The Science of Placebo, to Understand the Importance of Placebo Effects in OA:

A placebo is an inert substance that produces beneficial therapeutic effects. In the context of clinical trials, the placebo response refers to an improvement in symptoms in study participants assigned to placebo treatment. High placebo responses pose a substantial impediment to distinguishing active treatment effects and may in part account for the failure to identify effective treatments in clinical trials in osteoarthritis (OA) and other conditions. The treatment response to placebos has been to linked to an effect on opioid receptors in the brain. In addition, placebo responses in trials have been explained by regression to the mean, in which patients enroll in trials when their condition is flaring and the natural history is for their flare to resolve, leading to a natural improvement in their symptoms. Also, positive expectations of a therapeutic effect contribute to placebo response. Higher placebo effects have been linked to reduced treatment effects where treatment effect is defined as (treatment effect - placebo effect). In meta-analyses of OA trials, investigators at the University of Nottingham showed that placebo (so-called contextual) effects of OA treatments often accounted for more than half of the total treatment effect (treatment effect + placebo effect) for pain reduction among proven treatments. Investigators from Tufts University showed that among different modes of administration, placebo responses were greater for injected and topical than for oral treatments. In OA and other painful conditions, variability in pain scoring and initial high pain scores may also be predictive of large placebo effects.

Given the high placebo (or contextual) effects seen in OA, what should be done? This depends on the goal of the enterprise. If the goal is to enhance the patient response to treatment, the best approach is to maximize this response by encouraging a positive expectation for treatment and by choosing treatment modalities likely to maximize this response, injected or topical treatments. Further, such advice may need to be incorporated into treatment guidelines. On the other hand, if the goal is to optimize the likelihood of detecting the efficacy of a treatment, then minimizing placebo responses is necessary. This can be accomplished by providing no positive expectations for treatment and by minimizing pain reporting variability.

Conclusions: Placebo responses to treatment are critical elements contributing to treatment response. We need to consider them in both clinical treatment approaches and in testing potential treatments in OA.