Evidence on the global measurement model of the Mi

Authors: Olatz Garin Montse Ferrer Àngels Pont Ingela Wiklund Eric Van Ganse Gemma Vilagut Josué Almansa Aida Ribera Jordi Alonso

Publish Date: 2013/05/16

Volume: 22, Issue: 10, Pages: 2675-2684

PDF Link

Abstract

The Minnesota Living with Heart Failure Questionnaire MLHFQ is the most widely used healthrelated quality of life measure in both clinical and research settings Nevertheless its measurement model has never been confirmed This study aims to fill that gap with a large international sampleData from eight studies 3847 patients with heart failure from 21 countries were merged and analysed Common variables included MLHFQ scores functional capacity cardiovascular risk factors and the sociodemographic characteristics of the patient The measurement model of the MLHFQ was assessed by means of exploratory and confirmatory factor analyses EFACFA The reliability of MLHFQ scores was evaluated using Cronbach’s alpha coefficient and the MLHFQ’s ability to differentiate among known groups was assessed through severity levelsFindings from the EFA and CFA suggest that the MLHFQ total and domainspecific scores fall within a bifactor model The physical and emotional scores were supported within the sample as was the original total score Furthermore a third factor was revealed regarding social environment The reliability coefficient reached 09 for almost all physical and total scores All the MLHFQ mean scores showed the ability to differentiate among functional capacity groups with most of the effect size coefficients reaching 08Beyond the suitable degree of reliability and validity displayed by the MLHFQ scores in the different countryspecific versions our results confirmed for the first time the unidimensionality of the most commonly used score in HF patients the total MLHFQ score Moreover the social environment domain identified in this study can now be considered when assessing these patients’ HRQL especially as a relevant outcome with regard to disease managementThe authors are very grateful to the members of the IQOD Project managed by the MAPI Research Institute and to all the institutions and patients that have made it possible Furthermore we would like to thank JM Valderas for his special support in the initial preparation of this article and Carlos GarciaForero for his contribution to the interpretation of our results This study was supported by European Commission contract QLRICT200000551 IQOD CIRIT 2001SGR 00405 SAF200211009E and the CIBER de Epidemiología y Salud Pública CIBERESP

Keywords: