You’ve had a nagging ache on the outside of your hip for months. It flares when you lie on that side at night, when you cross your legs, and when you stand up after a long drive. So what actually fixes it: a cortisone shot, or the exercise program you keep putting off?
The study, briefly
A 2025 systematic review in Clinical Rehabilitation set out to answer that question with a deliberately high bar for evidence. Bremer, Nicklen, Fearon and Morrissey searched five databases from inception through August 2024, and then threw out almost everything they found. Only randomised controlled trials rated medium or high quality on the PEDro scale, judged low risk of bias on the Cochrane Risk of Bias Tool 2.0, and measuring both pain and function were allowed in. Four interventions from four trials survived that filter (Bremer et al., Clinical Rehabilitation, 2025;39(5):600-617).
The condition in question is gluteal tendinopathy, which is the most common reason for persistent pain on the outside of the hip in adults. It used to get labelled “bursitis,” and a lot of people still hear that word. What is usually happening is irritation and degeneration of the gluteus medius and minimus tendons where they attach to the greater trochanter, the bony bump on the side of your hip. Results were grouped into short, medium and long term windows, and effects were reported as standardised mean differences, which is a way of expressing effect size on a common scale so treatments studied with different questionnaires can be compared.
What they found
The headline number: exercise plus education produced the largest short term effect of any conservative option studied, with a standardised mean difference of 0.95 for pain (95% CI 0.58 to 1.33) and 0.91 for function (95% CI 0.53 to 1.28) compared with minimal intervention. The authors graded that as moderate strength evidence and a medium sized effect (Bremer et al., 2025).
- Corticosteroid injection helped, but less. Against the same minimal intervention comparison, a steroid shot produced a small short term pain effect: SMD 0.51 (95% CI 0.16 to 0.86). Real, but roughly half the size of the exercise and education effect.
- Platelet-rich plasma slightly beat steroid for function. PRP edged out corticosteroid on short term function with an SMD of 0.46, but the confidence interval ran from 0.00 to 0.91. When the lower bound of the interval sits exactly on zero, “no difference at all” is still inside the range of plausible results.
- Focused shockwave therapy looked better than steroid for long term pain, but the reported effect size was 5.77 (95% CI 4.84 to 6.71). An effect that large is not something you see in musculoskeletal rehab research. It almost always signals one small trial with unusual numbers rather than a robust finding.
- Benefits shrank over time. Past the short term window, the exercise and education effects on both pain and function dropped from medium to small.
The catch
This is where I want to be straight with you, because the review authors were straight too. Methodological heterogeneity prevented them from doing a meta-analysis. In plain terms, the four qualifying trials were built too differently from one another to be pooled into a single combined estimate. So every number above is an effect size carried over from one individual trial, not an average across a large body of research.
That matters in two ways. First, precision is limited. A single trial’s effect size can move a lot when the next trial is run. Second, that shockwave figure should be treated as an outlier awaiting confirmation, not as a reason to book a series of shockwave sessions. The authors’ own conclusion was that exercise and education can be “cautiously” recommended as the core approach, and that definitive trials are still needed before firm practice recommendations can be made (Bremer et al., 2025).
The other honest limitation: the review does not tell us total participant numbers, and the follow-up windows were described qualitatively rather than in weeks and months. So “short term” is doing some heavy lifting as a category. When you see a review this selective, you are trading breadth for quality, and the cost of that trade is a small evidence base.
What this means for you
The practical read is that the ordering of treatments matters more than the existence of any one treatment. Here is how I use this in the clinic.
Start with loading and education, not with a needle. A structured program for gluteal tendinopathy is not just hip strengthening. It is progressive load to the gluteal tendons, usually starting with isometric holds and side-lying or standing abduction work, then building toward loaded single-leg positions and step and stair mechanics. The education half is not filler. Tendons on the side of the hip get compressed when the thigh crosses the midline, so the specific instructions that matter are things like not sitting cross-legged, not standing hip-hitched with your weight parked on one leg, putting a pillow between your knees if you side-sleep, and not sleeping on the painful side while the tendon is angry. Patients often get more relief in the first two weeks from those position changes than from any exercise.
Use an injection as a window, not as a plan. A steroid shot did produce a small short term pain benefit. If pain is so high that you cannot tolerate the loading program at all, that window has value. What the data do not support is a shot as the whole treatment, because the effect is smaller than exercise and it fades. If you have already had two or three injections in the same hip without lasting change, that is useful information rather than a reason to try a fourth.
Plan for months, and plan for maintenance. The finding that effects shrank beyond the short term is the one people skip past. Tendons respond to load, and they also de-adapt when load stops. This is the same pattern that shows up across tendon research, including the exercise versus shockwave versus advice comparisons I wrote about in my breakdown of tennis elbow treatments. Two or three sessions a week of maintenance hip work after you feel better is not overkill. It is the reason you stay better.
Get the diagnosis confirmed first. Lateral hip pain is not always tendon. It can come from the lumbar spine, from the hip joint itself, or from a combination. If your pain travels down the leg past the knee, or your hip feels stiff and blocked in the morning, the plan changes. That distinction is worth an in-person exam, because it is what separates a program that works from six weeks of well-intentioned side-lying clamshells.
Where to go from here
If you’ve been managing this on your own with stretching and rest and it keeps coming back, the missing piece is usually load progression and the daily positions nobody told you about. That is exactly what an initial evaluation is built to sort out: a full hour, one on one, hands on testing to confirm whether the tendon is actually the driver, and a loading plan you walk out with. If you would rather use insurance for your care, CACC Physical Therapy is the insurance-based option.
References
- Bremer T, Nicklen P, Fearon A, Morrissey D. The efficacy of gluteal tendinopathy treatments: A systematic review. Clinical Rehabilitation. 2025;39(5):600-617. doi:10.1177/02692155251327298. PMID 40223303. https://pubmed.ncbi.nlm.nih.gov/40223303/
Leave a Reply