Background Physical activity confers a wide range of health benefits and is particularly important among older adults. Retirement is an important later life transition which may represent a critical period for physical activity in older age. Existing evidence on the association between retirement and physical activity is inconclusive, with some evidence suggesting that the physical activity implications of retirement may differ by occupational activity level. This study aimed to examine the physical activity implications of retirement and explore differences by gender and occupation.

Methods This study used data from waves 4–9 of the English Longitudinal Study of Ageing, which was designed to represent people aged 50 and over living in private households in England. Of the 10,693 participants included in the study, 2,199 (21%) retired during the study period. Within-individual change was examined using fixed effects regression models to evaluate whether there was an association between retirement and physical activity. Differences in the association between retirement and physical activity by gender and occupational activity were explored using interactions and stratified analyses.

Results Retirement was associated with a significant increase in physical activity β = 0.602METhrs/week (95% Confidence intervals [95% CI] = 0.490 to 0.713) and a tripling of the likelihood of moving into the highest category of physical activity compared to being inactive (relative risk = 2.827; 95% CI = 2.025 to 3.947). There were significant interactions between retirement and past occupational activity level (p<0.001), such that people retiring from sedentary or standing occupations experienced a significant increase in physical activity with retirement (0.580METhrs/week 95% CI = 0.375 to 0.785 and 0.330METhrs/week 95% CI = 0.095 to 0.565 respectively) but retirement from an occupation involving heavy manual labour was associated with a decrease in physical activity (-0.823METhrs/week 95% CI = -1.440 to -0.206). The physical activity implications of retirement did not differ between men and women. All analyses were conducted using Stata SE 17.

Discussion This study quantified the importance of retirement for later life physical activity. With demographic ageing, the population health importance of later life physical activity will likely become more important. These findings could help to inform the design of public health interventions to increase physical activity around the retirement transition, including being able to target those at greatest risk of reduced physical activity.