Effectiveness of screening protocols to reduce MRSA colonization in a pediatric hospital.

BACKGROUND: Methicillin-resistant Staphylococcus aureus (MRSA) is a major nosocomial pathogen that can be carried asymptomatically or cause invasive infections. Screening for asymptomatic carriers in healthcare settings reduces MRSA transmission and limits the burden of infections. METHODS: We calibrated an individual-based model to replicate MRSA colonization diagnoses in eight units at higher MRSA risk of an academic Italian pediatric hospital. We used the calibrated model to gain insights on transmission dynamics and on the epidemiological impact of different screening strategies. RESULTS: Under existing screening protocols, we estimated an effective reproduction number of 0.57 (95% Prediction Interval, PrI: 0.41-0.71), with highly heterogeneous transmission: 69% of carriers did not transmit MRSA, while 10% accounted for 58% of hospital-acquired colonizations. These superspreaders had significantly longer hospital stays (mean 81 days) compared to the remaining 90% (mean 30 days). Under all considered scenarios, the hospital setting remained a key reservoir and transmission hub for MRSA, given that MRSA transmission can occur repeatedly under non-outbreak conditions. Removing screening would increase the reproduction number to 1.01 (95% PrI: 0.68-1.38). Adding a biweekly screening to implemented protocols could reduce the incidence of MRSA colonizations by about 33% at the cost of a 56% increase in the number of screening tests. However, a targeted screening of patients selected by a trained machine learning classifier would achieve comparable reductions with only a 20% additional screening effort. CONCLUSIONS: These results suggest that targeted, risk-based screening algorithms may improve the cost-benefit of MRSA control strategies in hospitals and reduce the spillover to the community.