Does the Choice of Stepping Intensity Metric Influence Dose-Response Associations with Mortality? Analysis on UK...

Evidence on the potential mortality gain of higher free-living stepping intensity is limited and equivocal, potentially due to the inconsistent usage among various estimation metrics. To estimate the difference in the association with mortality risk across different stepping intensity metrics, we aimed to compare different metrics in terms of their multivariable-adjusted associations with all-cause, cardiovascular disease, and cancer mortality.

This cohort study included UK Biobank participants wearing wrist-worn accelerometers. We included eight peak cadence metrics, defined as the highest averaged steps/min across eight different time windows (1-, 5-, 10-, 15-, 20-, 25-, 30-, and 60-min), and two nonpeak cadence metrics including average daily cadence (steps/accelerometer wearing minutes) and purposeful cadence (averaged steps/min for minutes ≥40 steps). For each metric, we first standardized each individual's absolute cadence using (individual's absolute cadence - mean cadence)/standard deviation. We then estimated their dose-response associations using Cox-restricted-cubic-spline models and compared them on overlay plots.

Among 65,253 participants (mean age: 61.5 yr [standard deviation: 7.8]; 57% female) followed for 8.0 (median) yr, all peak cadence metrics and the average daily cadence exhibited similar positive dose-response associations with mortality. For example, the medians of the individual-level standardized cadence and hazard ratios (HRs) across peak 1-, 30-, and 60-min cadence were: all-cause mortality, -0.17 steps/min (HR: 0.71 [95% confidence interval (CI): 0.64-0.80]), -0.15 (0.66 [0.59-0.74]), and -0.15 (0.66 [0.59-0.75]), respectively; cardiovascular disease mortality, -0.17 steps/min (HR: 0.63 [95% CI: 0.51-0.78]), -0.15 (0.57 [0.46-0.71]), and -0.15 (0.57 [0.46-0.71]); cancer mortality, -0.17 steps/min (HR: 0.88 [95% CI: 0.75-1.03]), -0.15 (0.89 [0.75-1.04]), -0.16 (0.93 [0.78-1.09]). Purposeful cadence was not associated with mortality (e.g., all-cause mortality: median of the individual-level standardized cadence, 0.59 steps/min; HR, 0.99 [95% CI: 0.86-1.15]).

This study suggested that peak cadence and average cadence metrics can be used interchangeably to quantify the associations of stepping intensity with long-term health outcomes.