Objective: To explore the application value and predictive efficacy of three pressure ulcer risk assessment tools: Waterlaw scale, Braden scale and Norton scale, in patients with nasofacial pressure ulcers under noninvasive positive pressure ventilation. Methods: A total of 287 patients with chronic obstructive pulmonary disease who were treated with noninvasive positive pressure ventilation in our hospital from January 2017 to December 2018 were assessed with Waterlaw scale, Braden scale and Norton scale within 2 hours before or after noninvasive positive pressure ventilation. The area under the ROC curve, predictive value, critical value, sensitivity coefficient and specificity coefficient of the three pressure ulcer assessment tools were calculated. Results: There were 56 patients (19.5%) with nasofacial pressure ulcers. The scores of patients without nasofacial pressure ulcers on Waterlaw scale were significantly lower than those of patients with nasofacial pressure ulcers. The scores of Braden scale and Norton scale were significantly higher than those of patients with nasofacial pressure ulcers (P<0.001). The area under the ROC curve of Waterlaw scale, Braden scale and Norton scale was 0.769, 0.710 and 0.693, respectively, with statistical significance (P<0.05). The area under the curve of Waterlaw scale was significantly higher than that of Braden scale and Norton scale (Z=2.263 and 2.381, P<0.05). The positive and negative predictive values of the three scales were 61.191 and 89.082; 37.166 and 87.091; and 40.893 and 86.886, respectively. The critical scores were 23, 14 and 10, and the sensitivity and specificity coefficients were 53.570 and 91.770; 51.790 and 78.790; and 48.210 and 83.120, respectively. Conclusion: Waterlaw scale, Braden scale and Norton scale can predict the occurrence of nasofacial pressure ulcers in patients with noninvasive positive pressure ventilation to a certain extent. The predictive ability of Waterlaw scale was relatively high, while its sensitivity was relatively low.