Efficacy of Artificial Intelligence-Assisted Therapeutic Exercise in Patients With Knee Osteoarthritis

Condition(s) studied

Investigational drug(s) / intervention(s)

AI-Assisted Exercise Program: Participants will complete a 6-week home-based therapeutic exercise program delivered through an AI-assisted web application. The program will be individualized and updated based on participant-reported pain and perceived exertion, which will be used to guide exercise dose and progression (e.g., progression/regression, load adjustments, and exercise variants) according to guideline-informed decision rules embedded in the platform. Participants will also attend weekly in-person, arm-specific group sessions (experimental-only) for protocol reinforcement, questions, and adverse event monitoring.

Paper-Based Home Exercise Program: Participants will complete a 6-week home-based therapeutic exercise program provided in paper format (printed exercise materials). The program will follow a conventional, predefined progression consistent with usual care and will not incorporate real-time, individualized adjustments based on symptom/effort inputs. Participants will also attend weekly in-person, arm-specific group sessions (control-only) for protocol reinforcement, questions, and adverse event monitoring.

Study summary

Knee osteoarthritis is a common, chronic degenerative condition that causes pain, functional limitations, and reduced quality of life, especially in adults older than 50 years. Although clinical practice guidelines recommend therapeutic exercise and physical therapy as the first-line treatment, many patients do not receive or adhere to the recommended programs. Digital health tools may help expand access and improve adherence by supporting home-based exercise.

This study is a parallel-group, randomized controlled clinical trial designed to evaluate the effectiveness of an artificial intelligence (AI)-assisted therapeutic exercise program delivered through a web-based app compared with a conventional home exercise program provided on paper in adults with knee osteoarthritis. The primary goal is to determine whether the AI-assisted program produces greater improvements in pain and knee-related function. Secondary goals include assessing changes in health-related quality of life and patient-perceived overall improvement, and comparing implementation outcomes like adherence, usability, satisfaction, self-efficacy for exercise, and use of the prescribed exercise resource (paper vs web platform). Participants will be recruited from the Department of Physical Medicine and Rehabilitation at Hospital Universitario 12 de Octubre and randomly assigned to either the AI-assisted exercise group or the conventional exercise group. The AI-assisted web app will adjust exercise dose and progression based on participant-reported pain and perceived exertion, using decision rules informed by clinical practice guidelines and high-quality research evidence. Both groups will complete a 6-week home-based exercise program and will also attend weekly in-person group sessions that will be held separately by study arm (experimental and control groups will not mix) to reduce cross-contamination and support masking procedures. These sessions will be used to address questions, reinforce the protocol, and monitor for adverse events. Clinical outcomes will include pain intensity (visual analog scale), knee-related disability and function (WOMAC), health-related quality of life (EQ-5D), and patient global rating of change (GROC). Physical performance outcomes (30-second Chair Stand Test and Stair Climb Test) will be assessed at baseline and 1 week after completion of the 6-week intervention. Assessments will be conducted at baseline, one week after the 6-week intervention, and at 3 and 6 months after the intervention, with the exception of physical performance outcomes, which will be assessed at baseline and 1 week post-intervention.

Eligibility

Primary outcome measure(s)

• Pain Intensity — Baseline, 1 week post-intervention, 3 months post-intervention and 6 months post-intervention
  Pain intensity will be assessed using the Numeric Pain Rating Scale (NPRS), a single 11-point numerical rating scale widely validated across different patient populations, where 0 represents no pain and 10 represents the worst possible pain. Data obtained through the NPRS are easy to document, intuitively interpretable and meet regulatory requirements for pain assessment and documentation.

Trial sites (1)

More Universidad Complutense de Madrid trials in Spain

Other trials for the same condition