Construction Of Influencing Factors And Risk Prediction Model Of Sleep Disturbance In Elderly Patients With CCU

Background:Studies at home and abroad have proved that sleep disorders are common in the elderly.With the rapid approaching of China’s aging society,the social burden of sleep disorders among the elderly is further aggravated.The elderly with the increase of age,their own physiological function decline,physical function decline,sleep disorders will further reduce the immunity of elderly patients,induce a variety of diseases,increase the burden of disease.At present,related studies have reported the sleep disorder of the elderly at home and its influencing factors.Some studies have discussed the sleep disorder of the elderly in general wards and its influencing factors,and some studies have discussed the sleep disorder of the elderly in intensive wards and its influencing factors.However,there were few reports on sleep disorders in elderly patients with Coronary Care Unit（CCU）.The occurrence of sleep disorders in elderly patients with CCU was more likely to result in the decline of immune system and cardiovascular system functions,and increase the risk of infection and mortality.Therefore,it is necessary to study the status and influencing factors of sleep disorder in elderly patients with CCU.Objective:To investigate the status of sleep disorder in elderly CCU patients,determine the influencing factors of sleep disorder in elderly CCU patients,and establish a risk prediction model of sleep disorder in elderly CCU patients.To predict the risk of sleep disorder in elderly patients with CCU,and identify high-risk individuals with sleep disorder as early as possible,provide reference for taking intervention measures to prevent sleep disorder,and lay a foundation for improving sleep quality in elderly patients with CCU.Methods:Convenience sampling method was used to conduct a cross-sectional survey of 430 elderly CCU patients in a third class a hospital in Shiyan City.The sleep status and related factors of elderly CCU patients were investigated by general data questionnaire,Chinese Version of Richards-Campbell sleep scale,pain number score,simple psychological status Scale,acute physiological and chronic health status scale and pre-hospitalization sleep quality questionnaire.SPSS22.0 statistical tool was used to analyze the data.Single factor analysis was used to screen the significant variables of sleep disturbance related factors in elderly CCU patients,p<0.05 represents statistically significant difference in variables..Logistic regression analysis and decision tree classification were used to construct risk prediction models of sleep disorder,respectively,and to explore the independent risk factors of sleep disorder in elderly patients with CCU.The software Med Calc19.1 was used to draw receiver operating characteristic curve（ROC）,calculate the sensitivity and specificity of the model and the area under the ROC curve Curve,AUC),and the prediction efficiency of the analysis model.Results:（1）The prevalence of sleep disorder in elderly patients with CCU was 58.6%.（2）There was significant difference in univariate analysis between the sleep disorder group and the non-sleep disorder group（p<05）were gender,residence status,cardiac function grade,daily living ability score,whether invasive operation was performed at night,simple psychological status score,noise score,pain score,acute physiological and chronic health status score,and sleep quality score before hospitalization.（3）Logistic regression analysis showed that invasive operation at night,simple psychological status score Ⅱ,Ⅲ,noise score,pain score,acute physiological and chronic health status score were independent risk factors for sleep disturbance in elderly CCU patients.（4）The result of the decision tree model based on the Chi-square automatic interactive detection method shows that the decision tree has 13 nodes,with 7 middle nodes and a depth of 4.The independent influencing factors of sleep disorder were pain score,noise score,acute physiological and chronic health status score,and gender simple psychological status score.（5）The sensitivity and specificity of Logistic regression prediction model were87.7% and 89.9%,respectively,and AUC was 0.939.The sensitivity,specificity and AUC of decision tree model were 90.87%,80.90% and 0.927 respectively.There was no significant difference in AUC between the two models（Z =1.204,P =0.2285）.Conclusion:The risk prediction model of sleep disorder in CCU elderly patients based on Logistic regression analysis and classification decision tree has good predictive efficacy,and the two models have the same predictive efficacy.Pain score,noise score,acute physiological and chronic health status score,gender,simple psychological status score and night invasive operation were independent risk factors for sleep disturbance in elderly CCU patients.