基于LASSO-Logistic回归的冠心病患者Ⅱ期心脏康复依从性预测模型的构建与验证OA

Objective To analyze the risk factors associated with low compliance in stage Ⅱ rehabilitation of patients with coronary heart disease using LASSO-Logistic regression,and to establish a nomogram prediction model.Methods A total of 200 patients with coronary heart disease who underwent percutaneous coronary intervention at the Cardiovascular Center,Xiang'an Hospital Affiliated to Xiamen University from June 2019 to June 2024 were enrolled.Patients entered stage Ⅱ cardiac rehabilitation within 1-2 weeks after surgery.According to their willingness to comply with stage Ⅱ rehabilitation,they were divided into a low-compliance group(n=42)and a high-compliance group(n=158).General clinical data were compared between the two groups.Multivariate logistic regression analysis was used to identify influencing factors of stage Ⅱ cardiac rehabilitation compliance in patients with coronary heart disease,and a nomogram prediction model was established.The SHAP(SHapley Additive exPlanations)framework was applied to provide local and global interpretations for model outputs.The discrimination of the model was evaluated using the receiver operating characteristic(ROC)curve,and model stability was verified using the bootstrap method with 1000 resamplings.Radar charts were used to visually compare sensitivity,specificity,accuracy,positive predictive value,negative predictive value,and other indicators.Results The low-compliance group had significantly higher proportions of junior college education or below,rural residence,one-way travel time to hospital>30 min,psychological distress(depression),lower self-efficacy,lower family support,lower social support,no medical insurance reimbursement,lower disease-related knowledge,and higher total TSK-SV Heart(tampa scale of kinesiophobia Swedish version for the heart)score than the high-compliance group(P<0.05).Multivariate logistic regression analysis showed that total TSK-SV Heart score(OR=1.579,95%CI:1.297-1.921),education level(OR=3.880,95%CI:1.276-11.794),residence(OR=7.420,95%CI:2.160-25.480),family support(OR=7.329,95%CI:2.021-26.579),social support(OR=6.293,95%CI:1.850-21.409),and one-way travel time to hospital(OR=3.178,95%CI:1.018-9.919)were independent risk factors for low compliance in stage Ⅱ cardiac rehabilitation in patients with coronary heart disease(P<0.05).The nomogram prediction model constructed based on these factors showed favorable predictive performance.ROC curve analysis revealed that the area under the curve(AUC)was 0.965 in the training set and 0.932 in the validation set.Conclusions A prediction model for low compliance with stage Ⅱ cardiac rehabilitation in patients with coronary heart disease was established using LASSO-Logistic regression,and the model showed high discriminative efficacy.Higher total TSK-SV Heart score,lower educational level,rural residence,insufficient family and social support,and one-way medical travel time>30 min were independent risk factors for low compliance.Clinically,attention should be paid to patients'psychological and social support status,and health education and rehabilitation guidance should be strengthened to improve compliance with stage Ⅱ rehabilitation and prognosis.