Comparison of older adult subject and proxy responses on the SF-36 health-related quality of life instrument

Studies of older adults' health status and health-related quality of life (HRQoL) often rely on proxy responses when subjects have problems that affect their ability to respond. With the increased interest in outcomes research in health care, it is important to examine proxy reliability on HRQoL instruments. This study compares 32 pairs of subject-andproxy responses on the eight subscales and two summary scales of the Short Form 36 (SF-36). Subjects and their proxies, recruited from senior centers and residential facilities, were interviewed face-to-face within a seven-day period. Subjects were 60 years of age or older and had passed a brief cognitive screen, and proxies were geographically proximate and had seen the subject during the past week. Results showed that although moderate intra-class correlations were found on six of the eight measures, an item-level kappa statistic indicated poor to fair agreement on all subscales except items of Physical Functioning and Role Physical. Moreover, paired t-tests revealed proxy mean scores that were significantly lower on the Physical Functioning, Vitality, and Mental Health subscales. Given the mixed findings, until further research is done, researchers and clinicians should exercise caution when using proxy responses for older adults with the SF-36.