Construction and validation of a risk screening model for cognitive frailty in elderly patients with comorbidities
Objective To investigate the status and risk factors of cognitive frailty in elderly patients with comorbidities,and to develop a risk screening model.Methods Totally 738 elderly patients with comorbidities were recruited in Shang-hai Fifth People's Hospital from June 1,to October 31,2021 were selected and divided into a modeling group of 590 pa-tients and a validation group of 148 patients.Data were collected by a general questionnaire and cognitive frailty assess-ment tools.Logistic regression was used to determine the influencing factors,and SPSS 19.0 software was used to estab-lish a nomogram model to predict the risk of cognitive frailty.Bootstrap method was used for internal verification of the model.C statistic and calibration curve were used to evaluate the prediction performance of the model.Results The inci-dence of cognitive frailty in modeling group and a validation group were 33.6％and 35.1％,respectively.Logistic regres-sion analysis showed that age(70-＜80 years:OR=2.861,95％CI was 1.567-5.224,P=0.001;≥80 years:OR=13.333,95％CI was 5.744-30.951,P＜0.001),activities of daily living(OR=6.301,95％CI was 3.344-11.871,P＜0.001),nutritional status(OR=0.370,95％CI was 0.179-0.763,P=0.007),depression(OR=2.689,95％CI was 1.414-5.114,P=0.003),sleep disorders(OR=3.218,95％CI was 1.833-5.647,P＜0.001),number of co-morbidity ≥4(OR=2.126,95％CI was 1.075-4.203,P=0.030),diabetes(OR=3.054,95％CI was 1.766-5.280,P＜0.001),and chronic heart failure(OR=16.657,95％CI was 5.821-42.112,P＜0.001)were independent risk fac-tors for cognitive frailty among patients with comorbidities.The model included these eight independent predictors.The area under the curve of receiver operating characteristic of the model was 0.928(95％CI was 0.905-0.951,P＜0.001),the Hosmer-Lemeshow test was 4.863,P=0.772,the best cutoff value was 0.731,the sensitivity was 0.894,and the specificity was 0.837;The C statistics of internal and external validation were 0.901 and 0.934,respectively;calibration curve showed a good fit,and Brier scores were 0.113 and 0.093.Conclusions The incidence of cognitive frailty is high a-mong elderly patients with comorbidities.Elderly patients with age ≥70,impaired ability of daily living,diabetes,chron-ic heart failure,≥4 comorbid conditions,malnutrition,sleep disorders,and depression are more susceptible to cognitive frailty.The risk prediction model has good discrimination and calibration,which can be used by clinical medical personnel to evaluate the risk of cognitive frailty among elderly patients with comorbidities,and provide reference for early screening and formulating nursing strategies.
agedcomorbiditiescognitive frailtyinfluencing factorsnomogramsprediction model
NETL
NSTL
万方数据
