Cardiac rehabilitation-induced improvements in step outcomes: a harmonized dataset of 2,535 patients with cardiovascular diseases and 8,150 controls

Abstract Background To develop effective step count-based physical activity recommendations for patients with cardiovascular disease (CVD), it is essential to understand the acute and long-term impact of cardiac rehabilitation (CR) on step characteristics and how these differ from individuals without CVD. These recommendations are a promising strategy to promote an active lifestyle during and beyond CR as patients with CVD can easily monitor step goals with consumer devices (i.e. smartwatches, smartphones). Purpose We assessed the association of CR and accelerometer-derived step outcomes in patients with CVD and compared these findings with controls without CVD. Methods We harmonised accelerometery data and participant characteristics from 6 longitudinal and 3 cross-sectional studies. Raw accelerometer data were reanalysed if participants wore the accelerometer for ≥4 days including 1 weekend day. We calculated total step volume and time spent at moderate-to-vigorous intensity step cadence (minutes at ≥100 steps/min). In the longitudinal CVD cohorts, we assessed the association of CR and step outcomes from pre-CR to directly post-CR and CR maintenance (≥2 months post-CR) with a linear-mixed model, whilst adjusting for differences across CR programmes. We compared differences in step characteristics between patients with CVD (≥2 months post-CR or a history of CVD) and controls without CVD with a linear model, adjusting for wear location, age, sex, body mass index, smoking, and diabetes mellitus. Results The total dataset consisted of 2,535 patients with CVD (62±9 years old; 34% female) and 8,150 controls (60±9 years old; 48% female). Longitudinal data on CR-induced changes in step characteristics were available for 1,226 patients with CVD. From pre- to post-CR, step volume increased by 1,011 ((95% confidence interval (CI): 846; 1,178) steps/day; 5,974±2,734 versus 7,087±2,854 steps/day), without further changes from post-CR to CR maintenance (6,957±3,040 step/day; p=0.24). Moderate-to-vigorous intensity step time increased by +2 (95% CI: +1; +3) min/day from pre- to post-CR (7 2-16 versus 9 4-19 min/day), but reduced -2 (95% CI: -2; -1) min/day from post-CR to CR maintenance (7 2-17 min/day) and returned to pre-CR levels. Patients with CVD took fewer daily steps (-940 (95% CI: -1,146; -733) steps/day) and spent less time at moderate-to-vigorous intensity stepping (-2 (95% CI: -3; -1) min/day) compared to controls (8,294±3,784 versus 10,471±3,796 steps/day; 11 4-24 versus 19 10-32 min moderate-to-vigorous intensity stepping/day, respectively). Conclusion CR is associated with a sustainable increase in step volume, but step volume and step intensity remained significantly lower compared to values observed in controls without CVD. These findings highlight the need for dedicated interventions to increase step count, especially at a moderate intensity, during and beyond CR to improve clinical outcomes in patients with CVD.