Doctor Medication Guesses: an Indicator Of Clinical Improvement in Double-Blind Studies

A Contribution to the Methodology of Psychiatric Drug Evaluation.

¹ University of Pennsylvania.

In our controlled drug studies carried out with neurotic patients, medication guesses were unrelated to reported side reactions, but highly significantly influenced by clinical improvement. A physician appears to guess "drug" when his patient improves, and to guess "placebo" when his patient does not improve, irrespective of whether his patient is on drug or placebo. The more effective a drug is clinically, the more accurate the physician is in his medication guesses. In other words, the physician guesses more drug patients to be on drug than on placebo. The fact that a physician is most certain when he guesses drug correctly, while he is least certain when he guesses drug in correctly, seems to indicate that there may well exist minor cues of which we are not aware, that allow the physician to be more certain in his correct drug guesses. One obvious cue may well be the larger degree of improvement seen in these drug patients as compared to improved placebo patients and unimproved drug patients.

It is suggested that the physician's medication guesses may represent an important indicator of drug action and may differentiate between clinically active drugs and placebo as efficiently as the best rating scales available. If drugs do not produce major side effects, improvement must be the main cue used by the physician in his medication guess. Medication guesses carried out in drug evaluations performed in neurotic populations may well represent one of the most sensitive measures used to differentiate clinically active drugs from placebo.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?