Testing the psychometric properties of the Thai version of The Minnesota living with heart failure questionnaire

Abstract

The purpose of this study was to test the psychometric properties of the Thai version of the Minnesota Living with Heart Failure Questionnaire (MLHFQ). The subjects of the psychometric properties testing were 180 outpatients with chronic heart failure from cardiology and general medicine clinic at Phramongkutklao hospital. The Thai version of the MLHFQ and the Short Form-36 Health Survey (SF-36) version 1 were administered at baseline and next follow-up visit. Practicality was assessed with average time of administration and the floor and the ceiling effects. Reliability was evaluated using Cronbach’s alpha coefficients for internal consistency and intraclass correlation coefficients (ICCs) for test retest reliability assessment with telephone interview. Construct validity was supported with convergent and discriminant validity using correlation among the MLHFQ and the SF-36 scores, exploratory factor analysis using principal axis factoring and oblique rotation with direct oblimin, and known-groups validity that referred to ability to discriminate subjects among different New York Heart Association (NYHA) classes. Responsiveness was assessed with effect size. It was found that average time of administration approximately was five minutes and there were high floor (11.1% to 27.2%) and no ceiling effect on the MLHFQ scores. Cronbach’s alpha coefficients and ICCs of the MLHFQ were 0.86 to 0.93 and 0.61 to 0.77, respectively. There were significantly moderate or high correlations among the MLHFQ scores and the assumed corresponding SF-36 subscales and component summary scores (Spearman rank order correlation; rho = -0.49 to -0.56, p < 0.05). Twenty one items of the Thai version of the MLHFQ loaded on four factors such as physical, emotional, treatment, and symptoms dimensions. In addition, the MLHFQ scores discriminated among NYHA class I, II, and III (Kruskal-Wallis test, p < 0.001). Moreover, most observed effect sizes on the MLHFQ were moderate. These results suggest that the MLHFQ (Thai version) indicates acceptable preliminary psychometric properties and agrees with the original version