46. Templated Microbiology Comments with Candiduria to Enhance Antimicrobial Stewardship

Abstract Background Behavioral interventions have been shown to improve antimicrobial selection. Such practices are low cost and effective means of stewardship promotion. One area of overtreatment that contributes to unnecessary antifungal use is in hospitalized patients with candiduria. We implemented a templated microbiology comment to guide prescribing of antifungals for hospitalized patients with candiduria. Methods This was a quasi-experimental, multi-center, single health system study. When Candida is isolated, the following comment appears in the microbiology result section along with the urine culture result: “In the absence of symptoms, Candida is generally considered normal flora. No therapy indicated unless high risk (pregnant, neonate or neutropenic) or undergoing urologic procedure. If Foley catheter present, remove or replace when able.” We compared a pre-implementation cohort (June 2018-Janurary 2019) to a post-implementation cohort (June 2019-Janurary 2020). Patients were included in the study if they were inpatients, 18 years and older, with candiduria. The primary outcome was the rate of antifungal administration within 72 hours after culture results became available. Secondary outcomes include duration of therapy and rate of antifungal given within 73-240 hours after culture result. Results The study included a total of 297 patients between the two groups (156 pre-implementation, 141 post-implementation). The primary outcome was found to be significantly lower in the post-implementation group (48.1% vs 34.0%, p=0.014). A multivariate adjustment for baseline characteristics that were significantly different between groups revealed that post-implementation group maintained its effect (OR 0.49 (0.29, 0.82), p=0.0067). For secondary outcomes, no difference was found in patients requiring antifungal administration within 73-240 hours after microbiology results were available (1.3% vs 3.5%, p=0.199). There was no difference in mean antifungal duration (4 vs 3 days, p=0.449). Conclusion Adding a templated comment to urine cultures was associated with a significant reduction in the number of antifungals prescribed in patients with candiduria. This strategy is an effective low-cost, passive education technique to improve antimicrobial stewardship. Disclosures All Authors: No reported disclosures

Background. According to the WHO, carbapenems and fluoroquinolones (FQ) should be key targets for stewardship programs.
Methods. A multifaceted antimicrobial stewardship program (ASP) was implemented in July 2018 at a 160-bed tertiary care center serving the tristate area of Iowa, South Dakota and Nebraska. Carbapenem and FQ use during pre-ASP intervention period (P1: 12/01/2016-6/30/2018) was compared with ASP-intervention period (P2: 07/01/2018-1/31/2020). ASP interventions included: stewardship educational pearls in monthly physician newsletters; educational posters in provider areas; suppression of carbapenem results on microbiology susceptibility reports; provider counseling for appropriate ordering; creating carbapenem alternative alert in order-entry software; removing FQ and carbapenems from order-sets where appropriate; default antibiotic stop dates changed to 7 days in EMR (Epic); adverse effects warning fired as an alert when ordering FQ. Pharmacist interventions: procalcitonin protocol allowing pharmacists to reorder follow-up procalcitonin and make recommendations to discontinue therapy where appropriate.
Results. FQ use declined significantly from a mean of 133 days of therapy (DOT) per 1000 days to 46 DOT per 1000 patient days during P2 (p< 0.0001). Carbapenem use declined significantly from a mean of 65 DOT per 1000 patient days during P1 to 9 DOT per 1000 patient days in P2 (p< 0.001). All hospital units showed a significant decrease in use, with intensive care units (ICUs) noting 56% reduction (p< 0.00001) during P2 compared to P1. During P2, 55% of orders for carbapenems and FQ during P2 were found to be appropriate compared to 39% in P1 (p< 0.0001). Sensitivity profile for Pseudomonas aeruginosa improved from 86% carbapenem sensitivity during P1 to 89% in P2 and no Carbapenem-Resistant Enterobacteriaceae isolates were identified during the study period; FQ sensitivity remained stable at 81%. Cost savings of $757 per 1000 patient days were recognized in P2 as a result of reduced use.
Conclusion. With ASP and pharmacist interventions, a significant decline in total utilization of carbapenem and FQ, considerable cost savings and an increase in proportion of appropriate use were observed.
Disclosures. Background. Behavioral interventions have been shown to improve antimicrobial selection. Such practices are low cost and effective means of stewardship promotion. One area of overtreatment that contributes to unnecessary antifungal use is in hospitalized patients with candiduria. We implemented a templated microbiology comment to guide prescribing of antifungals for hospitalized patients with candiduria.
Methods. This was a quasi-experimental, multi-center, single health system study. When Candida is isolated, the following comment appears in the microbiology result section along with the urine culture result: "In the absence of symptoms, Candida is generally considered normal flora. No therapy indicated unless high risk (pregnant, neonate or neutropenic) or undergoing urologic procedure. If Foley catheter present, remove or replace when able. " We compared a pre-implementation cohort (June 2018-Janurary 2019) to a post-implementation cohort (June 2019-Janurary 2020). Patients were included in the study if they were inpatients, 18 years and older, with candiduria. The primary outcome was the rate of antifungal administration within 72 hours after culture results became available. Secondary outcomes include duration of therapy and rate of antifungal given within 73-240 hours after culture result.
Results. The study included a total of 297 patients between the two groups (156 pre-implementation, 141 post-implementation). The primary outcome was found to be significantly lower in the post-implementation group (48.1% vs 34.0%, p=0.014). A multivariate adjustment for baseline characteristics that were significantly different between groups revealed that post-implementation group maintained its effect (OR 0.49 (0.29, 0.82), p=0.0067). For secondary outcomes, no difference was found in patients requiring antifungal administration within 73-240 hours after microbiology results were available (1.3% vs 3.5%, p=0.199). There was no difference in mean antifungal duration (4 vs 3 days, p=0.449).

Conclusion.
Adding a templated comment to urine cultures was associated with a significant reduction in the number of antifungals prescribed in patients with candiduria. This strategy is an effective low-cost, passive education technique to improve antimicrobial stewardship.
Disclosures. All Authors: No reported disclosures