Determinants of Utility Based on the EuroQol Five-Dimensional Questionnaire in Patients with Chronic Heart Failure and Their Change Over Time: Results from the Swedish Heart Failure Registry
Jenny Berg, Peter Lindgren, Märit Mejhert, Magnus Edner, Ulf Dahlström, Thomas Kahan, Jenny Berg, Peter Lindgren, Märit Mejhert, Magnus Edner, Ulf Dahlström, Thomas Kahan
Abstract
Background: There is limited information on drivers of utilities in patients with chronic heart failure (CHF).
Objectives: To analyze determinants of utility in CHF and drivers of change over 1 year in a large sample from clinical practice.
Methods: We included 5334 patients from the Swedish Heart Failure Registry with EuroQol five-dimensional questionnaire information available following inpatient or outpatient care during 2008 to 2010; 3495 had 1-year follow-up data. Utilities based on Swedish and UK value sets were derived. We applied ordinary least squares (OLS) and two-part models for utility at inclusion and OLS regression for change over 1 year, all with robust standard errors. We assessed the predictive accuracy of both models using cross-validation.
Results: Patients' mean age was 73 years, 65% were men, 19% had a left ventricular ejection fraction of 50% or more, 23% had 40% to 49%, 27% had 30% to 39%, and 31% had less than 30%. For both models and value sets, utility at inclusion was affected by sex, age, New York Heart Association class, ejection fraction, hemoglobin, blood pressure, lung disease, diabetes, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, nitrates, antiplatelets, and diuretics. The OLS model performed slightly better than did the two-part model on a population level and for capturing utility ranges. Change in utility over 1 year was influenced by age, sex, and (measured at inclusion) disease duration, New York Heart Association class, blood pressure, ischemic heart disease, lung disease, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, and antiplatelets.
Conclusions: Utilities in CHF and their change over time are influenced by diverse demographic and clinical factors. Our findings can be used to target clinical interventions and for economic evaluations of new therapies.
Keywords: EQ-5D; chronic heart failure; utilities; value set.
Copyright © 2015. Published by Elsevier Inc.
Source: PubMed