Effect of diabetes on disability in middle-aged and older adults

Studied the effect of diabetes on disability in middle-aged and older adults. Data were analyzed from baseline and 2-year follow-up waves of the Health and Retirement Study (HRS) and the Study of Assets and Health Dynamics Among the Oldest Old (AHEAD), complementary health interview surveys with complex stratified multistage designs and nationally representative data on groups of community-dwelling middle-aged and older adults. Analytic samples included 8,001 HRS self-respondents aged 51-61 (mean age 55.6), with data for 1992 and 1994, and 5,478 AHEAD self-respondents aged 70 and older (mean age 76.8), with data for 1993 and 1995. Among HRS respondents, 661 (mean age 56.3) had diabetes and 7,340 (mean age 55.5) did not; for AHEAD respondents, 670 (mean age 76.0) had diabetes and 4,808 (mean age 76.9) did not. Ordinary least squares regression models revealed that diabetes was strongly associated with subsequent physical disability (measured by a composite variable combining activities of daily living, mobility, and strength tasks). Controlling for socioeconomic characteristics and common diabetes-related and unrelated comorbidities and conditions reduced the diabetes effect substantially, but it remained a significant predictor of disability in both groups.