This column, like the previous 2 columns in this series, shows how placebo is a control for the natural course of an illness. In the study presented in this column, the use of placebo enabled the investigators to detect that active treatment was deleterious to the patient. In the second column in the series, the use of placebo showed that adding a usually effective antibiotic for an infection did not alter the outcome because of the nature of the infection and the value of treatment as usual (TAU). The first column in the series described TAU in a double-blind randomized controlled trial (RCT). Parenthetically, that column also discussed how different TAU is in such studies versus TAU in clinical practice. Those differences likely contribute to why TAU in such trials can be quite effective and why they are frequently negative. That first paper also discussed how the Heisenberg Uncertainty Principle from physics can be applied to psychiatric clinical trials, further emphasizing the difference in care received in an RTC versus clinical practice. Thus, investigators need to take the Heisenberg Principle into account when designing their studies and readers need to consider it when assessing the results. In the studies described in the second and third columns in the series, the focus was on objective and clinically meaningful endpoints (ie, resolution of a localized infection in the second column and death in this third column) in contrast to the often subjective endpoints in many psychiatric clinical trials (eg, improvement on a rating scale of symptoms that are reported by a participant/patient). The field of psychiatry needs to aim to develop such objective and clinically meaningful endpoints for its clinical trials. Finally, all 3 columns in this series emphasize that placebo does not mean nothing but rather involves all the treatment that a patient receives during the trial, except for the investigational treatment being tested, as well as the natural history of the illness, which includes the natural fluctuation in symptoms.
Sheldon H. Preskorn (Thu,) studied this question.