Predictive Risk Factors for Methicillin-Resistant Staphylococcus aureus (MRSA) Colonisation among Adults in Acute Care Settings: A Systematic Review.

BACKGROUND Asymptomatically colonised MRSA carriers serve as a substantial reservoir for person-to-person transmission of MRSA in the acute care setting. Although many studies have evaluated prognostic risk factors for MRSA colonisation on patient admission to an acute care setting, a comprehensive review of all the prognostic risk factors was not identified in a preliminary search of the literature. OBJECTIVES A systematic review was performed to identify and evaluate the association between risk factors and MRSA colonisation. INCLUSION CRITERIA Studies that included all adult patients on admission in acute care settings were considered in this review.All independent risk factors of MRSA colonisation were analysed in this review.Cohort and case-control studies are main designs associated identifying the independent risk factors for MRSA colonisation.The primary outcome of interest was presence and absence of MRSA on admission, and then independent risk factors associated with MRSA colonisation on admission were identified. SEARCH STRATEGY MEDLINE, EMABSE, and CINAHL databases were searched for prognostic studies published between 1990 and 2010 that examined the association between risk factors and MRSA colonisation. The search included both published and unpublished studies written in the English language. METHODOLOGICAL QUALITY Included studies were assessed using a standardised critical appraisal instrument that was developed for prognostic studies in infection control field. DATA COLLECTION Data were collected from included papers in the review using the standardised data extraction tool from the JBI SUMARI Program; and the data extraction form was modified based on the characteristics of prognostic studies for infection control. DATA SYNTHESIS All risk factors in included studies were aggregated depending on their clinical characteristics. Data of any aggregated factors was pooled into meta-analysis based on univariate estimates and multivariate estimates separately when more than two groups of data in selected studies were available. RESULTS Fifteen prospective studies, including a total 16,467 patients, were eligible for inclusion in the meta-analyses. More than 30 independent risk factors were identified and aggregated. The risk factors associated with MRSA colonisation in the meta-analyses include hospitalisation within the last 24 months, previous admission to a long-term care facility (LTCF) or a rehabilitation facility within the last 18 months, antibiotic use within the past 12 months, the presence of skin lesion, surgical intervention within the last 60 months, indwelling urinary catheter, intensive care unit (ICU) admission in the last 5 years, previous MRSA colonisation, intra-hospital transfer, male sex, comorbidity of chronic health evaluation class C or D, and the presence of fatal illness. CONCLUSIONS The identification of risk factors for MRSA colonisation on admission may contribute to improved effectiveness and efficiency of current MRSA prevention strategies and control MRSA spread and acquisition in acute care settings. The outcomes of this review may facilitate prediction model development to quickly identify potential MRSA carriers before admission.