Analysis of influencing factors on compliance of rehabilitation exercise in stroke patients and construction of nomogram prediction model
Objective To explore and analyze the influencing factors of rehabilitation exercise compliance in stroke patients and construct a nomogram prediction model.Methods A retrospective analysis was conducted on the clinical data of 233 stroke patients admitted to the rehabilitation department of our hospital from February 2021 to December 2023 for rehabilitation exercise.Based on rehabilitation exercise compliance,160 cases were divided into a good compliance group(compliance rate≥75%)and 73 cases into a poor compliance group(compliance rate<75%).The predictive value of related indices for poor compliance with rehabilitation exercise in stroke patients was analyzed.Factors affecting poor compliance with rehabilitation exercise in stroke patients were explored.A nomogram prediction model for rehabilitation exercise compliance in stroke patients was established and verified and evaluated using calibration and decision curves.Results Compared with the poor compliance group,the good compliance group had a significantly increased proportion of patients with higher literacy,higher disease cognition,and more rehabilitation nursing guidance.SF-HBMS scores and SSRS scores were significantly higher,while HAMD scores were significantly lower(P<0.05).Multivariate Logistic regression analysis identified low education level,low disease cognition level,lack of rehabilitation nursing guidance,low SF-HBMS scores,low SSRS scores,and high HAMD scores as independent risk factors for poor compliance with rehabilitation exercise(P<0.05).ROC curve analysis showed AUCs of 0.642,0.708,and 0.625 for SF-HBMS scores,SSRS scores,and HAMD scores,respectively,with optimal cutoff values of>61 points,>33 points,and≤16 points,respectively,all of which had significant value in predicting rehabilitation exercise compliance(P<0.05).A nomogram prediction model for poor compliance with rehabilitation exercise in stroke patients was established based on independent influencing factors.The C-index was 0.863(95%CI:0.810～0.915),and the chi-square value of the goodness of fit test was 3.904,P=0.866>0.05.There was no significant difference between the predicted and observed values,indicating a high model fit.When the prediction threshold was>0.18,the clinical net benefit of the nomogram risk prediction model was higher than that of the independent predictor.Conclusions The nomogram prediction model,based on factors affecting compliance with stroke rehabilitation exercise such as education level,disease cognition level,rehabilitation nursing guidance,SF-HBMS scores,SSRS scores,and HAMD scores,is highly accurate and has high predictive value.
StrokeCompliance with rehabilitation exerciseInfluencing factorsNomogram type
万方数据
维普
