Introduction: Objective measurement of daily physical activity (PA) is an important predictor of cardiovascular risk among young- and middle-aged adults. However, data are sparse regarding the value of activity monitoring among older adults [[Unable to Display Character: –]] particularly among those with mobility limitations.
Hypothesis: We hypothesized that objectively-measured daily PA would be associated with the incidence of cardiovascular (CV) events among older adults in the Lifestyle Interventions and Independence for Elders (LIFE) study [[Unable to Display Character: –]] a Phase 3 randomized clinical trial comparing a structured exercise intervention (median 2.7 years) to health education among mobility-limited older adults.
Methods: CV events (myocardial infarction, stroke, cardiovascular death, silent myocardial infarction, angina/symptomatic coronary artery disease, peripheral artery disease, abdominal aortic aneurysm) were adjudicated by medical records review. Free-living activity data were collected by hip-worn accelerometer at baseline and 6, 12, and 24 months post-randomization. Linear regression, adjusted for accelerometer wear time, demographic and cardiovascular disease-related covariates, was used to model the relationship of incident CV events with accelerometry variables (steps/day and minutes/day spent in activity registering ≥ 100 or ≥ 500 counts/minute) at baseline and in a time-dependent fashion.
Results: Participants (n = 1618; age 78.9 ± [SD] 5.2 years; 67.2% female) had a cumulative CV event incidence rate of 14.5% (234 events). Event rates did not differ between randomized groups. At baseline, participants registered 2681 ± 1475 steps/day and spent 22.8 ± 8.2% of their time performing activities≥ 100 cts/min and 6.3 ± 4.3% of time in activities ≥ 500 cts/min. At baseline, CV event risk was10% lower for every 500 steps/day (HR: 0.90, 95% CI: 0.84-0.96, p=0.001). Each 30 minutes spent performing activities ≥ 100 cts/min (0.89, CI: 0.83-0.96, p=0.002) and ≥ 500 cts/min (HR = 0.76, CI: 0.65-0.90, p=0.001) was also associated with lower CV event risk. Across all follow-up visits, participants took 2544 ± 1592 steps/day and spent 21.3 ± 8.2% of their time performing activities> 100 cts/min and 5.8± 4.3% of time in activities ≥ 500 cts/min. In time-dependent analyses, both steps/day (HR: 0.92, CI: 0.86-0.97, p=0.004) and duration of activity ≥ 500 counts/min (HR: 0.80, CI: 0.68-0.94, p=0.007) were associated with lower CV event rates. Longitudinal performance of activity ≥ 100 cts/min was not statistically significant (HR: 0.93, CI: 0.87-1.01, p=0.067).
Conclusions: Objective measurements of PA collected via accelerometry predicted incident CV events among older adults with limited mobility. These data support the importance of daily PA in this population and suggest that accelerometry may be a useful tool to aid in assessing the risk of CV events.