![[ Visit Client Website ]](images/banner.jpg)
BACKGROUND: Tennessee acute care hospitals have been required to report Clostridium difficile infection (CDI) laboratory identified (LabID) events via the National Healthcare Safety Network (NHSN) since January 2012. In order to address CDI within Tennessee facilities across the healthcare spectrum, the Tennessee Department of Health (TDH) sought to target facilities within the same referral network with the intent of minimizing CDI transfer rates while also decreasing individual targeted facility rates. We outline the methodology utilized in prioritizing appropriate facilities for recruitment to ensure effective targeting of facilities to maximize impact.
METHODS: In order to create a community-onset (CO) CDI standardized infection ratio (SIR), community-onset rates were modeled using negative binomial regression. Risk modeling allowed for an estimate of the expected number of CO-CDI per facility. Hospital-onset (HO) standardized infection ratios were obtained from NHSN. A cumulative attributable difference (CAD) was calculated for each facility’s HO and CO SIR by subtracting the expected number from the observed number. Six rankings were developed by combining facility specific CDI counts, CADs, or rates from first quarter 2014.
RESULTS: The top 3 methodologies considered were: 1) Rank by HO-CAD, then CO-CAD, 2) a weighted rank of HO-CAD and CO-CAD relative to each corresponding absolute CDI count, 3) rank by CO-CAD, then HO-SIR. All rankings revealed commonalities; the same 8 hospitals were included in every ranking’s top 12 facilities. Out of the 3 rankings, TDH chose the combination of ranking facilities by their CO-CAD and HO-SIR. Facilities with the top 30 highest CO-CADs were ranked a second time by their hospital onset SIR. We plan to recruit the top 15 facilities for the collaborative.
CONCLUSIONS: Initial methodology based selection solely on high CO-CDI prevalence rates. However, CO-CDI rates do not account for important risk-adjustment variables, like test type or other meaningful facility differences. The Centers for Disease Control and Prevention (CDC) and TDH have developed metrics to address these important issues, which TDH applied to improve facility targeting. Using the CAD provides each facility’s excess infections, therefore offering an indication of the hospital’s infection burden. We hope that our chosen methodology will allow TDH to use risk-adjusted metrics for reducing CO-CDI and strengthen our targeted recruitment for the CDI prevention collaborative.