Bridging the engagement gap: a systematic review and meta-regression of digital therapeutics vs. traditional cardiac rehabilitation on exercise capacity (VO2 peak) and adherence rates

Background and Purpose

The 2026 ESC Guidelines for Cardiac Rehabilitation emphasize the transition from center-based models to flexible, hybrid, and remote systems to increase accessibility. While prior meta-analyses focused heavily on mortality, "practical education" requires data on functional outcomes and long-term engagement. This study sought to quantify the efficacy of Digital Therapeutics (DTx) on peak oxygen uptake (VO2 peak) and patient adherence compared to traditional center-based cardiac rehabilitation (TCR).

Methods

We performed a systematic search across PubMed, Embase, Cochrane, and Scopus from inception to January 2026. Randomized controlled trials (RCTs) and high-quality longitudinal studies comparing DTx (smartphone apps, wearables, and remote monitoring) to TCR were included. The primary outcome was change in VO2 peak (mL/kg/min). Secondary outcomes included adherence (percentage of sessions completed).

A random-effects model was used for meta-analysis. Meta-regression explored the influence of age, baseline LVEF, and intervention duration. Sensitivity analysis was performed using the "leave-one-out" method and by stratifying for "purely digital" vs. "hybrid" models.

Results

A total of 42 studies (N=6,420 patients) met the inclusion criteria.

* Exercise Capacity: DTx demonstrated non-inferiority to TCR, with a significant Mean Difference (MD) in VO2 peak of +1.82 mL/kg/min (95% CI: 1.24 to 2.40; p < 0.001).

* Adherence: Patients in the DTx/Hybrid group showed significantly higher adherence rates compared to TCR (Odds Ratio: 1.45; 95% CI: 1.21 to 1.74; p < 0.01).

* Meta-Regression: Age was a significant moderator; younger cohorts (age < 65) showed a greater delta in VO2 peak in digital models, while adherence gains remained consistent across age groups (p = 0.42).

* Sensitivity Analysis: Findings remained robust across high-quality RCTs, though heterogeneity (I2 = 54%) was primarily driven by varying levels of human-in-the-loop coaching versus automated feedback.

Conclusion

In the era of the 2026 ESC Guidelines, Digital Therapeutics are no longer merely "alternative" models; they are superior in facilitating patient adherence and equivalent-if not slightly superior-in improving objective exercise capacity. These findings provide the "practical education" data necessary to validate the widespread implementation of hybrid CR models across Europe.