[Skip to Content]
[Skip to Content Landing]
Article
September 9, 1992

Rapid Classification of Positive Blood Cultures-Reply

Author Affiliations

Brigham and Women's Hospital Boston, Mass

JAMA. 1992;268(10):1267. doi:10.1001/jama.1992.03490100060021
Abstract

In Reply.  —Dr Johnson asserts that the findings of this study are simple and intuitive and should be common knowledge. Unfortunately, much of what is "common knowledge" in medicine has been found incorrect when subjected to scrutiny. Furthermore, we have previously demonstrated that physicians expend substantial resources evaluating and treating patients with contaminant cultures, suggesting that the distinction between true-positive and false-positive blood cultures is not always obvious. Others have found that physicians routinely overestimate the probability of true bacteremia.1 Our hope is that the information we have presented will better allow physicians to focus on the important pieces of data and calibrate their estimates regarding the likelihood that a positive blood culture represents a true positive.Dr Chapdelaine notes that "Organism category 2" was included in the algorithm despite borderline statistical significance of this variable. As is apparent from Table 6, the organism categories were of major importance,

×