970. City-wide Collaboration to Reduce C. difficile Infections
Session: Poster Abstract Session: Clostridium difficile Infections: Treatment and Prevention
Friday, October 9, 2015
Room: Poster Hall
Posters
  • Poster 970.pdf (359.1 kB)
  • Background: The rate of hospital-onset (HO) Clostridium difficile infections (CDI) in Monroe County, NY in 2011 was 10.8 per 10,000 patient days; 21% higher than the state average. We formed a multidisciplinary city-wide collaborative to reduce CDI.

    Methods: In 2011, four acute care hospitals partnered to reduce HO and community-onset healthcare facility-associated (CO-HCFA) CDI rates by 30% over 3 years in 2 phases:

     Phase I (2011) - Environmental Cleaning and Infection Prevention:

    • Bleach use, ultraviolet light disinfection, observation of patient room cleaning, Adenosine triphosphate (ATP) bioluminescence testing.
    • Hand hygiene, decreased time to isolation, longer isolation length, dedicated equipment, and shared equipment disinfection.

    Phase II (2014) – Antimicrobial Stewardship:

    • Analysis of comparative antibiotic use and antibiotic indications to provide a benchmark and guide interventions.
    • Reduction of quinolone use due to the prevalence of the NAP1/027 C. difficile strain by targeting the 2 most common indications for quinolones: community-acquired pneumonia (CAP) and urinary tract infection.
    • Revision of CAP treatment guidelines to ceftriaxone and doxycycline, the latter due to its potential protective effect against CDI.
    • Varied implementation methods: post-prescription review, order set modification, and moxifloxacin restriction.

    CDI rates were tracked using the National Healthcare Safety Network. Generalized estimating equations using a negative binomial regression model were used to determine the effect of interventions.

    Results: Prior to adjusting for the longitudinal nature of the data, the HO CDI rate decreased from 10.8 to 7.9 cases per 10,000 patient days (Rate Ratio [RR] 0.73 (95% Confidence Interval [CI] .65, .83)) between 2011 and 2014. CO-HCFA CDI rates decreased from 2.9 to 2.1 (RR 0.70 (95% CI .55, .89)). Multivariate analysis showed that our multimodal approach significantly decreased the HO CDI rate (p<0.0001), with time since implementation of interventions contributing significantly to the downward trend.

    Conclusion: Reducing healthcare-associated CDI rates requires a multimodal approach with an intervention bundle based on national best practices and local data trends. Sharing of data and knowledge across hospitals is key to achieving community-wide collaboration and success.

    Christina Felsen, MPH1, Gail Quinlan, RN, MSN1, Nayef El-Daher, MD, PhD2, Donna Farnsworth, R.N.2, Paul Graman, MD, FIDSA, FSHEA3, Linda Greene, RN, MPS, CIC4, Maryrose Laguio, MD5, Mark Shelly, MD, FSHEA6, Ann Marie Pettis, RN, BSN, CIC7, Wan Tang, PhD8, Xin Tu, PhD9, Susan Messing, MS, MA3, Elizabeth Dodds Ashley, PharmD, MHS3 and Ghinwa Dumyati, MD, FSHEA3, (1)Center for Community Health, University of Rochester Medical Center, Rochester, NY, (2)Rochester Regional Health System, Rochester, NY, (3)University of Rochester Medical Center, Rochester, NY, (4)Infection Prevention, Highland Hospital, Rochester, NY, (5)Infectious Disease, Rochester General Hospital, Rochester, NY, (6)Infectious Disease, University of Rochester Medical Center, Rocheseter, NY, (7)Infection Prevention, UNIVERSITY OF ROCHESTER MEDICAL CENTER, Rochester, NY, (8)Biostatistics and Computational Biology, University of Rochester, Rochester, NY, (9)Department of Biostatistics and Computational Biology, University of Rochester, Rochester, NY

    Disclosures:

    C. Felsen, None

    G. Quinlan, None

    N. El-Daher, None

    D. Farnsworth, None

    P. Graman, None

    L. Greene, None

    M. Laguio, None

    M. Shelly, None

    A. M. Pettis, None

    W. Tang, None

    X. Tu, None

    S. Messing, None

    E. Dodds Ashley, Up to Date: Collaborator , Licensing agreement or royalty

    G. Dumyati, None

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.