216. Association Between Days to Initiate Appropriate Therapy and Hospital Length of Stay Among Adult Hospitalized Patients With Gram-negative Bloodstream Infections (GN-BSI)

BACKGROUND: The deleterious outcomes associated with delay receipt of appropriate therapy are well documented. However, scant data exists on the consequences of each day delay of appropriate therapy and subsequent outcomes among adult hospitalized patients with GN-BSIs. METHODS: Study design: a retrospective cohort analysis. Study population: consecutive adult, hospitalized patients with a GN-BSI (11 most prevalent pathogens) in 1 of 181 institutions contributing microbiology data to the Premier Healthcare Database (October 2010–Sep 2015). Exclusion criteria: age < 18 years; diagnosis of pregnancy or cystic fibrosis, died or discharged within 2 days of index GN-BSI culture, lack of sufficient antibiotic susceptibility or treatment data to determine appropriateness. Day of initiating appropriate therapy was defined as the first day when the patient received an antibiotic with in vitro activity against the GN-BSI post index culture. Results were summarized by Kaplan–Meier estimates, and Cox Proportional-Hazards (CPH) analyses modeling discharge to home were conducted. Time to initiate appropriate therapy (0, 1–2 days, 3–4 days, ≥5 days) was included in the CPH model as an ordinal variable. RESULTS: A total of 40,549 patients met selection criteria. Mean (SD) age was 67.5 (16.1) years and 54% were female. E. coli and K. pneumoniae were the most common GN-BSI (58.0% and 18.3%, respectively). Approximately 30% of patients were in the ICU at index GN-BSI and in-hospital mortality was 6.8%. The mean (SD) time to receive appropriate therapy post index GN-BSI culture was 0.6 (2.7) days, and 69.7%, 22.5%, 5.7% and 2.1% received appropriate therapy in 0, 1–2, 3–4, and ≥5 days of index GN-BSI, respectively. The mean/median LOS post index GN-BSI by 0, 1–2, 3–4, and ≥5 days delays in appropriate treatment were 8.3/6, 9.8/7, 11.5/8, and 19.2/11 days respectively. Kaplan–Meier plots are shown in Figure 1. In the CPH model, each interval delay in appropriate therapy was associated with a 21% decrease in the likelihood of being discharged home for patients with GB-BSIs. CONCLUSION: Hospital length of stay was found to increase when appropriate therapy was delayed. These findings highlight the critical need for early appropriate therapy among patients with GN-BSIs. [Image: see text] DISCLOSURES: All authors: No reported disclosures.