Program Schedule

Diagnostic Errors that Lead to Inappropriate Antimicrobial Use

Session: Oral Abstract Session: New Approaches to Antibiotic Stewardship
Friday, October 10, 2014: 9:00 AM
Room: The Pennsylvania Convention Center: 107-AB
Background:   We have reported that diagnostic errors factor into 89% of inappropriate inpatient antimicrobial (AM) courses.  Here we describe in detail the diagnostic errors that led to inappropriate courses.

Methods:   We studied AM courses given to 500 randomly selected VA hospital inpatients between 2007 and 2008.  We recorded information about the syndrome or disease that led to AM use (index condition), the initial provider diagnosis of the index condition (provider diagnosis), and AM drug(s), dose, route, and duration.  Each case summary was assigned randomly to 2 of 4 blinded reviewers. Based on data available to providers in real time, reviewers classified the provider diagnosis as correct, uncertain, not correct, or a sign or symptom of a potential infection rather than a syndrome or disease (sign or symptom) and the index AM course as appropriate or not, considering drug selection, route, dose, and duration.  

Results:  Overall, provider diagnoses were correct in 292/500 (58%) of cases, but accuracy varied by diagnosis. For the 6 most common diagnostic groups (376 [75%] of all cases), accuracy ranged from 90% for bacteremia or sepsis to 27% for cystitis, pyelonephritis, or urosepsis (P < .001).  Provider AM courses were considered appropriate in 191/500 (38%) of cases overall, but appropriateness varied in relation to diagnostic accuracy.  When provider diagnoses were correct, 181/292 (62%) of courses were appropriate compared with only 10/208 (5%) when diagnoses were incorrect, uncertain, or signs or symptoms (P < .001).  Considering the 111 cases in which provider diagnoses were correct and AM courses were inappropriate, reasons included wrong AM drug(s) (81, 73%), wrong duration (43, 39%), wrong dose (5, 5%), another type of drug error (1, 1%), and AM therapy not indicated (11, 4%).  In contrast, of the 198 cases in which provider diagnoses were incorrect or uncertain or were signs or symptoms and AM courses were inappropriate, AM therapy was not indicated in 171 (86%) cases. Other reasons included wrong AM drugs (31, 16%), incorrect duration (13, 7%), wrong dose (2, 1%), and another type of drug error (1, 0.5%).

Conclusion:   Efforts to improve AM courses should include strategies to help providers make accurate diagnoses of syndromes that might be due to infection.

Gregory A. Filice, MD1,2, Dimitri M. Drekonja, MD, MS3, Joseph R. Thurn, MD, MPH4,5, Galen M. Hamann, BSN6, Bobbie T. Masoud, PharmD6 and James R. Johnson, MD7, (1)Division of Infectious Disease, University of Minnesota, Minneapolis, MN, (2)Section of Infectious Disease, Minneapolis Veterans Affairs Healthcare System (MVAHCS), Minneapolis, MN, (3)Infectious Diseases, Minneapolis Veterans Affairs Health Care System, Minneapolis, MN, (4)Minneapolis VAMC, Minneapolis, MN, (5)University of Minnesota Medical School, Minneapolis, MN, (6)Minneapolis Veterans Affairs Medical Center, Minneapolis, MN, (7)University of Minnesota, Minneapolis, MN


G. A. Filice, None

D. M. Drekonja, None

J. R. Thurn, None

G. M. Hamann, None

B. T. Masoud, None

J. R. Johnson, None

Findings in the abstracts are embargoed until 12:01 a.m. EDT, Oct. 8th with the exception of research findings presented at the IDWeek press conferences.

Sponsoring Societies:

© 2014, All Rights Reserved.

Follow IDWeek