Abstract:

Objective To analyze the applicability and the optimal cut-off of Braden Scale for the diagnosis of pressure injury in the elderly through the data of multicenter cross-sectional study of pressure injury in the elderly in China，so as to provide bases for the correct use of Braden scale in clinical practice. Methods 1067 trained and qualified nurses from 52 different medical institutions across the country were recruited in a group of 2. They included the inpatients over 60 years old by purpose sampling method，examined the skin and relevant data of patients，and evaluated whether there was pressure injury and its stage，and used the Braden Scale to predict the risk of pressure injury. The statistical software was used to analyze the epidemic characteristics of pressure injury in the elderly，and the area under the working characteristic curve（AUC），sensitivity，specificity and Yoden index of those with pressure injury were predicted and diagnosed by the Braden Scale. Results In the effective data of 3 3769 cases，the total prevalence of pressure injury was 3.12%（n=1 054），of which the incidence of hospital acquired pressure injury was 0.79%（n=267）. Hierarchical analysis showed that the AUC predicted by the Braden Scale for the diagnosis of pressure injury in elderly patients of different ages，gender，races and medical institutions was 0.85~0.96，and the sensitivity and specificity were 0.83~0.96 and 0.71~0.86，respectively. The AUC of the 6 cut-off values of total score≤15，≤16，≤17，≤18，≤19 and ≤20 points was 0.67~0.82; the sensitivity and specificity were 0.67~0.84 and 0.54~0.68，respectively; the Youden index was 0.25~0.49. The Youden index and AUC of ≤20 points was the largest. Conclusion Braden Scale is suitable for elderly patients of different ages，gender，race and medical institutions to assess and predict the risk of pressure injury. The total score of≤20 points can be used as the best cut-off，which can improve the prediction efficiency.

Key words: Aged, Pressure Injury, Braden Scale, Risk Assessment, Predictive Cut-off, Nursing Care

摘要：

目的 探讨Braden量表诊断老年人压力性损伤的适用性和最佳界值，为临床正确使用Braden量表提供依据。 方法 招募全国52所医疗机构，1 067名培训合格的护士每2~3人一组，采用目的抽样法纳入60岁以上住院患者，检查其全身皮肤并收集相关资料，判断是否存在压力性损伤及其分期，同时采用Braden量表预测评分。使用统计软件分析老年人压力性损伤流行特征和Braden量表预测诊断压力性损伤受试者工作特征曲线下面积（area under the curve，AUC）、灵敏度、特异度和约登指数。 结果 在33 769例有效资料中，压力性损伤总现患率为3.12%（1 054/33 769），其中医院获得性压力性损伤发生率为0.79%（267//33 769）。分层分析显示，Braden量表预测诊断不同年龄、性别、种族和不同医疗机构中老年患者压力性损伤的AUC为0.85~0.96，灵敏度和特异度分别为0.83~0.96和0.71~0.86。总分 ≤15、≤16、≤17、≤18、≤19、≤20分6个界值的AUC为0.67~0.82，灵敏度和特异度分别为0.67~0.84和0.54~0.68，约登指数为0.25~0.49，≤20分的约登指数和AUC最大。 结论 Braden量表适用于不同年龄、性别、种族和不同医疗机构中的老年患者评估预测压力性损伤风险，总分≤20分作为最佳界值可提高预测效能。

关键词: 老年人, 压力性损伤, Braden量表, 危险评估, 预测界值, 护理