Participation bias in postal surveys among older adults: the role played by self-reported health, physical functional decline and frailty.

Postal survey is a simple and efficient way to collect information in large study samples. The purpose of this study was to find out differences between older adults who responded to a postal survey on health outcomes and those who did not, and to examine the importance of frailty, physical functional decline and poor self-reported health in determining non-response. We mailed out a questionnaire on general health twice at a year's interval to 1000 individuals ≥60 years, and members of the medical insurance scheme of the French national education system. At Year1, 535 persons responded to the questionnaire (65% women, 70.9 ± 8.4 years). A year later (Year2), we obtained 384 responses (63.3% women, 70.5 ± 7.8 years). Compared to respondents, non-respondents at Year2 were more frequently categorized as frail, reported more often to be in bad health, and had more physical functional declines. Frailty, physical functional decline and poor self-reported health increased the likelihood of not responding to Year2 questionnaire, with poor self-reported health weakening the association of physical functional decline and non-response. Respondents of this postal survey are fitter and healthier than non-respondents. This participation bias precludes the generalization of postal surveys results.