The results took them by surprise.

They found that the patients in the physical therapy group improved – but they did no better than the placebo group did; in fact, people receiving the sham treatments fared slightly better. Pain scores at the start of the study for both the active group and sham group were near the middle of the pain scale: 58.8 and 58.0, respectively. At 13 weeks, both groups had significantly lower pain scores: 40.1 for the active group and 35.2 for the sham group.

There were similar improvements in physical function. The physical function score for the active group was 32.3 at baseline and 27.5 at 13 weeks; sham group scores were 32.4 and 26.4, respectively.  

After nine months, about the same number of people in both groups reported improvements in pain and function. The groups did about the same on other measures, too, although the active group had slightly better balance but more periods of increased pain and stiffness (41 percent in the active group vs. 14 percent in the sham group).

“Unfortunately, as a [physical therapist], the results were not what we had hoped for; both groups showed the opposite of what we had hypothesized,” Bennell says. “But although there were no differences between them, both groups showed benefit, and we know that people who don't receive treatment don’t improve.”

So what caused similar rates of improvement among people who received real and sham treatments?

It’s possible that the active physical therapy program was “truly ineffective,” Bennell says. But another explanation is that combining exercise and manual therapy in a single session might reduce the effectiveness of both. Alternatively, the therapy used in the study might not have targeted the right muscles to improve strength and range-of-motion in arthritic hips.

Bennell says it’s possible the benefits were due not to the therapies themselves but to participants’ expectation of improvement as well as 12 weeks of close contact with sympathetic, caring therapists, both of which are known to influence healing. “Seeing a physical therapist is likely to lead to benefits, but not in the way we thought it was,” she says.

Kathleen Mangione, PhD, a physical therapy professor at Arcadia University in Philadelphia, says she believes the study was well constructed but that it may be saying more about the type and amount of exercise studied than the effectiveness of physical therapy overall.

“I don’t think we can say physical therapy isn’t effective. I think we can say four specific manual therapy techniques and a home-based program of unsupervised and infrequent stretching were not more effective than the sham therapy,” she says. “Physical therapy is a complex biobehavioral intervention and to lump all of physical therapy together – especially when the only physical therapy in this study were four manual therapy techniques – is oversimplifying what physical therapy is and what PTs do.”

David Mayman, MD, an orthopaedic surgeon at Hospital for Special Surgery in New York who was not involved in this research, says despite the findings, there is still a role for physical therapy in hip OA management. “We treat individual patients and cannot always generalize. There will be patients with osteoarthritis of the hip who do benefit from the manipulations and exercise used in physical therapy,” he says.

If that fails, he says, “There is a broad range of options that all have varying risks and benefits, including anti-inflammatory medications, nutritional supplements, physical activity, corticosteroid injections and hip replacement surgery.”