EARLY PREDICTION OF BACTEREMIA - TESTING IN A UNIVERSITY AND A COMMUNITY-HOSPITAL

BACKGROUND: Two rules (model 1 and model 2) were previously derived an d prospectively validated at the same institution to predict the likel ihood of bacteremia. The objective of the present study was to test an d compare the performance of the rules in patients admitted to two sit es of inpatient care: a university hospital and a community hospital. METHODS: Clinical and laboratory data (including the variables contain ed in the two models) were collected within 24 hours in all patients a dmitted to the Department of Medicine of the Beilinson Medical Center, a university hospital in central Israel, and Emek Hospital, a communi ty hospital in northern Israel, because of an acute infectious disease . The scores of the models were compared with the results of blood cul tures. RESULTS: The percentage of bacteremia was 15% in the university and 18.5% in the community hospital. The area under the receiver-oper ating characteristic curve was 0.56 +/- 0.04 SE for model 1, and 0.67 +/- 0.04 SE for model 2 in the university hospital; and 0.59 +/- 0.05 SE versus 0.63 +/- 0.04 SE, respectively, in the community hospital. A t the best calibration, model 1 defined low-risk groups of 205 patient s in the university hospital, and 66 patients in the community hospita l, with prevalences of bacteremia of 13% and 15%. The high-risk groups defined by model 1 had prevalences of 30% and 32%. Model 2 defined lo w-risk groups with prevalences of bacteremia of 7% (8 of 114) and 8% ( 6 of 76); and high-risk groups with percentages of 29% and 38%. CONCLU SIONS: The overall accuracy of the two models deteriorated significant ly. Both models defined groups at high risk of bacteremia, but the per centages of bacteremia and of death in the low-risk groups do not enco urage withholding blood cultures in these patients. The failure of the two models points toward the need for external validation, and for mo nitoring performance of prediction models over time.