Ideal 24-h physical activity trajectory reduces all-cause, cause-specific mortality and cardiovascular outcomes through aging deceleration and inflammation regulation: A UK biobank study

Physical activity (PA) is associated with mortality and cardiovascular disease (CVD). However, the effect of circadian PA trajectories remains ambiguous. This study aimed to explore ideal circadian PA patterns to reduce mortality and CVD, and potential mediators. 502,400 participants from UK Biobank were recruited between 2006 and 2010. Among them, 102,323 participants got valid continuously capturing acceleration data over 7 days by wrist-worn accelerometer. K-means cluster analysis was used to identify PA trajectories. The associations of PA with all-cause, cause-specific mortality and CVD were assessed by cox regression. A sensitivity test was also conducted, starting from the time of acceleration collection and excluding participants with corresponding disease prior to it. Furthermore, the mediation of aging and inflammation were explored. During a median follow-up of 12.9 years, 3482 deaths were recorded (704 were due to CVD). Five distinct PA trajectories were identified: Persistently Low, Moderate and Stable, Single Increase, Double Increase, and Vigorous patterns. Ideal PA trajectory patterns offered progressively protective benefits against all-cause, CVD caused mortality and CVD, especially in Double Increase and Vigorous patterns. Other cause-specific mortality and renal failure incidence showed similar trend. The sensitivity result was consistent. The mediating effects of phenotypic age and inflammation markers were statistically significant. Ideal PA trajectories offered protective benefits against all-cause, cause-specific mortality and CVD. The protection was associated with both intensity and circadian distribution. Double Increase and Vigorous activity patterns decreased these risks more significantly. Crucially, this protection was mediated by aging deceleration and inflammation regulation. •Continuous acceleration data was captured from 102,323 participants.•Five distinct PA trajectories were identified by machine learning cluster analysis.•Ideal PA trajectories offer progressively protective benefit against mortality&CVD.•The protection was mediated by aging deceleration and inflammation regulation.