2339. Clostridioides difficile: Impact of Active Screening of Asymptomatic Carriers and Testing Stewardship

BACKGROUND: We recently implemented a hospital-wide C. difficile testing algorithm and screening/isolation of C. difficile asymptomatic carriers primarily in heme-onc units. We aim to evaluate the impact of these interventions on the epidemiology of C. difficile + tests. METHODS: This retrospective cohort was performed in a 600-bed hospital in Milwaukee, WI, from January 1, 2016 to March 31, 2019. All clinical C. difficile tests included nucleic acid amplification (NAAT; Xpert C. difficile, Cepheid). On February 2017, all NAAT+ tests had toxin (tox) checked (Quick check complete, Alere). Testing algorithm (Figure 1) started mid 2016 until now. Screening phases included: Phase 1 (September 2016–May 2017): C. difficile screening cultures shared with units but not placed in electronic medical records (EMR). Patients + placed on enteric precautions (gown, gloves, hand hygiene). Phase 2 (May 2017–January 2018): C. difficile screening (NAAT) performed on admission and weekly thereafter, results placed in EMR, NAAT+ patients placed on enteric precautions. Phase 3 (January 2018–present): C. difficile screening (NAAT) on admission, results placed in EMR, NAAT+ patients placed on enteric precautions. Federal reporting changed to only reporting NAAT+tox+. Tests (NAAT+, NAAT+tox+, and NAAT+tox-) were analyzed using Poisson regression offsetting for log of patient-days using SAS, v9.4. RESULTS: Hospital-wide C. difficile tests decreased from 21 to 10.9 tests per 1,000 patient-days (P < 0.0001; Figure 2). This effect was seen in heme-onc units (41 to 15.7; P < 0.0001; Figure 3) and in all other units (18.9 to 9.9; P < 0.0001). All NAAT+ results decreased from 2.99 to 1.94 per 1,000 patient-days hospital wide (P < 0.0001) but remained unchanged in heme-onc units (4.6 to 3.7, P > 0.05). NAAT+tox+ results remained unchanged hospital wide and in heme-onc units (0.8 to 0.7 and 1.1 to 1.2, respectively; both P > 0.05); however, the frequency of NAAT+tox− tests decreased hospital wide (1.8 to 1.3; P = 0.0003) and in heme-onc units (3.8 to 2.4; P = 0.05). CONCLUSION: A C. difficile testing algorithm was successful decreasing the number of C. difficile tests performed and had a hospital-wide reduction of NAAT+tox− tests. The rate of NAAT+tox+ cases in heme-onc units and hospital wide remained unchanged despite active screening and isolation in selected units. [Image: see text] [Image: see text] [Image: see text] DISCLOSURES: All authors: No reported disclosures.