THE EFFECTIVENESS OF EXERCISE BASED DIGITAL HEALTH INTERVENTIONS (REQUIRING INTERNET) IN MANAGEMENT OF HIP AND KNEE OSTEOARTHRITIS: A SYSTEMATIC REVIEW AND META-ANALYSIS

ABSTRACT

Purpose: Osteoarthritis (OA) has a disproportionately significant effect on social, economic and healthcare domains. Systematic reviews and clinical guidelines indicate strong evidence for the management of OA through targeted exercise and education interventions. Throughout the COVID-19 pandemic, there has been a rapid increase in digital health utilization which has the potential to provide accessible, cost-effective and time-effective care. The aim of this study is to investigate the effectiveness of digitally delivered exercises in management of OA of the hip or knee. Methods: Databases of PubMed, PEDro, EMBASE & CINAHL were searched with the concepts of “Osteoarthritis”, “Digital” & “Randomised Controlled trials”. Synonyms within each concept were search with the OR operator, and combined between concepts with the AND operator. Titles and s were screened for eligibility by two reviewers [AS1] and conflicts were resolved by consensus, or a third reviewer. Articles were screened according to the following criteria Participants- people with hip or knee OA;Intervention- had to include a form of exercise prescription being delivered digitally (requiring internet);Comparison- no intervention or comparison intervention;Outcome- pain was the primary outcome for this review. Full-texts of potentially eligible articles were screened by one reviewer and checked by a second. Means, standard deviations or 95% Confidence Intervals (CI) of post-test results were used to calculate standardised mean differences (SMDs). Data was stratified by time points, mode of delivery and OA type (knee or hip). Data were pooled in a random effects meta-analysis where two or more studies were identified within a subgroup. Results: 7327 studies were identified through database searching. After title and , and full-text screening, 10 studies were included within this systematic review. There was a total of 3,402 participants included within this meta-analysis. Studies included participants with hip or knee OA (n=3 studies), knee OA only (n=7 studies). There were no studies on hip OA only. Interventions ranged from exercise delivered via web-based, app-based and video-conferencing, with comparisons including waitlisting, usual care or usual physiotherapy, alternate modes of delivery, and education only via digital means. Overall Pain Results: Digitally delivered exercise had significantly better outcomes for pain at short term (≤3 months) SMD and 95% CI [-0.44 (-0.73, -0.15)], medium term (3 to 6 months [-0.24(-0.46, - 0.02)]) and long term (>6 months [-0.20 (-0.38, -0.03)]). Knee Osteoarthritis only 7 studies reported on knee OA. Digital exercise interventions were superior to the comparison at short term (95%CI -0.68(-1.11, -0.24)), and long term (95%CI -0.30(-0.49, -0.12). There was no difference between digital exercise intervention and the comparison at the medium term (95%CI -0.29(-0.60, 0.02) time-point. App Based intervention 4 studies used an app based digital intervention to deliver exercise prescription. App based was superior to the comparison in the short term [-0.45(-0.79, -0.10)], but there was no difference at the medium term [-0.12(-0.34, 0.10)] Web Based intervention 5 studies used a web-based platform to deliver exercise interventions. These were superior to the comparison for the outcome of pain short term [-0.16(-0.32, -0.01)], but there was no difference in the medium [95%CI -0.29(-0.60, 0.02], and long term [95%CI -0.16(-0.33, 0.02)]. Conclusions: The data indicates that exercises delivered digitally are more effective than any comparisons including waitlisting, usual care or physiotherapy, alternate modes of delivery, or education component only in management of hip and knee osteoarthritis. There is robust evidence to suggest an increased effect within the first 3 months of implementation across all modes of delivery (Video, Web or App-based). Improvements in pain symptoms is an important outcome as this can be associated with improvement in function, health, and returning to activiti s of daily living. Additionally, pain is the predominant factor for seeking healthcare, so improvements within community-dwelling patients with OA may reduce burdens on hospital and health services, and subsequent costs for hospital admissions due to OA. The significant benefit of a digital service which provides exercises for patients with OA is the ability to inform, educate and promote exercise with patients from the convenience of their own home regardless of geographical proximity to a health service. This would suggest that this can be a highly-cost effective method of delivering high-quality care across a multitude of regions & population groups. Given that exercise delivered digitally is not-inferior to usual care, alternate modes of delivery of exercises, or education only, the uptake of digital care on the future of management in OA must be considered. Further research may be required on user engagement with technology to determine optimisation of delivery service.