Skip to main content
Effective Health Care Program

Cluster Randomized Trial of Hospitals to Assess Impact of Targeted Versus Universal Strategies to Reduce Methicillin-Resistant Staphylococcus aureus (MRSA) in Intensive Care Units

Abstract Archived

Topic Abstract

Background: Despite rising MRSA disease and deaths among the chronically and critically ill, it is not known how best to respond. Several viable options have been implemented in intensive care units (ICUs) and even put forward as mandates in a handful of states. However, there is little information about which of these options is most effective in reducing MRSA disease. The implications of legislating one of these options increases the urgency for an intervention trial to identify the most effective strategy.

Objective: There is a compelling need to develop a solid evidence base to guide policy and practice regarding prevention of healthcare associated MRSA infection. The Randomized Evaluation of Decolonization versus Universal Clearance to Eliminate MRSA (REDUCE MRSA) Trial is a cluster randomized trial of the comparative effectiveness of three strategies to prevent MRSA in intensive care units.

Study Design and Methods: Each of the 3 study arms will be comprised of approximately 15 Hospital Corporation of America (HCA) facilities (~50 total). The three strategies employed by each arm to be evaluated are: a) screening on admission followed by isolation of MRSA+ patients, b) screening on admission followed by isolation and decolonization of MRSA+ patients, and c) universal decolonization on admission with no screening. The decolonization regimen involves bathing with chlorhedixine plus intra-nasal application of mupirocin.


Study outcomes will consist of an evaluation of routinely reported central line associated blood stream infections (CLBSI), an assessment of trial impact on monthly percent of ICU patients with any nosocomial MRSA+ clinical culture; and an asessment of trial impact on monthly nosocomial rates of MRSA and all-pathogen bacteremias and urinary tract infections.


The investigators will perform regression analysis across all ICUs predicting outcomes of interest, while controlling for a number of variables. Bias will be minimized by using randomized hospital assignments. Analyses will be assessed by randomization assignment, regardless of whether cross over occurs.

The study is a partnership between AHRQ, the CDC, the CDC Prevention Epicenters, and the Hospital Corporation of America.

Expected outputs: Three scientific reports.

Expected date of project completion: 3/30/2011