[Skip to Content]
[Skip to Content Landing]
Citations 0
December 17, 2003

Premature Application of Trial Results to Clinical Practice—Reply

Author Affiliations

Letters Section Editor: Stephen J. Lurie, MD, PhD, Senior Editor.

JAMA. 2003;290(23):3072. doi:10.1001/jama.290.23.3072-a

To the Editor: Dr Clark and colleagues1 found an increased use of an experimental, unproved therapy (apheresis) by community physicians during the time that randomized controlled trials of that therapy were being performed. The authors called this behavior "jumping the gun."

The authors' examples, however, suggest that physicians may have anticipated the trials' results prior to publication. For example, Clark et al found that use of apheresis for thrombotic thrombocytopenic purpura increased gradually during the relevant trial.2 That increase continued after the trial found a positive effect. In contrast, during a trial of apheresis for multiple sclerosis3 (which found the treatment to be ineffective after a slight trend favoring apheresis during the first 2 years), the use of apheresis among nontrial patients increased during the first 2 years of the trial but decreased in the last year of the trial, before its lack of effectiveness was demonstrated. This decrease contrasts with the increased use of apheresis among the multiple sclerosis trial participants up to the end of the trial.