Temporal changes in personal activity intelligence and mortality: Data from the aerobics center longitudinal study

Abstract

Background: Personal activity intelligence (PAI) is a metric developed to simplify a physically active lifestyle for the participants. Regardless of following today's advice for physical activity, a PAI score ≥100 per week at baseline, an increase in PAI score, and a sustained high PAI score over time were found to delay premature cardiovascular disease (CVD) and all-cause mortality in a large population of Norwegians. However, the association between long-term temporal change in PAI and mortality in other populations have not been investigated.

Objective: To test whether temporal change in PAI is associated with CVD and all-cause mortality in a large population from the United States.

Methods: We studied 17,613 relatively healthy participants who received at least two medical examinations in the Aerobics Center Longitudinal Study between 1974 and 2002. The participant's weekly PAI scores were estimated twice, and adjusted hazard ratios (AHR) and 95% confidence intervals (CI) for CVD and all-cause mortality related to changes in PAI between baseline and last examination were assessed using Cox proportional hazard regression analyses.

Results: During a median follow-up time of 9.3 years [interquartile range, 2.6-16.6; 181,765 person-years], there were 1144 deaths, including 400 CVD deaths. We observed an inverse linear association between change in PAI and risk of CVD mortality (P=0.007 for linear trend, and P=0.35 for quadratic trend). Compared to participants with zero PAI at both examinations, multivariable-adjusted analyses demonstrated that participants who maintained high PAI scores (≥100 PAI at both examinations) had a 51% reduced risk of CVD mortality [AHR, 0.49: 95% CI, 0.26-0.95)], and 42% reduced risk of all-cause mortality [AHR, 0.58: 95% CI, 0.41-0.83)]. For participants who increased their PAI scores over time (PAI score of zero at first examination and ≥100 at last examination), the AHRs were 0.75 (95% CI, 0.55-1.02) for CVD mortality, and 0.82 (95% CI, 0.69-0.99) for all-cause mortality. Participants who maintained high PAI score had 4.8 (95% CI, 3.3-6.4) years of life gained. For those who increased their PAI score over time, the corresponding years gained were 1.8 years (95% CI, 0.1-3.5).

Conclusion: Among relatively healthy participants, an increase in PAI and maintaining a high PAI score over time was associated with reduced risk of CVD and all-cause mortality.

Condensed abstract: Our objective was to investigate the association between temporal changes in PAI and mortality in a large population from the United States. In this prospective cohort study of 17,613 relatively healthy participants at baseline, maintaining a high PAI score and an increase in PAI score over an average period of 6.3 years was associated with a significant reduction in CVD and all-cause mortality. Based on our results, clinicians can easily recommend that patients obtain at least 100 PAI for most favourable protection against CVD- and all-cause mortality, but can also mention that significant benefits also occur at maintaining low-to-moderate PAI levels.

Keywords: Activity metric; Cardiovascular disease; Exercise; Mortality; Physical activity.