Using the SF-36 with older adults: a cross-sectional community-based survey

Objectives: to assess the practicality and validity of using the 36-item short-form health survey (SF-36) in a community-dwelling population over 65, and to obtain population scores in this age group. Design: postal survey, using a questionnaire booklet containing the SF-36 and other health-related items, of all those aged 65 or over registered with 12 general practices in Sheffield. Non-respondents received up to two reminders at 3-weekly intervals. Sample: 9897 subjects, aged 65-104 years. Main outcome measures: scores for the eight dimensions of the SF-36 and a modified version of the physical functioning dimension. Results: the SF-36 achieved a response rate of 82% (n=8117) and dimension completion rates of 86.4-97.7%. Internal consistency measured by Cronbach's alpha exceeded 0.80 for all dimensions except social functioning. These results compare favourably with postal surveys of younger adults. We calculated scores for older adults by age and sex. Comparison with data from younger people showed how physical health declines steeply with age, in marked contrast to mental health. Conclusions: the SF-36 is a practical and valid instrument for use in postal surveys of older people living at home. The population scores provided here may facilitate its use in future surveys of older adults.