Evaluation of the usefulness of EQ-5D as a patient-reported outcome measure using the Paretian classification of health change among patients with chronic heart failure

Purpose: The aim of this study was to evaluate the usefulness of EQ-5D as a patient-reported outcome measure using different analytical methods. Especially we used the Paretian Classification of Health Change, to see if this gave better information compared to measures that are more traditional. For the evaluation we used data from patients with chronic heart failure (HF).

Methods: We compared results of EQ-5D at baseline and at 1 year's follow up for HF patients with preserved or reduced ejection fraction (EF), HFpEF (EF > 50%, n = 930) and HFrEF (EF < 40%, n = 3831) using individual patient data from the Swedish Heart Failure Registry. Statistical analysis included EQ-5D index and proportions for all five dimensions of the EQ-5D. In addition, we also used the Paretian classification of Health Change to judge overall improvements (improved in at least one dimension and not worsened in any other dimension) or worsening (vice versa) in EQ-5D profiles.

Results: Mean EQ-5D index showed minor changes at the one-year follow-up, likewise in both groups. The proportions reporting moderate, or severe, problems increased for all five dimensions of the EQ-5D in the HFpEF group. In the HFrEF group this was seen only for three dimensions, with no change for "anxiety/depression" and reduction of problems for "usual activities ". The Paretian classification showed that 24% (n = 200) of the HFpEF group and 34% (n = 1059) of the HFrEF group reported overall improvement while 43% (n = 355) and 39% (n = 1212) respectively reported overall worsening. Multiple logistic regressions showed different patterns of determinants e.g. that treatment in a cardiology clinic only affected overall health outcome in the HFrEF group.

Conclusion: The usefulness of EQ-5D is dependent on the analytical method used. While the index showed minor differences between groups, analyses of specific dimensions showed different patterns of change in the two groups with better prognosis for the HFrEF group. The Paretian classification of Health Change could further identify subgroups that showed overall improvements or overall worsening. This method can therefore help to identify needs for more tailored interventions in health services.