LAUSR

Background Previous studies linking social activity and disability have been limited by focussing on self-reported physical performance in older adults (>65). We examined whether social participation in mid-life is associated with objective and subjective measures of physical performance in older age. Methods Participants of the Medical Research Council National Survey of Health and Development reported their involvement in social activities at ages 43 and 60-64 years; frequency of such involvement was classified into thirds. Physical performance was measured at age 60-64 using: grip strength; standing balance; chair rises; timed get-up-and-go; self-reported physical function from the Short Form-36. Multivariable regression was used to examine longitudinal associations between social participation and each physical performance measure. We also investigated whether change in social participation between 43 and 60-64 was associated with each outcome. Results In fully adjusted models, higher frequency of social participation at 43 was associated with faster chair rise (1.42 repetitions/min, 95% CI 0.45-2.39) and timed get-up-and-go speed (2.47 cm/s, 95% CI 0.27-4.67) and lower likelihood of self-report limitations (OR of low physical function 0.67, 95% CI 0.50-0.91) at 60-64 compared with low frequency. Better performance in objectively measured outcomes was observed only if higher social participation persisted over time whereas lower odds of self-reported limitations were found in all groups when compared to those with persistently low participation (ORs 0.43-0.56, all P≤0.02). Conclusions Our findings suggest that associations between higher levels of social participation in mid-life and better physical performance exist only if this social participation persists through to older age.