Analysis of interventions to reduce the incidence of Clostridium difficile infection at a London teaching hospital trust, 2003-2011

Background: Since 2008 there has been a substantial fall in the incidence of Clostridium difficile infection (CDI) in the UK, though it is unclear what contribution local and governmental interventions have made to this reduction. Aim: To assess the value of retrospective analysis using a segmented regression model and single factor analysis of variance (ANOVA) in assessing the impact of interventions to reduce hospital-acquired CDI. Methods: A retrospective time-series analysis of 28 interventions implemented over an eight-year period by a large teaching hospital trust. Single factor ANOVA was used to analyse the association between CDI rates and changes in antibiotic usage and hand hygiene compliance data. Findings: Whereas several interventions were associated with a reduction in CDI rates, insufficient time-points were available between interventions to allow meaningful interpretation of all the data. Reduction of cephalosporin and quinolone use was associated with reduction in CDI rates. There was little association between observed variations in hand hygiene compliance and CDI rates. Conclusion: Though several interventions were associated with a reduction in CDI, this study illustrates the inherent limitations of retrospective time-series methodology for the investigation of infection control measures where large numbers of interventions are introduced over a short time-period. Single factor ANOVA demonstrated an association between fall in CDI and restriction of high-risk antibiotic use. Little association was demonstrated between compliance rates documented in hand hygiene audits and CDI, though compliance was relatively high throughout the period of study. (C) 2014 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.