TOPLINE

Land-based exercise interventions for hip osteoarthritis (OA) yielded modest reductions in pain and improvements in physical function compared with no treatment or usual care, and their effects “are unlikely to be clinically meaningful,” according to findings from a Cochrane Collaboration systematic review.

METHODOLOGY

• Researchers updated a Cochrane review first published in 2009 and last updated in 2014, including 18 randomized controlled trials (eight new trials) from February 2013 to 2025 evaluating land-based exercise in adults with symptomatic hip OA diagnosed using clinical criteria or self-reported chronic hip joint pain.
• A total of 1368 participants (mean age range, 53-73 years; 63% women; mean BMI range, 24.7-30.5) were included; the studies compared exercise vs attention control or placebo (two studies, 123 participants); exercise vs no treatment, usual care, or limited education (10 studies, 494 participants); and exercise plus a nonsurgical co-intervention vs the co-intervention alone (seven studies, 751 participants). Meta-analyses were conducted separately for each comparator category.
• Exercise interventions varied considerably in type (eg, aerobic, resistance, neuromuscular, mixed, mind-body, and stretching), duration (2-52 weeks), frequency, and delivery method (supervised or unsupervised, individual or group, and in-person or online).
• The critical outcomes assessed were severity of hip pain, physical function, and/or quality of life measured immediately after treatment, at 3-6 months, and after 6 months of cessation of treatment, with follow-up ranging from 2 weeks to 24 months. Only a subset of studies reported outcomes at 3-6 months and after 6 months.

TAKEAWAY

• Exercise compared with attention control or placebo may have improved physical function (mean difference [MD], -7.44 points; 95% CI, -13.86 to -1.01; low-certainty evidence), with the CI of the MD crossing the threshold for a clinically important improvement of 13 points for physical function.
• Exercise compared with no treatment, usual care, or limited education probably reduced pain slightly immediately after treatment (MD, -7.19 points; 95% CI, -10.70 to -3.68; nine studies, 449 participants; moderate-certainty evidence) and probably improved physical function slightly (MD, -8.79 points; 95% CI, -12.00 to -5.41; nine studies, 447 participants; moderate-certainty evidence), although these improvements did not meet the threshold for clinically important differences of 12 points for pain and 13 points for physical function.
• Exercise plus a co-intervention compared with the co-intervention alone probably had little to no effect on pain, physical function, or quality of life immediately after treatment, with moderate-certainty evidence for all. However, exercise combined with another intervention was probably associated with a reduced risk for adverse events compared with the co-intervention alone (risk ratio, 0.75; 95% CI, 0.58-0.97; six studies, 731 participants; moderate-certainty evidence).
• Little to no effect was noted for exercise vs no treatment, usual care, or limited education and for exercise plus a co-intervention vs the co-intervention alone on quality of life after treatment (moderate-certainty evidence) and on participant-reported treatment success (low-certainty evidence). Little to no effect was also noted across all three comparator categories for study withdrawals.

IN PRACTICE

“Exercise is often prescribed as part of a broader physiotherapy program in clinical practice, and our findings are only applicable to the effect of exercise alone. Thus, caution is necessary when extrapolating to combined clinical programs,” the authors wrote.

“There just isn’t a huge body of evidence out there,” said Belinda Lawford, co-lead author from the University of Melbourne, Melbourne, Australia, in a news release. “For some people struggling with hip pain, exercise can really be their only hope, but I also don’t want to give patients false hope. It’s important future research is done with larger, better-quality trials, examining what types of exercise work specifically for different people.”

SOURCE

The study was led by Michelle Hall, Musculoskeletal Research Hub, Charles Perkins Centre, University of Sydney, Sydney, Australia, and Belinda J. Lawford, Centre for Health Exercise and Sports Medicine, University of Melbourne. It was published online on July 23, 2026, in the Cochrane Database of Systematic Reviews.

LIMITATIONS

More than half of the studies had high risk for performance and detection bias; thus, the evidence certainty was downgraded across all comparators due to risk for bias. The findings may not be applicable to younger adults with hip OA. Most studies assessed outcomes immediately after treatment, with fewer trials evaluating quality of life, participant-reported treatment success, and adverse events, thereby limiting certainty about these outcomes.

DISCLOSURES

The study had no direct funding. Three authors were supported by National Health and Medical Research Council Investigator Grants. Some authors disclosed receiving grant funds/research funding, royalties, or travel expenses for meetings from various sources or serving as a board member or committee member of various councils or groups.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.