Development of Data-Driven Models for Just-in-Time Digital Self-Management Advice to Improve Physical Functioning in...

Clinical guidelines recommend a stepped-care strategy for patients with hip and knee osteoarthritis that begins with nonoperative approaches, including education, pain medication, and self-care. However, the implementation of stepped-care remains limited. Digital self-management interventions have the potential to support patients in applying lifestyle advice and self-care strategies, but current tools often provide generic support without long-term continuity. Artificial intelligence offers new opportunities to deliver personalized, just-in-time self-management interventions for people with osteoarthritis. The development of such artificial intelligence algorithms is limited due to a lack of rich, longitudinal datasets.

The primary objective of this study is to develop and evaluate data-driven models that support personalized recommendations on the optimal timing and optimal advice (physical activity promotion, sleep optimization, weight management, no program) for individuals with hip or knee osteoarthritis.

This prospective cross-over study aims to include 600 people with hip or knee osteoarthritis, meeting the National Institute for Health and Care Excellence criteria. The study is registered at ClinicalTrials.gov (registered February 2, 2026, NCT07423858). We aim to screen digital health literacy in our sample. Participants will be recruited across the Netherlands and will use the e-cOAch web app, with 3 self-care programs (ie, physical activity promotion, weight management, and sleep optimization). Each participant will complete all three 12-week programs and one 12-week control period in a randomized sequence. Participants will be followed for 12 months, with biweekly assessments conducted via the app. The primary outcome for model development will be deterioration in physical functioning, measured by a decrease in the Hip Disability and Osteoarthritis Outcome Score subscale activities of daily living of 6.7 or the Knee Injury and Osteoarthritis Outcome Score subscale activities of daily living of 8.2. Secondary outcomes will be pain and participation. Additional measures will include patient characteristics (date of birth, sex, height, level of education, comorbidity, ethnicity, use of a walking device, use of pain medication, device for e-cOAch, duration of osteoarthritis complaints, health and digital literacy, smoking, alcohol use), physical activity, sleep quality and insomnia, psychosocial factors, behavioral determinants, and engagement with the app. These data will inform the development of data-driven models using supervised (causal) machine learning.

The funding for the study was granted in 2023. At the time of manuscript submission, 520 participants had been recruited. Recruitment is expected to be completed in March 2026, with data collection projected to conclude in April 2027. The publication of the results is anticipated in spring 2028.

This study will provide data-driven models that forecast changes in physical functioning over time and support personalized recommendations on the optimal timing of specific self-care programs for people with hip or knee osteoarthritis. These models will be integrated into a new iteration of a self-management app, e-cOAch (version 2), to provide personalized support for people with osteoarthritis across varying levels of digital health literacy.