Accuracy of Clinical Tests to Identify Hip Arthritis

Hip arthritis is a common problem linked with aging. Adults in their mid-50s are often the first to notice changes that suggest degenerative joint disease. Early diagnosis and treatment are the keys to staying active and healthy.

In this study, a group of physical therapists (PTs) from the U.S. Army-Baylor program evaluated the accuracy of clinical tests commonly used with patients who have unilateral (one sided) hip pain. The goal was to find a way to predict which patients are more likely to have hip osteoarthritis (OA) based on the results of those tests. This is referred to as a prediction rule. In other words, when certain tests are positive, hip OA is probably present.

Since physical therapists often see patients with unilateral hip pain, it’s important to know which tests to use for an accurate, reliable diagnosis. A prediction rule could help cut down on the time it takes to identify (or rule out) hip arthritis. Patients could get the help they need sooner.

The patients in this study were adults from Brooke Army Medical Center (BAMC) over the age of 40. Pain in the buttock, groin, or front of the thigh was a requirement to be in the study. Each person (78 total) was examined by the physical therapist using commonly applied clinical tests and measures. These included hip joint range-of-motion, the squat test, the scour test, and Patrick’s test. There were a total of five tests.

Pain level was recorded for each patient before and after each test. X-rays were taken to show evidence of OA. The results of the clinical tests were compared with the results of the X-rays. In this way, the therapists could look and see which tests were most likely to point to a diagnosis of OA. Finding a successful prediction rule may help patients avoid X-rays when they aren’t needed.

The authors report that 30 per cent of the 78 patients had arthritis based on X-ray findings. Within this group with positive X-ray indications of arthritis, four out of the five clinical tests were also positive. That worked out to be a 91 per cent probability that the patient had arthritis. Analysis of the data showed that when three of the five tests were positive, the likelihood of OA increased to 68 per cent. This was significant but not quite as predictable as when four of the five tests were positive.

Before these tests can be used as a clinical prediction rule, a larger study with more people is needed. Further examination of the tests is also needed. There may be other tests that could serve more effectively as a prediction rule. The next step is to find the best combination of tests to rule out OA (rather than rule in arthritis).