| Objective:To investigate the current status and analyze the risk factors of delirium in patients with cervical spinal cord injury in ICU,and construct a prediction model of delirium in patients with cervical spinal cord injury in ICU.Verify the predictive efficiency of the model,in order to provide evidence for ICU medical staff to identify and intervene in high-risk patients with delirium in advance,thereby reducing the incidence of delirium in target patients.Methods:1.The relevant variables of delirium in ICU patients with cervical spinal cord injury included in the study were initially selected through a literature review,and then the above variables were discussed and modified through expert meetings,and finally the variables that may affect delirium of patients with cervical spinal cord injury in ICU were determined.2.Non-probabilistic sampling method was used to select patients with cervical spinal cord injury who were admitted to ICU of a tertiary hospital from November 2020 to November 2022 as the research objects.Data were collected by consulting the intensive care information system and paper medical records.The final included subjects were assigned to the modeling group and the validation group at a ratio of 7:3 according to the time sequence of ICU admission.The model was constructed with the patients in the modeling group.The risk factors of delirium in patients with cervical spinal cord injury in ICU were screened out by Logistic regression analysis based on factors with statistically significant differences in univariate analysis(P<0.05).The risk prediction model of delirium in patients with cervical spinal cord injury in ICU was constructed by combining the risk factors and the partial regression coefficient β value and intercept value of each factor.3.The area under the receiver operating characteristic curve(AUC)and Hosmer-Lemeshow goodness of fit test were used to evaluate the discrimination and calibration of the prediction model in the modeling group and the verification group.Results:1.Based on the literature review and expert meeting method,this study included variables that may affect delirium in patients with cervical spinal cord injury in ICU,including general data,disease-related data,evaluation data,ICU treatment data,and laboratory indicators.2.A total of 397 patients were finally included in this study,of which 84 patients developed delirium,with an incidence of 21.16%.Of the 397 patients,278 patients entered the modeling group and 119 patients entered the validation group.Univariate analysis showed that there were significant differences in age,gender,APACHE Ⅱ score,nerve injury level,shock,mechanical ventilation,sedative use,and constraint variables between the delirium and the non-delirium group(P<0.05).Logistic regression analysis found that independent risk factors for delirium in patients with cervical spinal cord injury in ICU(P<0.05)were age[OR=1.059,95% CI(1.026～1.093)],APACHE Ⅱ score [OR=1.063,95% CI(1.011～1.119)],nerve injury level [OR=2.323,95% CI(1.178～4.481)],shock[OR=4.808,95% CI(1.597～14.472)],sedative use [OR=3.210,95% CI(1.283～8.028)].The constructed prediction model for ICU patients with cervical spinal cord injury is,and is an exponential function.=-7.910+(0.057×Age)+(0.061×APACHE Ⅱ score)+(0.843×Nerve injury level assignment)+(1.570×Shock assignment)+(1.166×Sedative use assignment).3.The area under the receiver operating characteristic curve(AUC)of the model in the modeling group was 0.818(95% CI: 0.759～0.878,P < 0.001),and the Hosmer-Lemeshow goodness of fit test show that =8.440,P=0.392(P>0.05).When the optimal critical value of the prediction model was 0.223,the corresponding combination of sensitivity and specificity were the best,0.833 and0.734 respectively,the Youden index was 0.567.The results of the validation group showed that the area under the receiver operating characteristic curve(AUC)of the model in the validation group was 0.761(95% CI: 0.641～0.882,P<0.001),and the Hosmer-Lemeshow goodness of fit test showed that =9.988,P=0.266(P>0.05),the sensitivity and specificity of the model prediction were 0.625 and 0.926.The model divided the delirium risk levels of target patients into low risk(0～1 points),medium risk(2～3 points)and high risk(4～6 points).Conclusions:1.This study found that the incidence of delirium in patients with cervical spinal cord injury in ICU was 21.16%.Age,APACHE Ⅱ score,nerve injury level,shock,and sedative use were independent risk factors.On this basis,a risk prediction model for delirium in patients with cervical spinal cord injury in ICU was constructed.2.The risk prediction model constructed in this study has good prediction effect,which can effectively evaluate the occurrence and risk level of delirium in target patients.It is helpful for ICU staff to identify delirium high-risk patients early,take personalized intervention measures in time,and reduce the incidence of delirium in patients. |
PDF Full Text Request