1034. Factors Associated with De-escalation in Patients with Healthcare-Associated Pneumonia
Session: Poster Abstract Session: Antibiotic Stewardship: General Acute Care Implementation and Outcomes
Friday, October 28, 2016
Room: Poster Hall
Background: Antimicrobial resistance is a significant healthcare quality and safety issue. Physicians routinely prescribe broad-spectrum therapy for nosocomial infection to avoid inappropriate treatment associated with greater mortality; competing with this goal is the need to prevent resistance. Antibiotic de-escalation is a strategy used to optimize antimicrobial use. Strategies to identify patients suitable for de-escalation would have major benefits. The clinical impact of such a strategy has not been widely studied. The aim of this study was to identify factors associated with de-escalation in patients with healthcare-associated pneumonia (HCAP) in the non-intensive care unit (non-ICU) setting.

Methods: We performed a retrospective electronic health record review of all adult patients admitted to the Queen’s Medical Center with a diagnosis of HCAP from January 1, 2013 to December 31, 2014. Patient demographics, comorbidities, microbiologic data, infectious diseases specialty consultation, assignment to resident teams, and data elements for pneumonia severity index (PSI) and CURB-65 scores were collected. Patients were followed for de-escalation for a maximum of 5 days. A logistic regression model was used for analysis.

Results: A chart review revealed 217 cases admitted to the non-ICU setting with pneumonia and subsequent positive blood cultures and/or sputum cultures. Seventy-four cases met predetermined criteria for HCAP, with 58 (78.4%) cases having been de-escalated. In univariate analysis, factors associated with the implementation of de-escalation included antibiotic use within 30 days of admission (OR, 3.11, CI 1.09-9.84; p<0.05), malignancy (OR, 5.29, CI 1.34-35.5; p<0.05), assignment to resident teams (OR, 6.77, CI 1.73-45.3; p<0.05), and PSI score (OR, 0.987, CI 0.973-1.00; p<0.05). De-escalation was associated with low in-hospital mortality (p<0.05) and high number of hospital free-days (p<0.05). Even in higher severity defined as PSI risk class V (OR, 0.148, CI 0.0187-0.800; p<0.05), de-escalation was associated with decreased mortality.

Conclusion: De-escalation occurred in patients who were less ill. However, de-escalation may benefit mortality regardless of severity.

Lorrance Majewski, DO, Nobuhiro Ariyoshi, MD, Zao Zhang, MD, Nattawat Klomjit, MD and Heath Chung, MD, Internal Medicine, John A. Burns School of Medicine, University of Hawaii at Manoa, Honolulu, HI


L. Majewski, None

N. Ariyoshi, None

Z. Zhang, None

N. Klomjit, None

H. Chung, None

Findings in the abstracts are embargoed until 12:01 a.m. CDT, Wednesday Oct. 26th with the exception of research findings presented at the IDWeek press conferences.