Background: Approximately 90% of all blood cultures grow no organisms (ie, are true negatives), and 5% are thought to represent contaminants (ie, are false positives). The cost effectiveness of blood cultures could therefore be improved by developing rules that safely decreased the number of cultures drawn from patients with a low likelihood of having bacteremia and/or by improving the process of obtaining cultures, thereby decreasing the number of contaminants. We analyzed the potential effects of these two approaches.
Methods: We annualized the hospital costs and lengths of stay for patients with true-negative and false-positive blood cultures from a retrospective analysis of 939 sets of cultures drawn in January 2002.
Results: Of the 939 blood culture sets, 816 (87%) were true negatives and generated annualized costs of approximately 750,000 dollars. Although only 56 (6%) of the blood culture sets were false positives, they resulted in annualized costs of 1.4-1.8 million dollars and added an estimated 1450-2200 extra hospital days/year.
Conclusions: Despite there being nearly 15 times as many true-negative blood cultures as false positive ones, far greater improvements in resource utilization would result from reducing the number of contaminated blood cultures than by reducing the number of true negatives. The potential savings from this approach are of sufficient magnitude to justify investing considerable resources to attaining this goal.
(c) 2006 Society of Hospital Medicine.