The Use of Placebos in Clinical Trials and Clinical Practice
Kirsch, Irving, Canadian Journal of Psychiatry
These 2 In Review articles1,2 provide useful contributions to understanding the use of placebos in clinical trials and clinieal practice. They also reveal opposing trends: the rise of intentional placebo use in clinical practice and the dissatisfaction with its use in clinical trials. In this commentary, I take an opposite track: placebo controls need to be strengthened in clinical trials, and alternatives to placebo use need to be found in clinical practice.
Following an excellent and illuminating history of the use of placebos in clinical trials, Dr Edward Shorter1 argues that "for truly effective drugs, RCTs [randomized controlled trials] are probably not indispensable ... if the drug clearly works, vast and expensive trials would seem unnecessary,"p 196 and he notes that chlorpromazine and Imipramine were approved based on "clinical observation of patients whose improvement was undeniable and as plain to see as the betterment of patients on penicillin."p196
Actually, Imipramine provides a telling example of why clinical trials are necessary in evaluating psychiatric drugs. As a treatment for depression, Imipramine is no better than selective serotonin reuptake inhibitors (SSRIs) or any other class of antidepressants (ADs),3 which means that it is not much better than a placebo, even for most patients with very severe depression.4,5 RCTs have shortcomings, but they are far better than clinical observation. Relying on clinical observation would be a step backwards, opening the door to the approval of many new drugs with no therapeutic benefits beyond the placebo effect. We need to strengthen clinical trials, not abandon them.
The main problem with clinical trials is not that they are expensive, but rather that they are too lax. The trials are supposed to be double blind, but psychiatric drugs produce substantial side effects leading patients and doctors break blind. In one study, 89% of patients and 90% of doctors accurately guessed that they were in the active drug condition.6 Thus the true drug effect of ADs may be nonexistent, and the small apparent effect might be an enhanced placebo effect.7 Supporting this contention is that SSRIs are therapeutically indistinguishable from selective serotonin reuptake enhancers.8 In other words, drugs with opposing chemical mechanisms have the exact same clinical effects - hardly what one would expect were the effects owing to the chemical composition of the drugs.
As Dr Shorter1 notes, the use of active comparators has been increasingly advocated as a replacement for conventional, placebo-controlled RCTs. This, too, would represent a step backwards. The problem is that finding new ADs that are significantly more effective than existing ones has proven impossible and is not the goal of comparator trials. Instead, these trials are aimed at showing that a new drug is equivalent to an existing drug. The problem, of course, is that both may be placebos. The inclusion of placebos in tests of psychiatric drugs is indispensible, but the methodology of these trials needs to be strengthened so that the double blind is maintained. At the very least, assessment of the integrity of the blinding, which would add nothing to the cost of clinical trials, should become a routine requirement of RCTs.
The data reported by Dr Amir Raz and colleagues2 indicate that a substantial number of Canadian physicians prescribe placebos, and an even larger number prescribe what he calls pseudoplacebos, which he defines as "placebolike interventions that may be active in principle but unlikely effective for the condition being treated.""204 Pseudoplacebo is a lovely term, in as much as placebos might be defined as pseudotreatments. The problem is that, for the most part, these placebos are administered deceptively, a practice that is ethically questionable and carries the risk of underlying the trust that is essential to the doctor-patient relationship. …