ABSTRACT
BACKGROUND: Colonization with Staphylococcus aureus is a well-defined risk factor for disease in hospitals, which can range from minor skin infections to severe, systemic diseases. However, the generalizability of this finding has not been thoroughly investigated outside of the hospital environment. We aimed to assess the role of S. aureus colonization as a risk factor for disease in the community. METHODS: We performed a meta-analysis of observational studies and searched PubMed for articles published between December 1979 and May 23, 2016. We included cohort, cross-sectional, and case-control studies that reported quantitative estimates of both S. aureus colonization and disease statuses of all study subjects. We excluded studies on recently hospitalized subjects, long-term care facilities, surgery patients, dialysis patients, hospital staff, S. aureus outbreaks, and livestock-associated infections. Our meta-analysis was performed using random-effects analysis to obtain pooled odds ratios (ORs) to compare the odds of S. aureus disease with respect to S. aureus colonization status. RESULTS: We identified 3477 citations, of which 12 articles on 6998 subjects met the eligibility criteria. Overall, subjects colonized with S. aureus were more likely to progress to disease than those who were non-colonized: (OR 1.87, 95% CI 1.21-2.88, n = 7 studies). We observed a larger effect with methicillin-resistant S. aureus colonization (7.06, 4.60-10.84, n = 7 studies). However, the methicillin-sensitive S. aureus colonization was not associated with greater odds of disease (1.20, 0.69-2.06, n = 4 studies). Heterogeneity was present across studies in all of the subgroups: S. aureus (I2 = 95.0%, χ2 = 120.3, p < 0.001), MRSA (I2 = 92.8%, χ2 = 82.8, p = p < 0.001), and MSSA (I2 = 86.3%, χ2 = 21.8, p < 0.001). CONCLUSIONS: While the majority of papers individually support the assumption that colonization is a risk factor for S. aureus disease in the general population, there is marked heterogeneity between studies and further investigation is needed to identify the major sources of this variance. There is a shortage of literature addressing this topic in the community setting and a need for further research on colonization as a focus for disease prevention.
