This Article  • Full text  • PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Citing articles on ISI (3)  • Contact me when this article is cited  Related Content  • Related articles  • Similar articles in JAMA  Topic Collections  • Adverse Effects  • Alert me on articles by topic

	Since this article does not have an abstract, we have provided the first 150 words of the full text and any section headings.

In Reply: I strongly agree with Drs Hauben and van Puijenbroek that true signals should emerge from clinical judgment and that statistical algorithms, such as PRRs,^1-5 should be used as supplements to clinical and epidemiological judgment, not replacements. I also agree that the value of statistical algorithms even in that role remains unproven. Unfortunately, however, statistical algorithms are too often used alone, in publications and in the courtroom, as if they represent analyses useful for hypothesis testing, which is inappropriate.

I also agree that sometimes signals are acted upon without waiting for epidemiological confirmation, despite the desirability of waiting. However, there is always a risk of acting prematurely and incorrectly. When the signal is made stronger by combining it with clinical judgment, and if the action is relatively minor (such as product label changes), an early decision is easier to make.

Regarding the utility of publishing case reports, that is . . . [Full Text of this Article]

Brian L. Strom, MD, MPH
bstrom@cceb.med.upenn.edu
Center for Clinical Epidemiology and Biostatistics
University of Pennsylvania School of Medicine
Philadelphia

RELATED ARTICLES

Evaluation of Suspected Adverse Drug Reactions
Manfred Hauben and Eugène P. van Puijenbroek
JAMA. 2005;293(11):1324.
EXTRACT | FULL TEXT  

Potential for Conflict of Interest in the Evaluation of Suspected Adverse Drug Reactions: A Counterpoint
Brian L. Strom
JAMA. 2004;292(21):2643-2646.
EXTRACT | FULL TEXT