Time-series analysis of the relationship of antimicrobial use and hand hygiene promotion with the incidence of healthcare-associated infections

We analyzed the effect of antimicrobial use and implementation of a hand hygiene program on the incidence of healthcare-associated infections (HAIs) and healthcare-associated methicillin-resistant Staphylococcus aureus (HA-MRSA) infections at the Chung Shan Medical University Hospital (Taichung, Taiwan). Monthly data were retrospectively reviewed from January 2004 to December 2010. Use of antimicrobials and alcohol-based hand cleaner were separately regressed against the incidences of HAIs and HA-MRSA infections. Infection incidence was expressed as persons per 1000 patient days (PDs), monthly use of i.v. antibiotics was expressed as defined daily doses per 1000 PDs and monthly alcohol-based hand cleaner use was expressed as bottle per 1000 PDs. Multivariate analysis indicated that use of hand cleaner was associated with reduced incidence of HAIs (P=0.0001) and HA-MRSA infections (P<0.0001). Time-series analysis indicated that increased use of hand cleaner was significantly associated with significant decreases in the incidences of HAIs and HA-MRSA infections. Total antibiotic use had no significant effect on HAIs, but was associated with more HA-MRSA infections. In addition, the use of J01CR01 antibiotics (combinations of penicillins, including Β-lactamase inhibitors) in particular was correlated with significantly increased incidence of HA-MRSA infections. Our forecasting model demonstrates the efficacy of a hand hygiene program and the need to limit the use of certain restricted antimicrobials in order to reduce the incidence of HAIs and HA-MRSA infections.