European Review of Aging and Physical Activity · Nov 19, 2026
Physical activity loss and re-entry across biological age and functional adversity in older adults
By Lo WLA, Wu D, Chen L et al.
Physical activity participation changes as older adults’ health and circumstances evolve, yet the factors associated with activity loss and re-entry may differ. The United States Health and Retirement Study data from 2014 to 2020 were analysed, comprising 30,139 loss-risk person-wave intervals from 15,189 participants and 12,466 re-entry-risk intervals from 8,010 participants aged 50 years or older. Separate transition cohorts were defined for loss of weekly moderate or vigorous physical activity among participants at baseline and for re-entry, defined as transition from no weekly MVPA at baseline to weekly MVPA at follow-up. Modified Poisson models estimated adjusted risk ratios for baseline physical limitation, cognition, family structure, social participation, and PhenoAge acceleration. High physical limitation was associated with greater risk of activity loss (risk ratio [RR] = 1.39, 95% confidence interval [CI]: 1.31, 1.47) and lower probability of re-entry (RR = 0.80, 95% CI: 0.75, 0.85). Low social participation showed a similar pattern (loss RR = 1.30, 95% CI: 1.15, 1.48; re-entry RR = 0.78, 95% CI: 0.70, 0.87). Poor cognition was associated with loss but not re-entry, while family structure was associated with neither transition. Higher PhenoAge acceleration was associated with greater loss risk and lower re-entry probability but interaction analyses did not detect statistically significant effect modification by physical or cognitive adversity. Preventing activity loss and supporting re-entry are distinct public health challenges, with physical function and social participation representing potentially important factors to consider in supporting later-life activity participation.