Becoming Active After 50 Tied to Less Disability and Greater Recovery

Adults aged 50 or older who took up physical activity after a period of inactivity experienced slower functional deterioration and faster recovery than peers who remained inactive, according to a longitudinal analysis of five population-based aging cohorts published in BMC Public Health. The study pooled data from cohorts in which a total of 266,204 participants met the age criterion. Of these, 115,140 contributed eligible paired physical-activity and functional-state data, and 112,855 entered the primary complete-case analysis.

The findings suggest that the timing of activity change, not just the level of activity at a single point in time, carries information about how disability unfolds and reverses in later life. The research team classified each participant's change in physical activity between two assessments as stable inactive, became active, became inactive, or stable active. The intervals between the first and second assessments had a median of two years in four of the five cohorts, while the Mexican cohort had a median interval of three years with an interquartile range of two to nine years.

Functional status was measured across three states: fully Independent, disability limited to instrumental activities of daily living, and disability extending to basic activities of daily living. Death was not modeled as an outcome state; instead, observation ended at each participant's last valid functional-state assessment. To quantify how activity change related to movement between functional states, the researchers fitted continuous-time three-state Markov models separately within each cohort and combined the cohort-specific estimates using random-effects meta-analysis.

Within the stratum of participants who were inactive at baseline, initiating activity versus remaining stably inactive was associated with a lower transition intensity from Independent to instrumental-activity disability, with a hazard ratio of 0.58 (95% confidence interval, 0.46 to 0.72). Initiation was also associated with a higher recovery intensity from basic-activity disability back to full independence, with a hazard ratio of 1.73 (95% confidence interval, 1.40 to 2.13). In the opposite direction, among participants active at baseline, ceasing activity versus continuing it was associated with a higher deterioration intensity from Independent to instrumental-activity disability (hazard ratio, 1.28; 95% confidence interval, 1.03 to 1.58) and a lower recovery intensity from basic-activity disability (hazard ratio, 0.66; 95% confidence interval, 0.61 to 0.72).

Across the primary analysis, becoming active and remaining active were generally associated with less deterioration and greater recovery than stable inactivity, and stable activity showed the most favorable overall profile of transitions between functional states. Becoming inactive, by contrast, was associated with less favorable subsequent transitions than continued activity. The complete-case analysis served as the primary approach, and a sensitivity analysis using multiple imputation produced results that were directionally concordant, meaning the pattern of associations held when missing data were handled differently.

The authors flag substantial statistical heterogeneity between cohorts. These measures, combined with the observational design, mean the study shows associations between activity change and subsequent functional transitions, not causal effects, and the authors explicitly state that cautious interpretation is warranted. Several design features shape how the results should be read.

Because death was not modeled as an outcome, the estimates describe transitions among living participants between functional states and do not capture how mortality might select or censor the observed population. The Markov framework treats the functional states as a progression in which instrumental-activity disability lies between full independence and basic-activity disability, allowing the models to estimate both deterioration intensities moving downward and recovery intensities moving upward. The five cohorts also differ in follow-up intervals, with the Mexican cohort's wide interquartile range of two to nine years reflecting far more variable assessment spacing than the roughly two-year intervals elsewhere.

The paired-assessment structure allowed the researchers to separate people who changed their activity status from those who did not, and to show that initiation and cessation carried opposite patterns of association with deterioration and recovery. The study's conclusions are stated in associational terms: becoming active after a period of inactivity and remaining active were associated with less functional deterioration and greater recovery, whereas becoming inactive was associated with less favorable subsequent transitions than continued activity. The substantial variation between cohorts and the observational design are cited as reasons for cautious interpretation.

The framework linking measured activity change to modeled transitions among independence, instrumental-activity disability and basic-activity disability provides a basis for examining how robust these patterns prove across aging populations with different follow-up rhythms and baseline activity levels.

Subject of Research: Changes in physical activity and transitions between functional states among adults aged 50 years or older: a longitudinal study of five aging cohorts

Article Title: Changes in physical activity and transitions between functional states among adults aged 50 years or older: a longitudinal study of five aging cohorts

Article References: Luo, Y., Xu, S., Lu, Z., & Song, X. (2026). Changes in physical activity and transitions between functional states among adults aged 50 years or older: a longitudinal study of five aging cohorts. BMC Public Health. https://doi.org/10.1186/s12889-026-29979-2

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29979-2

Keywords: Changes, physical, activity, transitions, between, functional, states, adults, aged, years, older, longitudinal