OBJECTIVES: Hospital-acquired pressure ulcers are localized skin injuries that cause significant morbidity and mortality, and are costly to treat. Following the International Guidelines for pressure ulcer prevention reduce rates over standard care, but require risk-assessment among all patients each nursing-shift using the Braden Scale. The relative value of only continually risk-assessing at-risk patients – about 7% of hospitalizations – is uncertain. Our objective was to analyze the cost-effectiveness of nursing resources to risk-assess all patients or select risk-groups identified by machine learning compared to standard care. METHODS: We obtained patient-level data on 43,787 encounters from a university hospital EHR between 2011-2014, including daily Braden scores. Longitudinal arrays of adult inpatients with ≥10 Braden scores were classified into 5 risk-groups: very high-risk [6-9]; high-risk [10-11]; moderate-risk [12-14]; at-risk [15-18]; low-risk [19-23]. We constructed a 10-state Markov model using supervised machine learning to calculate age-adjusted transition probabilities between risk-groups, pressure ulcer states and discharge states. Costs (2015 $US) and lifetime quality-adjusted life years (QALYs) related to pressure ulcer outcomes were added to the Markov model to compare the cost-utility of standard care, continual risk-assessment in all risk-groups or only follow-up risk-assessment in higher risk-groups. Probabilistic sensitivity analysis (PSA) tested model uncertainty at $100,000/QALY willingness-to-pay from the U.S. societal perspective. RESULTS: Risk-assessing all patients every nursing-shift was cost-effective ($5,193, 15.43 QALYs) compared to standard care ($5,139, 14.00 QALYs) at an increment cost-effectiveness ratio of $37/QALY. Follow-up risk-assessment only among patients with Braden scores <19 dominated standard care ($1,745, 14.03 QALYs) in 50.89% of PSA simulations with small QALY gains. Follow-up among patients with lower Braden scores decreased costs at fewer QALYs. CONCLUSIONS: Our analysis using real-world data maintains that follow-up risk-assessment among all patients cost-effectively prevents pressure ulcers. Hospitals should encourage nurses’ adherence to International Guidelines on risk-assessment to prevent patient harm.