Trajectories of self-reported physical activity and predictors during the transition to old age: a 20-year cohort study of British men

Background: Maintenance of physical activity (PA) during later life is associated with optimal health; however, the long-term trajectories of PA into old age and their predictors have not been extensively researched using latent class methods. This study aimed to identify trajectories of self-reported PA and their predictors in men transitioning from midlife to old age. Methods: 7735 men (aged 40-59 years) recruited in 1978-80 were followed up after 12, 16 and 20 years, reporting PA, health status, lifestyle behaviours and socio-demographic characteristics. Group-based trajectory modelling identified the trajectories of PA and associations with time-stable and time-varying covariates. We considered a range of sociodemographic and health and lifestyle factors as potential covariates. Results: 4952 men (mean baseline age 49.1 +/- 5.6 years) providing PA data at >= 3 time points were included in analyses. Three distinct 20-year trajectories were identified: low decreasing (24.6%, n = 1218), light stable (51.1%, n = 2530) and moderate increasing (24.3%, n = 1204). Being older, having a manual occupation, having never married or had children, residing in the midlands or North of England, suffering from a range of health conditions, being a smoker/ ex-smoker and never consuming breakfast cereal or alcohol were independently associated with reduced odds of belonging to the moderate increasing trajectory group compared to the low decreasing group. Of the time-varying covariates considered, leaving employment was associated with a decrease in PA in the low decreasing group (beta -0.306, p < 0. 001) but an increase in the light stable (beta 0.324, p < 0.001) and moderate increasing groups (beta 0.847, p < 0.001). Developing cardiovascular-related conditions was associated with a decrease in PA in the low decreasing (beta -0.408, p < 0.001) and light stable groups (beta -0.118, p < 0.001) but no association was observed in the moderate increasing group (beta -0.060, p = 0.313). Conclusions: Three distinct trajectories of PA were identified in men transitioning from midlife to old age, of which nearly a quarter had persistently low levels of PA. Promotion efforts may need to focus attention prior to middle age and consider a number of sociodemographic, health and lifestyle factors to sustain PA into old age. The effects of retirement and development of cardiovascular disease may vary according to PA trajectories.