Community-onset (CO) methicillin-resistant Staphylococcus aureus (MRSA) pneumonia is an evolving problem, and there is a great need for a reliable method to assess MRSA risk at hospital admission. A new MRSA prediction score classifies CO-pneumonia patients into low, medium, and high-risk groups based on objective criteria available at baseline. Our objective was to assess the effect of initial MRSA therapy on mortality in these three risk groups. We conducted a retrospective cohort study using data from the Veterans Health Administration. Patients were included if they were hospitalized with pneumonia and received antibiotics within the first 48 hours of admission. They were stratified into MRSA therapy and no MRSA therapy treatment arms based on antibiotics received in the first 48 hours. MRSA risk groups were analyzed separately. The primary outcome was 30-day patient mortality. Multivariable logistic regression was used to adjust for potential confounders. A total of 80,330 patients met inclusion criteria, of which 36% received MRSA therapy and 64% did not receive MRSA therapy. The majority of patients were classified as either low (51%) or medium (47%) risk, with only 2% classified as high-risk. In the high-risk group, unadjusted 30-day mortality was lower among patients who received initial MRSA therapy (40% versus 58%; p<0.0001). Likewise, multivariable logistic regression analysis also demonstrated that initial MRSA therapy was associated with a lower 30-day mortality in the high-risk group (adjusted odds ratio 0.57; 95% confidence interval 0.42-0.77). There was no benefit of initial MRSA therapy in the low or medium-risk groups. This study demonstrated improved survival with initial MRSA therapy in high-risk CO-pneumonia patients. The MRSA risk score should not replace clinical judgment, but it might be a useful tool to spare MRSA therapy for only those patients who are most likely to benefit. / text