Frederick M. Hecht, MD
To the Editors: For decision analysis to be clinically relevant, it must rest on accurate information and model realistic choices. I believe that the decision analysis used by Johanson and Sonnenberg (1) in their recent article on the evaluation of diarrhea in the acquired immunodeficiency syndrome (AIDS) is based on problematic assumptions. The results are likely to be misleading and potentially dangerous.
The investigators make a critical assumption that diarrhea will resolve in 67% of patients through symptomatic treatment with an antimotility agent (diphenoxylate hydrochloride), regardless of cause and with no recurrences. As a consequence of this assumption, in the
Hecht FM. Diarrhea and AIDS. Ann Intern Med. 1990;113:804–805. doi: 10.7326/0003-4819-113-10-804
Download citation file:
Published: Ann Intern Med. 1990;113(10):804-805.
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use