Paradoxically, it is easier to perform a sham trial in surgery, the most invasive physical act, than in physiotherapy because the patient is asleep when it is delivered. Physical therapy involves physical acts that are hard to imitate as placebo treatment, but the influence of the patient-therapist interaction makes it important to tease out any placebo effect. Researchers have, however, performed sham trials in physiotherapy.
A study from JAMA in 2014 (here) tested a 12 week multimodal physical therapy program (combined with home exercises) for patients with hip osteoarthritis, against sham physiotherapy involving an inactive ultrasound machine and some inert gel (combined with home application of the gel). Even though patients tended to guess what group they were in (those getting real physiotherapy thought they were getting the real treatment), the results still showed no difference in pain and function scores at 24 weeks, with the pain scores slightly favouring the sham group. There were also less adverse events in the sham group. It seems that physiotherapy hurts, and provides no benefit beyond placebo.
The same group did a similar study of knee osteoarthritis published in 2005 (here), also showing that the active multimodal treatment was not significantly better than sham treatment.
These are important studies because they tease out the effect of the interaction with the therapist, an important factor in getting patients to feel better. A review (here) of the effect of this ‘treatment alliance’ between the patient and the therapist has shown that it positively affects patient outcomes.
The placebo effect in studies of osteoarthritis has been previous summarised (here), concluding that placebo treatment is better than no treatment, particularly in patients with more severe symptoms and if the intervention is ‘active’, with treatments involving needles doing best.
The bottom line
Sham studies of physical therapy for hip and knee osteoarthritis show no benefit over placebo. Given that open (not blinded or sham) studies of physical therapies often show only marginal benefits (often less than what is considered clinical meaningful), perhaps we should be looking at doing more sham studies of physical therapies. And maybe doing a little less physical therapy in the meantime.