This descriptive correlational study explored the predictive validity of the Braden Scale and factors affecting it A Braden score was determined within 4 hours of admission for 50 adult medical/surgical inpatients. Independent skin assessments were made three times a week and at discharge. Fourteen patients (28%) developed pressure ulcers. A Braden score cutoff of 18 or less resulted in a 71% sensitivity, 83% specificity, 63% predictive value of a positive test, and 88% predictive value of a negative test. Three of the four patients incorrectly predicted to be not at risk scored "inadequate" on the nutrition subscale. Two of the four also were underweight. Of the six patients incorrectly predicted at risk for a pressure ulcer, three had been placed on air mattresses and were receiving levothyroxine (Synthroid). This study provides further evidence of the Braden Scale's predictive validity. The results suggest that patients who are underweight or getting inadequate nutrition be considered at increased risk for pressure ulcers.