| Research Purpose:To identify the risk factors of delirium in elderly patients after abdominal surgery,establish logistic regression model and verify the model.To provide theoretical basis for medical staff to take corresponding preventive measures in time,so as to reduce the incidence of postoperative delirium of patients undergoing abdominal surgery,ensure the smooth progress of patients'treatment,reduce the length of stay and cost in hospital,promote the harmonious relationship between doctors and patients,and improve the satisfaction of patients and their families.Research Methods:In this study,the data of elderly patients who received abdominal surgery in a top three hospital in Huzhou City from January 2019 to November 2019 were investigated retrospectively.Through VIP,CNKI,Wanfang,CBM and other chinese databases and PubMed,Cochrane Library,JBI evidence-based health care database,EMBASE,web of Science and other foreign databases were used to retrieve the literature and research progress on the risk factors of postoperative delirium in elderly patients undergoing abdominal surgery."Elderly patients","delirium","acute encephalopathy syndrome","abdominal surgery","postoperative delirium","dangerous factors"and"risk factors"are the key words in Chinese and"older adults","elderly patients","delirium","acute encephalopathy syndrome","abdominal operation","postoperative delirium","dangerous factors"and"risk factors"are the key words in English.Referring to a number of domestic and foreign clinical practice guidelines,systematic review,expert consensus results,consulting experts of general surgery and anesthesiology department to select risk factors that may be related to postoperative delirium of elderly patients undergoing abdominal surgery,the hospital electronic medical record information system and access to paper medical records of patients were used to collect relevant information.The data were collected by Excel and analyzed by spss21.0.The measurement data subject to normal distribution were represented by meanąstandard deviation,and the comparison between groups was performed by t test;the measurement data of non normal distribution were described by median and quartile,and the comparison between groups is performed by Wilcoxon rank sum test.Frequency and rate(%)were used for counting data,and chi square tests were used for inter group comparison.The collected patient medical record information was divided into modeling group and validation group according to the time sequence of collection.Using modeling queue data to build the model,the risk factors entering the model were selected by the advanced single factor analysis method,and then the partial regression coefficient?value of each factor in the prediction model was established by logistic regression analysis,and the regression equation was established.Hosmer lemeshow(H-L)chi square test was used to verify the compliance of the model.It was statistically significant that there was a significant difference between the prediction ability and the actual incidence of postoperative delirium in elderly patients undergoing abdominal surgery,and the degree of compliance is poor.Otherwise,the consistency between the two was good.The validation queue data was used to evaluate the prediction efficiency of the model,and the working characteristic curve of the subjects was selected the ability to distinguish the area under curve(AUC)response prediction model,that is,the ability to correctly distinguish postoperative delirium patients and non delirium patients,the closer AUC is to 1,the better the ability to distinguish the model.Research Results:In this study,we collected 360 effective patient information materials,among which 55patients actually had delirium.According to the single factor results,there are 12 risk factors related to postoperative delirium in elderly patients undergoing abdominal surgery,including age,APACHE-II score,high blood pressure,heart disease,admission status,operation level,sleep disorder,infection,mechanical ventilation,sedative and analgesic drugs,passive restraint and postoperative environment.The difference of 12 clinical variables was statistically significant(P<0.05).There was no significant difference between other remaining indexes and postoperative delirium in elderly patients(P>0.05).After binary logistic regression analysis,the interaction between single factors and the mixed factors were excluded.The results showed that:age[or=1.180,95%confidence interval(1.097-1.269)],heart disease[or=3.256,95%confidence interval(1.121-9.461)],admission status[or=3.415,95%confidence interval(1.198-9.740)],sleep disorder[or=3.333,95%confidence interval(1.172-9.478)]which the four risk factors were statistically significant(P<0.05)between the delirium group and the non delirium group of the elderly patients undergoing abdominal surgery,which indicated that the above four risk factors were independent risk factors of pod in the elderly patients undergoing abdominal surgery,and the admission status might be a marker independent risk factor of pod in the elderly patients undergoing abdominal surgery.No correlation was found between APACHE-?score,hypertension,surgical grade,infection,mechanical ventilation,sedative and analgesic drugs,passive restraint,postoperative environment and postoperative delirium in elderly patients undergoing abdominal surgery.The prediction model was p=e ~x/(1+e ~x)×100%,e was exponential function,x=-19.521+0.165×age+1.181×heart disease assignment+1.228×admission status assignment+1.204×sleep disorder assignment.The H-L chi square test result of the model is?~2=5.198,P=0.736(P>0.05),which shows that the prediction probability of the model for pod is significantly different from that of the actual observation,and there is no statistical significance.We think that the prediction model constructed in this study has a good degree of compliance.The area under ROC curve of the model is 0.841(0.8-0.9),95%confidence interval(0.736-0.946),P<0.001.When the risk threshold of the prediction model is 36.73%,the combination of sensitivity and specificity is the best,in which the sensitivity is 0.588,the specificity is 0.956,and the Jordan index is 0.544.The results show that the sensitivity,specificity and accuracy of the model are 52.9%,96.7%and 89.8%,respectively.The risk of delirium was classified as low risk(0-1),medium risk(2)and high risk(3-4).Research Conclusion:In clinical practice,there are many factors related to the occurrence of pod in elderly patients undergoing abdominal surgery.Through this study,we found that age,heart disease,admission status and sleep disorder were independent risk factors of pod,and admission status might be a marker independent risk factor of pod in elderly patients undergoing abdominal surgery.The probability of pod can be predicted by constructing logistic regression model.The prediction efficiency of the model is tested by Hosmer lemeshow(H-L)chi square test,prediction effect test,and ROC analysis.The model is easy to operate,which can help medical staff to identify patients with high risk of delirium,and provide objective theoretical basis for them to take corresponding prevention and intervention measures in time. |
PDF Full Text Request