THE double-blind technique is only one way of dealing with bias in drug studies. Three other approaches that can validly be used are herein described and evaluated.
"Bias" as an experimental design concept may be defined simply as any cause, factor, variable, or suspected variable which could conceivably make the conclusions from an experiment invalid or render them ambiguous. More specifically, if a factor can affect one group or one condition in an unknown way it can be considered a source of bias.
From this point of view, biases may enter into an experiment in many different ways. For example, if two groups of depressed patients are being compared in terms of response to two drugs, and one is a group of clinic patients and the other is a group of private patients, then any differences in reaction could be related to several factors. These include such variables as patients'
Plutchik R, Platman SR, Fieve RR. Three Alternatives to the Double-Bind. Arch Gen Psychiatry. 1969;20(4):428–432. doi:10.1001/archpsyc.1969.01740160044007
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: