Determinates of Clostridioides difficile infection (CDI) testing practices among inpatients with diarrhea at selected acute-care hospitals in Rochester, New York, and Atlanta, Georgia, 2020-2021

• Authors: Scott K Fridkin, Emory University; Udodirim N Onwubiko, Emory University; William Dube, Emory University; Chad Robichaux, Emory University; Jessica Traenkner, Emory University; Dana Goodenough, Atlanta Veterans’ Affairs Medical Center; Frederick J Angulo, Pfizer Vaccines, Collegeville; Joann M Zamparo, Pfizer Vaccines, Collegeville; Elisa Gonzalez, Pfizer Vaccines, Collegeville; Sahil Khanna, Mayo Clinic Rochester; Christopher Myers, University of Rochester; Ghinwa Dumyati, University of Rochester
• Journal Title: INFECTION CONTROL AND HOSPITAL EPIDEMIOLOGY
• Volume: Volume 44, Number 7
• Publisher: CAMBRIDGE UNIV PRESS | 2022-09-14, Pages 1085-1092
• Type of Work: Article | Final Publisher PDF
• Publisher DOI: http://doi.org/10.1017/ice.2022.205
• Subjects: Health Sciences, Public Health; Environmental Sciences; Engineering, Biomedical
• Keywords: Science & Technology; Life Sciences & Biomedicine; Public, Environmental & Occupational Health; Infectious Diseases; UNDERDIAGNOSIS; RATES

Abstract

Objective: We evaluated the impact of test-order frequency per diarrheal episodes on Clostridioides difficile infection (CDI) incidence estimates in a sample of hospitals at 2 CDC Emerging Infections Program (EIP) sites. Design: Observational survey. Setting: Inpatients at 5 acute-care hospitals in Rochester, New York, and Atlanta, Georgia, during two 10-workday periods in 2020 and 2021. Outcomes: We calculated diarrhea incidence, testing frequency, and CDI positivity (defined as any positive NAAT test) across strata. Predictors of CDI testing and positivity were assessed using modified Poisson regression. Population estimates of incidence using modified Emerging Infections Program methodology were compared between sites using the Mantel-Hanzel summary rate ratio. Results: Surveillance of 38,365 patient days identified 860 diarrhea cases from 107 patient-care units mapped to 26 unique NHSN defined location types. Incidence of diarrhea was 22.4 of 1,000 patient days (medians, 25.8 for Rochester and 16.2 for Atlanta; P <.01). Similar proportions of diarrhea cases were hospital onset (66%) at both sites. Overall, 35% of patients with diarrhea were tested for CDI, but this differed by site: 21% in Rochester and 49% in Atlanta (P <.01). Regression models identified location type (ie, oncology or critical care) and laxative use predictive of CDI test ordering. Adjusting for these factors, CDI testing was 49% less likely in Rochester than Atlanta (adjusted rate ratio, 0.51; 95% confidence interval [CI], 0.40-0.63). Population estimates in Rochester had a 38% lower incidence of CDI than Atlanta (summary rate ratio, 0.62; 95% CI, 0.54-0.71). Conclusion: Accounting for patient-specific factors that influence CDI test ordering, differences in testing practices between sites remain and likely contribute to regional differences in surveillance estimates.