Impact of a Healthcare Provider Educational Intervention on Frequency of Clostridium difficile Polymerase Chain Reaction Testing in Children: A Segmented Regression Analysis.

Abstract

Although Clostridium difficile infections (CDIs) are increasingly diagnosed in children, many children diagnosed with CDI lack classic risk factors. Frequent use of highly sensitive tcdB polymerase chain reaction (PCR) testing in low-risk patients leads to CDI misdiagnosis and unnecessary CDI antibiotic use in children with C difficile carriage. For this quasi-experimental study, we developed and implemented an educational intervention (EI) to inform healthcare providers (HCPs) about tcdB PCR test limitations. We provided HCP didactic education and built an electronic notification into the tcdB PCR test order that describes scenarios in which carriage is more likely than CDI. Segmented regression analysis assessed changes in level (ie, overall rates) and trend of C difficile testing rate ([TR] number of tests performed per 1000 patient encounters) and test positivity rate ([PR] number of positive tests per 1000 patient encounters) between the pre- (August 2009-August 2013) and postintervention (February 2014-July 2015) periods. Hospital-wide, absolute TR reduction was 0.71 (P[level] = .0067; P[trend] = .0042) and absolute PR reduction was 0.14 (P[level] = .22; P[trend] = .018). In the outpatient setting, absolute TR reduction was 0.30 (P[level] = .0015; P[trend] < .001) and absolute PR reduction was 0.09 (P[level] = .0069; P[trend] = .046). The incidence density of healthcare facility-associated CDI did not significantly change after the EI. The EI was associated with avoidance of 574 tests and 113 positive tests (and subsequent antibiotic courses) during the postintervention period, which saved approximately $250 000 in patient charges related to CDI testing and treatment. Healthcare provider education can cost-effectively reduce the frequency of C difficile testing and CDI misdiagnosis by improving test utilization among low-risk children.