Feb 6, 2014
Published in: Medical Care, v. 52, no. 2, suppl. 1, Feb. 2014, p. S66-S73
Posted on RAND.org on January 31, 2014
BACKGROUND: The Surgical Care Improvement Project (SCIP) has developed a set of process compliance measures in an attempt to reduce the incidence of surgical site infections (SSIs). Previous research has been inconclusive on whether compliance with these measures is associated with lower SSI rates. OBJECTIVES: To determine whether hospitals with higher levels of compliance with SCIP measures have lower incidence of SSIs and to identify the measures that are most likely to drive this association. DATA AND METHODS: Analysis of linked SCIP compliance rates and SSIs on 295 hospital groups observed annually over the study period 2007–2010. A hospital group comprises all hospitals sharing identical categories for location by state, teaching status, bed size, and urban/rural location. We used a generalized linear model regression with logistic link and binomial family to estimate the association between 3 SCIP measures and SSI rates. RESULTS: Hospital groups with higher compliance rates had significantly lower SSI rates for 2 SCIP measures: antibiotic timing and appropriate antibiotic selection. For a hospital group of median characteristics, a 10% improvement in the measure provision of antibiotic 1 hour before intervention led to a 5.3% decrease in the SSI rates (P<0.05). Rural hospitals had effect sizes several times larger than urban hospitals (P<0.05). A third-core measure, Timely Antibiotic Stop, showed no robust association. CONCLUSIONS: This analysis supports a clinically and statistically meaningful relationship between adherence to 2 SCIP measures and SSI rates, supporting the validity of the 2 publicly available healthcare–associated infection metrics.