The aim of this study was to develop a modified Braden scale, to evaluate its predictive validity, and to identify a more valid pressure ulcer risk calculator for application in acute care hospitals in Mainland China among the modified Braden, Braden, and Norton scales. The initial modified Braden scale, with the addition of skin type and body build for height, was proposed in this study. Four hundred twenty-nine subjects who were admitted to two acute care hospitals in Mainland China within 24 hr and free of pressure ulcers upon admission were assessed with the initial modified Braden, Braden, and Norton scales by three nurse assessors. This was followed by a daily skin assessment to note any pressure ulcer by a nurse assessor. Nine subjects had pressure ulcers detected at Stages I (89%) and II (11%) after an average stay of 11 days. The descriptive analysis of each subscale scoring item in the initial modified Braden scale indicated that skin type and body build for height were the most distinct predictive factors whereas nutrition was the least distinct factor for predicting pressure ulcer development. Based on these findings, the modified Braden scale was further developed with the addition of skin type and body build for height and by exclusion of nutrition. The predictive validity test reported that the modified Braden scale demonstrated a better balance of sensitivity (89%) and specificity (75%) at a cutoff score of 16, with a higher positive predictive value (7%), than the Braden and Norton scales. This finding revealed that for this sample, the modified Braden scale is more effective in pressure ulcer risk prediction than the other two scales. Because the modified Braden scale is not 100% sensitive and specific, to increase clinical efficacy in the prevention of pressure ulcer, it is recommended that it be adopted combined with nursing judgment to predict pressure ulcer development in acute care settings in Mainland China.