Sensitivity and specificity of the Braden Scale in the cardiac surgical population.
PURPOSE: A descriptive study was conducted to investigate the sensitivity and specificity of the Braden Scale for Predicting Pressure Ulcer Risk in a cardiac surgical population. PATIENTS AND SETTING: A convenience sample of 337 pressure ulcer-free patients undergoing cardiothoracic surgery at a large midwestern national referral center were enrolled in the study. METHODS: Systematic skin and Braden Scale assessments were completed independently on the day of surgery and on postoperative days 1, 3, and 5. The presence of a pressure ulcer was determined and classified using the 4-stage scale developed by the WOCN Society. RESULTS: Sixteen patients (4.7%) developed a total of 22 pressure ulcers. Sensitivity and specificity of Braden scores were calculated for the day of surgery and for postoperative days 1, 3, and 5. The established Braden "cutoff" score of < or = 16 to identify those "at risk" had poor specificity and sensitivity in this patient population. The appropriate cutoff score varied by hospital day. A preoperative Braden score of 22 correctly classified 50% of the pressure ulcer-positive patients. The appropriate cutoff scores on postoperative day 1, 3, and 5 were 13, 14, and 20, respectively. Those scores correctly classified 67% of the pressure ulcer-positive patients on postoperative day 1, 57% on postoperative day 3, and 50% on postoperative day 5. CONCLUSION: These results illustrate that optimum prediction of pressure ulcer risk can only be accomplished with reassessments and determination of the Braden cutoff score or scores that are reflective of the patient's changing clinical condition throughout the hospitalization.