The Braden Scale is a globally recognized pressure injury risk prediction tool with good reliability and validity. However, due to different application populations, its optimal cut-off value is controversial. The optimal cut-off value of the scale in clinical application in China is ≤ 16 points, while most European countries and the US take ≤18 points as the cut-off value. There is insufficient research on the predictive value of the scale in hospitalized patients in China.

To carry out a cross-sectional study on the epidemiological characteristics of pressure injuries among adult inpatients in 46 tertiary hospitals in 13 provinces/autonomous regions across China, and to examine the predictive values and optimal cut-off values of the Braden Scale and its six subscales.

Established by the Eastern Theater General Hospital, the research group of this multicenter study publicly recruited 46 eligible grade A tertiary hospitals containing at least 500 beds in 13 provinces/autonomous regions of China as the multicenter research settings after qualification review and signing a bilateral agreement on participating in the research. Adult patients (≥18 years old) with hospitalization time ≥24 hours were included as the patient participants, and senior wound care nurses with at least two years of working experience (n=1 060) were included as investigators. On March 30 and May 29, 2021, the nurses used a special survey record form to record the general situation of the patients, and checked their skin of the whole body to find out if there was a pressure injury, and used the six subscales of the Braden Scale (sensory perception, moisture, activity, mobility, nutrition, friction and shear) to assess the risk of pressure injuries. ROC curves of the scale and its subscales were plotted to assess their predictive values and optimal cut-off values for pressure injuries.

In total, the valid data of 60 555 patients were obtained. The prevalence of pressure injuries, and hospital-acquired pressure injuries was 1.67% (1 010/60 555), and 0.74% (448/60 555), respectively. The prevalence of pressure injuries in males was higher than that of females〔2.08% (677/32 518) vs 1.19% (333/28 037) 〕 (χ²=73.394, P<0.001). And the prevalence of pressure injuries in those hospitalized in the ICU〔10.58% (324/3 061) 〕 and those aged 80 years and over was higher〔5.98% (311/5 198) 〕. Logistic regression analysis showed that age, sex, serum albumin, hemoglobin, incontinence type, bed rest and Braden score were associated with pressure injuries (P<0.05). The area under the ROC curve (AUC) of the Braden Scale for predicting pressure injuries was 0.95〔95%CI (0.94, 0.95) 〕, with 96% sensitivity and 83% specificity, and 0.79 Youden index, as well as its corresponding optimal cut-off value of 17.50 points. In assessing the risk of pressure injuries in patients in different departments and age groups, the AUC of the Braden Scale ranged from 0.78 to 0.97, with a sensitivity ranging from 83% to 97%, a specificity ranging from 61% to 91%, and the Youden index ranged from 0.47 to 0.88, the optimal cut-off value ranged from 13.50 to 18.50 points. The AUC of the friction and shear predicting the risk of pressure injuries was 0.87, with 0.56 as the Youden index, and 1.50 points as its associated optimal cut-off value, and the AUC of mobility was 0.84, with 0.56 as the Youden index, and 2.50 points as its associated optimal cut-off value, and the AUC of activity was 0.80, with 0.54 as the Youden index, and 1.50 points as its associated optimal cut-off value.

The prevalence of pressure injuries, and hospital-acquired pressure injuries among adult inpatients in 46 tertiary hospitals in 13 provinces/autonomous regions across China is 1.67% (1 010/60 555), and 0.74% (448/60 555), respectively. The Braden Scale is applicable to adult inpatients in tertiary hospitals in China. The total score ≤18 points may be the optimal cut-off. Three subscales, namely friction and shear, mobility and activity have higher predictive values. In case of insufficient medical workers or special settings, the three subscales can be used to predict the risk of pressure injury, but their optimal cut-off values still need to be further verified in applications.

Braden量表是全球公认的有良好信效度的压力性损伤危险预测工具，但因应用人群不同，其最佳预测界值也存有争议。该量表在我国临床应用的预测界值为≤16分，欧美国家多以≤18分为界值，但其在我国住院患者中的预测作用的研究较少。

横断面调研全国13个省/自治区、46所三级医院成年住院患者压力性损伤的流行病学特征，分析Braden量表及其6个分量表预测压力性损伤发生的效能及其最佳界值。

由解放军东部战区总医院负责成立多中心合作研究课题组，在全国公开招募≥500张床位的三甲医院，经过资格审查和签订合作研究双向协议后，共纳入13个省/自治区、46所符合条件的医院作为多中心研究单位，纳入住院时间≥24 h、年龄≥18岁的患者为研究对象。入选护龄≥2年的伤口护理骨干护士作为调研人员，共入选1 060名调研护士。在2021-03-30和2021-05-29，使用专用调查记录表记录患者的一般情况；检查全身皮肤，判断是否发生压力性损伤及分期情况；采用Braden量表（从感知觉、潮湿度、活动能力、移动能力、营养、摩擦和剪切力6个分量表逐项评分）判断压力性损伤的发生危险。绘制受试者工作特征（ROC）曲线，分析Braden量表及其6个分量表预测成年住院患者发生压力性损伤的效能及其最佳界值。

本次调研共获得60 555例有效数据，其中压力性损伤现患率为1.67%（1 010/60 555），医院获得性压力性损伤现患率为0.74%（448/60 555）。男性压力性损伤现患率〔2.08%（677/32 518）〕高于女性〔1.19%（333/28 037）〕（χ²=73.394，P<0.001）。≥80岁和ICU患者压力性损伤现患率最高，分别为5.98%（311/5 198）和10.58%（324/3 061）。多因素Logistic回归分析结果显示，年龄、性别、血清白蛋白、血红蛋白、失禁类型、卧床、Braden量表总分是成年住院患者发生压力性损伤的影响因素（P<0.05）。Braden量表预测成年住院患者发生压力性损伤的ROC曲线下面积（AUC）为0.95〔95%CI（0.94，0.95）〕，灵敏度和特异度分别为96%和83%，约登指数为0.79，对应的最佳界值为17.50分。Braden量表预测不同科室和年龄患者发生压力性损伤的AUC为0.78~0.97，灵敏度为83%~97%，特异度为61%~91%，约登指数为0.47~0.88，最佳界值为13.5~18.5分。摩擦和剪切力、移动能力和活动能力3个分量表预测成年住院患者压力性损伤发生危险的AUC分别为0.87、0.84、0.80，约登指数分别为0.56、0.56、0.54，最佳界值分别为1.50、2.50、1.50分。

本次调研得出我国13个省/自治区、46所三级医院成年住院患者压力性损伤现患率为1.67%（1 010/60 555），医院获得性压力性损伤现患率为0.74%（448/60 555）。Braden量表适用于我国三级医院住院患者预测压力性损伤发生危险，总分≤18分有更好的预测作用，其中摩擦和剪切力、移动能力和活动能力分量表的预测作用最大，在人员紧张或特殊环境下可采用此3个分量表预测压力性损伤危险，最佳界值的预测作用需在应用中进一步验证。