Poor adherence is commonly observed as one of the characteristics of exercise interventions and there is little consensus as to whether digital interventions promote exercise adherence in people undergoing physiotherapy, especially in older adults with knee diseases.

To investigate the effects of digital health for improving the exercise adherence and intervention outcomes.

A systematic search was conducted on 4 databases; PRISMA reporting guidelines were followed. Journal articles in English (published till June 30, 2024) reporting digital health on exercise adherence and intervention outcomes for older adults with knee diseases were searched, and eligible articles underwent data extraction and a thematic synthesis.

Of the 1015 potentially relevant trials, 13 studies totaling 1258 participants were eligible for inclusion. Digital health was slightly better but not significant than non-digital health on total adherence (SMD 0.29, 95% CI −0.02 to 0.60; P = 0.07; I² = 77%). At short-term follow-up, digital health improved adherence compared with nondigital health (SMD 0.70, 95% CI 0.39-1.01; P < 0.001; I² = 4%), with a very low certainty of evidence. At mid- and long-term, digital health was no better than non-digital health on adherence (mid-term: SMD 0.01, 95% CI -0.55 to 0.58; P = 0.97; I² = 86%; long-term: SMD 0.09, 95% CI -0.31 to 0.49, P = 0.66; I² = 48%). For the intervention outcomes, digital health reduced the pain only at short-term and was no better than non-digital health on physical function, ADL or QoL at any time point.

The available evidence is insufficient to draw the conclusion that digital health technology improves exercise adherence and intervention outcomes in older people with knee disease, and there is very low to low certainty of evidence supporting improvements of adherence and pain only at short-term. Larger-scale, more reliable studies and strategies are needed to enhance mid- and long-term adherence and intervention outcomes in older adults.