Online exercise shows benefits, with uncertainty about some outcomes
Online programs improved several outcomes compared with no exercise. Comparisons with in-person exercise were less definitive, despite the abstract's claim of noninferiority.
Based on the published abstract. The full paper may contain additional methods, results and limitations.
The 30-second takeaway
Online exercise produced favorable results for several physical and mental health measures compared with no exercise. Evidence certainty was low to moderate, and some estimates remained compatible with no difference. Most comparisons with face-to-face exercise were not statistically significant, which alone does not establish that the formats are equivalent or equally effective.
The authors report few studies and substantial differences between trials for some outcomes. This abstract-based account cannot establish equivalent benefits between delivery formats, and the endurance and depression estimates remain inconclusive.
What researchers found
Pooled strength outcome
Online exercise versus no exercise; follow-up timeframe was not reported. The pooled confidence interval excluded no difference, but the abstract does not express the improvement in everyday strength units.
Benefits differed by endpoint
Strength, balance, physical activity and several wellbeing measures favored online exercise. Endurance and depression estimates were less conclusive. The review therefore supports an outcome-specific reading rather than assuming every measured benefit was established.
Comparing delivery formats
Gait speed improved more with face-to-face exercise. For other comparisons, the abstract describes nonsignificant differences as noninferiority, but supplies no noninferiority margin. Those results alone cannot confirm that online delivery preserves the same benefits.
The original publication
The Effectiveness of Online Exercise on Physical Activity, Motor Function, and Mental Health: Systematic Review and Meta-Analysis.
Bhundoo AK, Pillay JD, Wilke J
J Med Internet Res · 2025
- PubMed ID
- 40815775
- Record checked
AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.
One more question, understood.
Keep track of the research you’ve explored.