Randomised or quasi-randomised controlled trials of exercise (any form of physical activity or training) or self-management interventions for patients with knee osteoarthritis, were eligible for inclusion. Studies could not use medication as treatment for knee osteoarthritis, and had to report pre- and post-intervention quantitative assessments of physical and/or psychological health. Psychological, physical (including pain), and physiological (e.g. walking/balance tests) were the outcome categories of interest.
The mean age of participants was 65.8 years. Most studies either compared exercise to self-management and/or control, or compared different types of self-management. Exercise programmes varied in the type and intensity of workouts, and self-management programmes also varied greatly. Control groups received no treatment, standard care, attention control, or sham electric stimulation. Duration of intervention ranged from a single session to multiple sessions over two years (most lasted around 10 weeks, with weekly one to two hour sessions). A variety of self-report scales were used to assess outcomes.
The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.