Construction and validation of social isolation prediction model for patients with cervical cancer during chemotherapy

Lu Yan; Lu Yun; Qin Yunxia; He Chenyun; Qiu Xiaoli

doi:10.3969/j.issn.1674-3245.2025.01.031

Abstract：Objective To analyze the risk factors of social isolation in patients with cervical cancer during chemotherapy,and to construct and verify the risk prediction model.Method A total of 154 patients with cervical cancer during chemotherapy who were treated in our hospital from April 2019 to February 2023 were retrospectively selected as the training set.According to the score of social network scale(LSNS-6),they were divided into social isolation group(<12 points)and non-social isolation group(≥12 points).Univariate analysis and multivariate logistic regression were used to analyze the related factors of social isolation in patients with cervical cancer during chemotherapy,and a risk prediction model was constructed.Hosmer-Lemeshow(H-L)was used to verify the goodness of fit of the model and the area under the receiver operating characteristic(ROC)curve(AUC)was used to test the prediction effect of the model.A total of 66 patients with cervical cancer who were treated in our hospital from March 2023 to April 2024 were selected as the validation set.Results There were 58 cases of social isolation in 154 patients with cervical cancer during chemotherapy,with an incidence of 37.66%.Multivariate logistic regression analysis showed that age<50 years old(β=1.022,OR=2.780,95%CI=1.061～7.282),low education level(β=1.247,OR=2.851,95%CI=1.850～8.983),family per capita monthly income(β=1.636,OR=5.135,95%CI=1.905～13.842),combined anxiety and depression(β=1.376,OR=3.959,95%CI=1.477～10.610),low social support score(β=-0.159,OR=0.853,95%CI=0.790～0.920),and low self-esteem score(β=-0.106,OR=0.900,95%CI=0.812～0.997)were risk factors for social isolation in patients with cervical cancer during chemotherapy.The H-L test results of the training set showed that χ2=4.559,P=0.803,the AUC of the ROC curve was 0.903(95%CI=0.855～0.951,P<0.05),the sensitivity was 0.845,and the specificity was 0.844.The H-L test results of the validation set showed that χ2=5.607,P=0.691,the AUC of the prediction model was 0.887(95%CI=0.835～0.938,P<0.05),the sensitivity was 0.810,and the specificity was 0.833.Conclusion The risk prediction model based on the six risk factors in the results has a good predictive effect,which can effectively provide reference for clinical intervention and reduce the occurrence of social isolation in patients with cervical cancer during chemotherapy.