老年轻症急性胰腺炎患者跌倒风险与住院时间延长的关系：基于 RCS曲线与阈值效应分析 [中文引用][英文引用]

目的 探讨老年轻症急性胰腺炎患者跌倒风险与住院时间延长的关系，为临床护理及管理决策提供依据。方法 选取 2021 年 10 月至 2024 年 5 月本院收治的 232 例老年轻症急性胰腺炎患者为研究对象。采用一般资料调查表、修订版 Morse 跌倒评估量表（Morse fall scale，MFS）进行调查，并根据住院时间划分为延长组、非延长组。采用 Logistic 回归分析住院时间延长的影响因素，通过控制不同变量的 Logistic 回归分析跌倒风险对住院时间延长的影响。结果 229 例患者完成研究。患者的平均住院时间（9.8±5.5）d，中位住院时间 9（6，12）d，住院时间延长者 53 例（23.14%）；跌倒风险评分（5.9±3.0）分，中位评分 6（4，8）分。多因素 Logistic 回归分析显示，糖尿病（OR=3.005，95%CI=1.380~6.542）、有营养风险（OR=2.393，95%CI=1.220~4.695）、跌倒风险（OR=1.233，95%CI=1.098~1.385）是老年轻症急性胰腺炎患者住院时间延长的影响因素（均 P<0.05）。通过控制不同变量多模型进行 Logistic 回归分析，结果显示，分别在控制了性别、年龄、体质量指数，吸烟史、饮酒史、高血压、糖尿病，文化程度、居住方式，主要病因、发病至入院时间、医疗费用支付方式、营养风险后，跌倒风险与住院时间延长仍存在相关（OR=1.278，95%CI=1.126~1.451，P<0.001）。限制性立方样条（restricted cubic spline，RCS）曲线分析显示，跌倒风险与住院时间延长之间存在非线性关系（Pfor overall<0.001，Pfor nonlinear<0.05）。阈值效应分析显示，当跌倒风险评分≥ 8 分时，评分每增加一个单位，住院时间延长的风险增加 2.3 倍（OR=2.300；95%CI=1.454~3.637；P<0.001）。结论 跌倒风险与糖尿病、营养风险共同影响老年轻症急性胰腺炎患者住院时间延长，其中跌倒风险对住院时间延长还具有独立影响性；当跌倒风险评分≥ 8 分时住院时间延长的风险随评分升高而增加。

Objective To explore the relationship between the risk of falls and the prolonged hospital stay in elderly patients with acute pancreatitis and to provide evidences for the decision-making of nursing and patient management. Methods A total of 232 elderly patients with acute pancreatitis admitted to our hospital from October 2021 to May 2024 were included as the study objects. General information questionnaire and revised Morse fall scale（MFS）were used for investigation. The patients were divided into a prolonged group and a non-prolonged group according to the length of hospital stay. Logistic regression was used to identify the factors that influenced the prolonged hospital stay. The effect of a risk of falls on prolonged hospital stay was analysed by Logistic regression with controlled variables. Results A total of 229 patients completed the study. The average hospital stay of the 229 patients was （9.8±5.5）days, with a median of 9 days（interquartile range：6, 12 days）. A total of 53 patients（23.14%）had a prolonged hospital stay. The average score of the risk of falls was（5.9±3.0）, with a median score of 6（interquartile range：4, 8）. Multivariate Logistic regression analysis showed that the factors influencing the prolonged hospital stay were diabetes（OR=3.005, 95%CI=1.380-6.542）, nutritional risk （OR=2.393, 95%CI=1.220-4.695）, risk of falls（OR=1.233, 95%CI=1.098-1.385）（all P<0.05）. Further logistic regression analysis was conducted using multiple models and controlled variables, including gender, age, body mass index, history of smoking, history of alcohol consumption, hypertension, diabetes, education, lifestyle, primary disease, time from morbidity to admission, payment method for medical care, and nutritional risk. The results showed that the risk of falls was associated with prolonged hospital stay even after the adjustment of variables（OR=1.278, 95%CI=1.126-1.451, P<0.001）. Restricted cubic spline（RCS）curve analysis demonstrated that there was a nonlinear relationship between the risk of falls and the prolonged hospital stay（Pfor overall<0.001, Pfor nonlinear<0.05）. Threshold effect analysis indicated that while the score of risk of falls was ≥ 8, then per additional increase in the score was associated with a 2.3-fold of increase in the risk of prolonged hospital stay（OR=2.300, 95%CI=1.454-3.637, P<0.001）. Conclusion The risk of falls, in conjunction with diabetes and a nutritional risk, affects the prolonged hospital stay in elderly patients with acute pancreatitis. The risk of falls also affect independently on a prolonged hospital stay. When the score of risk of falls is ≥ 8, the risk of prolonged hospital stay increases with the increment of the fall risk score.