基于老年综合评估的睡眠障碍列线图的构建与应用

目的探讨老年住院患者睡眠障碍的影响因素,并基于老年综合评估的睡眠障碍构建列线图,以指导临床实践。方法通过回顾性分析本院老年住院患者临床资料,采用老年综合评估指标进行Logistic回归分析,确定睡眠障碍的影响因素,并构建预测模型;比较使用该模型的观察组与不使用该模型的常规诊疗对照组患者PSQI评分,评估模型的有效性。结果多因素Logistic回归分析显示,年龄、尿失禁(尿频)、疼痛评分、焦虑评分、抑郁评分是睡眠障碍的独立危险因素(P<0.05)。构建的模型预测老年住院患者睡眠障碍的ROC曲线下面积(AUC)为0.802(95%CI:0.755~0.850),表明模型具有较好的预测能力。观察组采用该模型进行老年住院患者睡眠障碍评估和针对性干预后,患者PSQI评分降低至(6.30±3.33)分,与对照组的(7.60±3.20)分比较,差异有统计学意义(P<0.05)。结论睡眠障碍预测模型具有较高的区分度及准确度,有助于临床医护人员准确评估老年住院患者的睡眠障碍,并进行有效干预,从而改善住院患者的睡眠质量。

Objective To explore the factors affecting sleep disorders in hospitalized elderly patients and to construct a nomogram for sleep disorders based on comprehensive geriatric assessment to guide clinical practice.Methods By retrospectively analyzing the clinical data of hospitalized elderly patients in our hospital,logistic regression analysis was performed using comprehensive geriatric assessment indicators to determine the factors affecting sleep disorders and to construct a predictive model.The effectiveness of the model was evaluated by comparing the Pittsburgh sleep quality index(PSQI)scores of patients in the observation group using the model with those in the control group without using the model.Results Multivariate logistic regression analysis showed that age,urinary incontinence(frequency of urination),pain score,anxiety score,and depression score are independent risk factors for sleep disorders(P<0.05).The receiver operating characteristic(ROC)curve area(AUC)of the constructed model for predicting sleep disorders in hospitalized elderly patients was 0.802(95%CI:0.755~0.850),indicating good predictive ability.After using the model for sleep disorder assessment and targeted intervention in the observation group,the PSQI score of patients was reduced to(6.30±3.326),which was statistically significantly different from the control group's score of(7.60±3.20)(P<0.05).Multivariate logistic regression analysis showed that age,urinary incontinence(frequency),pain score,anxiety score,and depression score were independent risk factors for sleep disorders(P<0.05).The area under the ROC curve(AUC)of the nomogram for predicting sleep disorders in elderly hospitalized patients was 0.802(95%CI:0.755-0.850),with the internal calibration curve close to the standard curve.After using the nomogram to assess sleep disorders in elderly hospitalized patients and implementing targeted interventions,the PSQI score of the intervention group was effectively reduced to 6.30±3.33,lower than the PSQI score of the control group(7.60±3.201),with statistical significance(P<0.05).Conclusion The sleep disorder prediction model has high discrimination and accuracy,which helps clinical medical staff to accurately assess sleep disorders in hospitalized elderly patients and to carry out effective interventions,thereby improving the sleep quality of hospitalized elderly patients.