± 4.9 years. There were 182 persons, 119 female and 63 male, mean age 81.4 ± 4.8 years, who did not respond to the postal invitation and who were untraceable for a variety of reasons (moved house, deceased). There were 186 persons who did not initially agree to have a health check and were contactable. Of these, 44 (23.7%) persons, 27 female and 17 male, mean age 81.6 ± 4.3 years, actively declined the invitation, i.e., they wrote back stating that they were in good health and did not wish to have a health check. The other 142 persons, 92 female and 50 male, mean age 80.8 years ± 3.6, who did not respond to the postal invitation, when subsequently contacted by telephone agreed to have the check. There were 120 (64.5%) persons who did not respond because they had already seen their general practitioner within the previous 12 months (108 persons had seen their general practitioner within the previous 6 months and 61 persons had seen their general practitioner within the previous month). Of the 22 (11.8%) who did not respond and who had not seen their general practitioner in the previous 12 months, none reported ill health. Seven persons did not respond because of ill health. Of these, four had seen their general practitioner within the previous month, and the other three had seen their general practitioner within the previous 6 months. In this study, the older people who did not respond to an invitation to have an annual health check refrained from doing so mainly because they had already seen their general practitioner within the previous year or because they were well. In evaluation of nonresponse among older people, therefore, reasons for nonresponse should be considered. Older nonresponders are not a homogeneous group, and it should not be assumed that nonresponse always indicates poor health status. They are a heterogeneous group who differ in their reasons for nonresponse according to the type of invitation issued.