Research On Structural Equation Modeling And Subtype Study Of Attributional Effect On Physical Activity Among Patients With Coronary Heart Disease

Background:Although regular physical activity(PA)can significantly reduce the mortality of patients with coronary heart disease(CHD),only 20%to 50%of patients maintain regular PA.Different attribution style will induce different effects on future behavior.Dividing patients into different subtype for individualized intervention should help improve the effectiveness of the intervention.However,subtype classification based solely on attribution style and behavioral outcome may not be able to provide strong evidence for precise intervention.Therefore,it is necessary to clarify the pathway and key predictors for the attribution style of cardiac rehabilitation exercise in patients with CHD which affects their later behavior.This allows classification of the different attributional effects on behavioral outcomes to provide more precise interventions for patients with CHD.Objectives:The purpose of this study was to explore the main pathway of attribution to PA and key predictors affecting PA by constructing Structural Equation Modeling(SEM)of attributional effect on PA and to explore the subtype characteristics of PA by conducting a subtype study among patients with CHD based on the key predictors related to PA in the model,therefore providing a theoretical basis for the PA intervention on patients with CHD.Methods:This study included four parts:First part:Through the integration of Attribution Theory and Self-Efficacy Theory,a primary attributional effect model for PA in patients with CHD was established.Second part:The psychometric evaluation,the revised Causal Dimension Scale-Ⅱ(CDS-Ⅱ)and the Multidimensional Exercise Self-Efficacy Scale(MSES)were culturally adapted.The Individual-Oriented Achievement Motivation Scale-Short Form(IOAMS-SF)and Exercise Success Expectation Questionnaire(ESEQ)were revised based upon the goals of this study.The validity and reliability of the four instruments were tested.Third part:Emperical study,a cross-sectional study conducted from July to September 2020,in 703 patients with CHD from three tertiary general hospitals in Tangshan,who were discharged from hospital for more than 6 months and had no exercise contradictions.The patients were studied using the self-designed questionnaire and exercise diary to collect data on their PA.The Positive and Negative Affective Scale(PANS),ESEQ,the Chinese version of CDS-Ⅱ and the Chinese version of MSES and IOAMS-SF were used to collect data from the 703 patients.Using these data,the Structural Equation Modeling(SEM)of attributional effect on PA was constructed and evaluated.Forth part:This was a descriptive study.Based on a model of attributional effect on PA among patients with CHD,subdivided variables of the subtype study were determined.The subtype study was conducted on the 703 patients with CHD using K-means cluster analysis and discriminant analysis.Characteristics of each subgroup were explored,based on the patient demographic and clinical characteristics.The intervention strategies for each subgroup were put forward according to its classification and characteristics.Results:Part one:The conceptual framework of the attributional effect model of PA in patients with CHD was constructed:the attribution style of PA in patients with CHD indirectly affects PA through Exercise Success Expectation,Emotional Response,Exercise Self-Efficacy and Exercise Achievement Motivation.Part two:the I-CVI index of CDS-Ⅱ,ESEQ,MSES and IOAMS-SF were 0.88～1.00,S-CVI/Ave were 0.98,1.00,0.92 and 0.98 respectively.The factor analysis of CDS-Ⅱ and MSES extracted four and three factors respectively,same with the original scale.The Cronbach’s α values of CDS-Ⅱ,ESEQ,MSES and IOAMS-SF were 0.854,0.969,0.918 and 0.938 respectively.Part three:The attributional effect model fit well(χ2/df=3.922,GFI=0.964,AGFI=0.929,NFI=0.966,CFI=0.974,IFI=0.974,TLI=0.957,RMSEA=0.065).Part four:The patients with CHD were divided into four subtypes according to the key predictors in the model of attributional effect on PA:the type of most active exercise(22.8%),the type of more active exercise(35.6%),the type of helplessness no exercise(39.6%)and the type of regulated exercise(16.8%).There were significant differences in age,education,job participation,income level and duration of CHD among the four subtypes(P<0.05).Conclusions:The model of attributional effect on PA among patients with CHD reveals that Emotional Response,Exercise Success Expectation,Exercise Achievement Motivation and Exercise Self-Efficacy are the key mediate variables between attribution and PA.The subtype study suggests that the patients with CHD can be divided into four subtypes,and helps identify the different needs of the subtypes.The subtype study provides a theoretical basis for the intervention in clinical practice.