1114. Utility of a Bedside Diagnostic Testing Algorithm to Screen for Hospital-Onset Clostridium difficile Infection

Abstract

BackgroundMolecular assays have improved C. difficile detection in hospitalized patients. However, asymptomatic carriers have been misclassified as hospital onset C. difficile infection (HO-CDI), which has implications for management and infection prevention programs. At our facility, we implemented robust antibiotic stewardship policies in 2016 and had an SIR for HO-CDI of 0.73 for the year. In Q1 2017, this increased to 1.88. These cases revealed that nearly all tests, found positive for C. difficile, did not meet the standard definition of clinically significant diarrhea (CSD). Moreover, many patients did not have a clinical change in condition that supported a diagnosis of C. difficile. We reasoned that an algorithm for appropriate testing for C. difficile would significantly reduce our perceived rates of HO-CDI. We also reasoned that this tool could efficiently be used at the bedside during a clinical assessment.MethodsTo determine which patients had CSD, we designed, educated on and implemented an algorithm to screen for appropriate testing. It required three major elements: three or more loose stools in 24 hours, no gastric motility agents 48 hours prior, and a clinical change in condition (e.g., leukocytosis, fever, abdominal cramping). The completed algorithm accompanied the stool specimen and was required for testing. We evaluated each submitted algorithm for method validation. From this, we determined testing appropriateness and algorithm tool selectivity.ResultsOne year pre- and post-algorithm periods (PR-A and PO-A, respectively) were defined. Following its introduction, we noted a 57% decline in rates of HO-CDI (23 cases PR-A vs. 10 cases PO-A), and a 44% reduction in tests sent for C. difficile (average of 41 tests/month PR-A vs. 23 tests/month PO-A). We only used NAAT testing. We also noted a marked rise in adherence to the algorithm as time elapsed. The PDSA tool was used to refine the algorithm, with improved utilization by providers.ConclusionA simple bedside algorithm leads to more appropriate testing of patients for HO-CDI. A significant decline in reported rates of HO-CDI was noted. There is an additional benefit of diagnostic stewardship, as fewer tests are sent. This tool can be used immediately and independent of an electronic health record, is very cost effective, and is applicable to hospitals with low rates of HO-CDI.Disclosures R. V. Nathan, Merck, Allergan, The Medicines Co.: Speaker’s Bureau, Speaker honorarium.