Umbrella review finds mixed benefits from digital health interventions

Researchers searched six databases through April 2025 for meta-analyses of randomised controlled trials comparing digital health interventions with inactive controls or standard therapies. The umbrella review included 232 publications, 1,133 unique outcomes and 559,792 participants across 12 condition categories. The team assessed review quality, evidence certainty and credibility, and reanalysed results using random-effects methods.

Against inactive controls, high-certainty findings included lower systolic blood pressure with telemonitoring for hypertension (Hedges’ g 0.55, 95% CI 0.31–0.78), lower diastolic blood pressure with telephone interventions (0.17, 0.09–0.24), reduced depression with internet-based cognitive behavioural therapy in insomnia (0.38, 0.23–0.54), and lower weight with telemedicine in type 2 diabetes (0.12, 0.06–0.17). Specific virtual-reality interventions improved upper-limb function after stroke compared with standard therapies (0.23, 0.06–0.40).

Other high-certainty comparisons found no significant advantage over standard care for several outcomes, including hospitalisation, mortality or stroke in pacemaker patients, HbA1c in type 2 diabetes, and physical function in knee osteoarthritis. Review quality varied: 37% of included publications were rated high quality, while 19% were critically low. The findings indicate that effects depend on the intervention, comparator and outcome; they do not establish that digital approaches are generally superior to usual therapies.