By Patrick D Hahn
Hot on the heels of the publication of David Healy’s Shipwreck of the Singular comes a paper in Neuropsychopharmacology by Elan Cohen and seven of his colleagues which demonstrates why Shipwreck was needed in the first place.
Seven of the eight authors of the paper are drug company employees. The subject of the paper is the “mitigation” of the placebo response in placebo-controlled randomized clinical trials.
As Dr. Healy points out in Shipwreck, the placebo-controlled randomized clinical trial once was held up as a shield to protect us from useless or dangerous medicines but in fact has turned into a vehicle to deliver adverse consequences with impunity. There is no requirement that a given drug be shown to be better than existing remedies already on the market. The drugmakers have been tasked with manufacturing and controlling the evidence purporting to show their wares are safe and effective and they control every aspect of the trials, including but not limited to the choice of patients, dosing regimen, duration, outcome, and types of statistical tests used. The rest of us (including your doctor) cannot see the patient-level data, and instead have a rely on a carefully selected and manipulated subset. The pharmaceutical companies can and do switch outcomes and hide data on adverse events up to and including death, and when all else fails, entire studies can be deep-sixed.
But apparently, all this isn’t enough for them. Dr. Cohen and his co-authors eloquently describe the plight of the drugmakers:
Placebo response rates have steadily increased over the past several decades which has been described as ‘a major obstacle in CNS drug development’ and ‘a growing crisis.’
Luckily, Drs. Cohen et al. are coming to the rescue by means of by means of something called the Placebo-Control Reminder Script, or PCRS, “a brief interactive procedure that educates patients about factors known to cause placebo response,” and which was the subject of a randomized controlled trial.
Al the patients enrolled in the two-week trial had been diagnosed with either Major Depressive Disorder, Schizophrenia, or Schizoaffective Disorder. All were told they would have a fifty percent chance of receiving either the active drug or placebo. In fact, all the patients were given a placebo pill.
In plain English, the investigators blatantly lied to the patients who had put their trust in them, but never mind that for now. The randomization consisted of assigning half of the patients to the PCRS, while the remaining half were not. The PCRS, which was read to study subjects by raters at the beginning of Week One and again at the beginning of Week Two, contained information about the nature of placebo-controlled trials, including these words:
Remember that the placebo is inactive and should do nothing to help your symptoms and should have no side effects. [Emphasis added.]
All subjects were assessed by means of the Beck Depression Inventory-II, a diagnostic instrument which, in the words of the authors, “has the combinatory characteristics of stability and malleability needed to evaluate the placebo response” – which may sound an awful lot like saying “We can make this mean whatever we want it to mean.” But again, never mind that for now.
The patients in the PCRS group exhibited a significantly smaller decrease in depression than the Non-Intervention Group – although the graph in Figure 2 shows the disparity falls short of the minimum clinically important difference needed to be considered meaningful by a patient. But once again, never mind that for now.
Why is this important? Drs. Cohen et al. explain:
A variety of complicated and expensive clinical trial design features, such as double-blind placebo-run-in periods and remote raters, have thus far failed to show clear and consistent benefits for mitigating the placebo response. [Emphasis added.]
Notably, although the [PCRS] script is directed at the participant, it concurrently serves as a reminder to rates to manage their own behaviors and expectations – factors which have also been shown to enhance the placebo response. [Emphasis added.]
By enhancing signal detection, such procedures may help drug developers progress compounds to faster approval and reach patients who are suffering sooner.
In other words, both rates and subjects alike were repeatedly admonished to keep in mind that the presence side effects was an indication that they were getting the active drug, and their absence a sign they were getting placebo. This is a blatant violation of the very essence of a placebo-controlled trial, which is that neither patients nor clinicians should have any idea whether a given patient is getting the active drug or not.
Even more disturbing. Dr. Cohen and his co-authors give no indication – none – that they understand the placebo effect may be a useful signal that the drugs don’t work. Rather, they make it clear they see it as just one more hurdle that needs to be surmounted on the way to getting the drugs to market, never explaining why we need to “reaching patients who are suffering sooner” with useless or dangerous “medicines.” With that, the drugmakers have dropped all pretense of being on our side.
Buy “Shipwreck of the Singular” on Amazon
Shipwreck of the Singular
by David Healy
Shipwreck of the Singular: