Abstract
<sec> <title>BACKGROUND</title> <p>Background: Exercise rehabilitation is a core component of knee osteoarthritis management, but sustained delivery may be constrained by time, geography, and limited health care resources. Digital technologies may improve access to prescribed exercise and support its delivery and monitoring; however, their overall effectiveness and the factors influencing treatment response remain uncertain.</p> </sec> <sec> <title>OBJECTIVE</title> <p>Objective:To systematically evaluate the effects of digital technology–supported exercise rehabilitation on pain, patient-reported function, performance-based physical function, and quality of life in people with knee osteoarthritis, and to investigate potential moderators of treatment effects.</p> </sec> <sec> <title>METHODS</title> <p>Methods: PubMed, Web of Science, Embase, EBSCO, the Cochrane Library, China National Knowledge Infrastructure, and Wanfang Data were systematically searched from inception to May 20, 2026. Randomized controlled trials were eligible if they evaluated exercise-based rehabilitation supported by digital technologies, including telerehabilitation, mobile applications, or virtual reality. Three-level random-effects models were used to account for dependencies among multiple effect sizes arising from different outcomes, measurement instruments, or assessment time points within the same study. Statistical inference was further supported by cluster-robust variance estimation with CR2 small-sample correction. Subgroup analyses and meta-regression were conducted to examine the potential moderating effects of knee osteoarthritis severity, type of digital technology, comparator condition, sensor use, intervention duration, and weekly intervention frequency. The certainty of evidence was assessed using the GRADE approach.</p> </sec> <sec> <title>RESULTS</title> <p>Results: Twenty randomized controlled trials involving 1,222 participants with knee osteoarthritis were included. Three-level meta-analyses showed that digital technology–supported exercise rehabilitation significantly reduced pain (Hedges’ g = 0.60, 95% CI [0.15, 1.06], 95% PI [−1.29, 2.50]) and improved patient-reported function (Hedges’ g = 0.72, 95% CI [0.24, 1.20], 95% PI [−1.42, 2.87]), performance-based physical function (Hedges’ g = 0.54, 95% CI [0.11, 0.97], 95% PI [−0.99, 2.07]), and quality of life (Hedges’ g = 0.51, 95% CI [0.16, 0.85], 95% PI [−0.46, 1.48]). These effects remained statistically significant after CR2 small-sample correction. Subgroup analyses indicated greater improvements in quality of life among participants with mild-to-moderate knee osteoarthritis. Meta-regression further showed that longer intervention duration was associated with greater pain reduction and improvements in quality of life, whereas a higher weekly intervention frequency was associated with greater improvements in patient-reported function. Sensitivity analyses supported the robustness of the findings for patient-reported function and quality of life. Although the pooled estimates consistently favored digital technology–supported exercise rehabilitation, all 95% PI crossed the line of no effect, indicating substantial variability in the treatment effects that may be observed across future clinical settings.</p> </sec> <sec> <title>CONCLUSIONS</title> <p>Conclusions: Digital technology–supported exercise rehabilitation may improve pain, function, and quality of life in people with knee osteoarthritis. Its principal clinical value may lie in expanding access to exercise rehabilitation and supporting the implementation of prescribed exercise. however, the optimal mode of delivery remains to be established through further high-quality research.</p> </sec> <sec> <title>CLINICALTRIAL</title> <p>Trial Registration: PROSPERO CRD420261404719; https://www.crd.york.ac.uk/PROSPERO/view/CRD420261404719</p> </sec>