Disease-specific health-related quality of life questionnaires for heart failure: a systematic review with meta-analyses
Authors:
Garin, O., Ferrer, M., Pont, A., Rue, M., Kotzeva, A., Wiklund, I., Van Ganse, E., and Alonso, J.
Abstract:
BACKGROUND: Heart failure (HF) is an increasingly common condition affecting patients' health-related quality of life (HRQL). However, there is little literature comparing HF-specific instruments. Our aim was to evaluate and compare data on the conceptual model and metric properties (reliability, validity and responsiveness) of HF-specific HRQL instruments, by performing a systematic review with meta-analyses.
METHODS AND RESULTS: Of 2,541 articles initially identified, 421 were full-text reviewed. Ninety-four reported data on five questionnaires: Minnesota Living with Heart Failure Questionnaire (MLHFQ), Chronic Heart Failure Questionnaire (CHFQ), Quality of Life Questionnaire for Severe Heart Failure (QLQ-SHF), Kansas City Cardiomyopathy Questionnaire (KCCQ) and Left Ventricular Dysfunction (LVD-36) questionnaire. Metric properties (reliability, validity and responsiveness) were summarised using meta-analysis for pools above five estimates. Cronbach's alpha coefficients were generally high (0.83-0.95) for overall scores and scales measuring physical health. Associations with four validity criteria (New York Heart Association [NYHA] class, six-minute walk test [6MWT] and short form-36 [SF-36] 'Physical' and 'Social Functioning') were moderate to strong (0.41-0.84), except for those between two CHFQ domains (fatigue and dyspnoea) and the NYHA (0.19 and 0.22). Pooled estimates of change from eight meta-analyses showed the MLHFQ to be highly responsive, with changes in overall score ranging from -9.6 (95% confidence interval [CI]: -4.1; -15.2) for placebo to -17.7 (95%
CI: -15.3; -20.2) for pacing devices. The CHFQ and KCCQ also showed good sensitivity to change.
CONCLUSIONS: Most of the questionnaires studied met minimum psychometric criteria, though current evidence would primarily support the use of the MLHFQ, followed by the KCCQ and CHFQ.
