Effect of Testing Methods on Incidence of Clostridioides difficle Infection Rates in Veterans’ Affairs Medical Centers

Abstract

Background: Healthcare-associated Clostridioides difficile infection (CDI) rates have been decreasing in Department of Veterans’ Affairs (VA) acute-care medical facilities since the CDI Prevention Initiative began in 2012. Assessment of rates, however, is complicated by changing surveillance definitions and diagnostics. Over the past 2 years, the VA has adopted the less stringent surveillance definitions of the NHSN for hospital-onset healthcare-facility–associated (HO-HCFA) CDI (onset on or after day 4 of hospitalization) than was originally used (>48 hours after admission). New diagnostic testing methods have been developed, but variabilities in the reliability of testing methods for the diagnosis of CDI in patients have been detected, thus not yielding a gold standard test. As a result, some facilities use nucleic acid amplification testing (NAAT) for CDI diagnosis and reporting whereas others use a 2-step process of NAAT followed by a toxin enzyme immunoassay (EIA) test, with the latter determining positivity for reporting (as allowed by the NHSN). We reviewed CDI rates at facilities performing one-step and two-step testing to determine whether the testing protocol may be influencing CDI rate reporting. Methods: Data on HO-HCFA CDI rates entered monthly in fiscal year (FY) 2018 (October 2017 through September 2018) and FY2019 (October 2018 through September 2019) by each acute-care facility into the VA Inpatient Evaluation Center (IPEC) database were analyzed. HO-HCFA CDI rates in facilities that used NAAT in FY2018 and switched to in the 2-step NAAT plus EIA in FY2019 were compared to rates in facilities used NAAT alone for both FY2018 and FY2019. Statistical regression analysis was performed. Results: From FY 2018 through FY2019, 70 facilities performed NAAT for the entire 2-year period. Overall, 7 facilities performed NAAT for FY 2018 and then switched to NAAT + EIA in FY2019. We detected no significant decrease in HO-HCFA CDI rates in FY 2018 when both groups were using NAAT (P = 0.21) (Fig. 1). However, in FY2019, there was a significant decrease in HO-HCFA CDI rates for those facilities that performed the 2-step testing versus those facilities that continued to use strictly NAAT alone (P < .0001). Conclusions: HO-HCFA CDI rates decreased for those VA acute-care facilities that switched to 2-step testing, and this finding highlights implications for assessing rates over time. Given the variable reliability of the toxin test, individual patient consideration for therapeutic decisions is reasonable.

Funding: None

Disclosures: None