Comparison of the PROMIS Preference Score (PROPr) and EQ-5D-5L Index Value in General Population Samples in the United Kingdom, France, and Germany.

Autor: Klapproth CP; Department of Psychosomatic Medicine, Center for Internal Medicine and Dermatology, Charité - Universitätsmedizin Berlin, Berlin, Germany. Electronic address: christoph-paul.klapproth@charite.de., Sidey-Gibbons CJ; Department of Symptom Research, MD Anderson Cancer Center, University of Houston, TX, USA., Valderas JM; Health Services & Policy Research Group, University of Exeter, Exeter, England, UK; NIHR Peninsula Collaboration for Leadership in Applied Health Research and Care, Exeter, England, UK., Rose M; Department of Psychosomatic Medicine, Center for Internal Medicine and Dermatology, Charité - Universitätsmedizin Berlin, Berlin, Germany; Department of Quantitative Health Sciences, University of Massachusetts Medical School, Worcester, MA, USA., Fischer F; Department of Psychosomatic Medicine, Center for Internal Medicine and Dermatology, Charité - Universitätsmedizin Berlin, Berlin, Germany.
Jazyk: angličtina
Zdroj: Value in health : the journal of the International Society for Pharmacoeconomics and Outcomes Research [Value Health] 2022 May; Vol. 25 (5), pp. 824-834. Date of Electronic Publication: 2021 Dec 15.
DOI: 10.1016/j.jval.2021.10.012
Abstrakt: Objectives: The Patient-Reported Outcome Measurement Information System (PROMIS) Preference score (PROPr) can be used to assess health state utility (HSU) and estimate quality-adjusted life-years in cost-effectiveness analyses. It is based on item response theory and promises to overcome limitations of existing HSU scores such as ceiling effects. The PROPr contains 7 PROMIS domains: cognitive abilities, depression, fatigue, pain, physical function, sleep disturbance, and ability to participate in social roles and activities. We aimed to compare the PROPr with the 5-level EQ-5D (EQ-5D-5L) in terms of psychometric properties using data from 3 countries.
Methods: We collected PROMIS-29 profile and EQ-5D-5L data from 3 general population samples (United Kingdom = 1509, France = 1501, Germany = 1502). Given that cognition is not assessed by the PROMIS-29, it was predicted by the recommended linear regression model. We compared the convergent validity, known-groups construct validity, and ceiling and floor effects of the PROPr and EQ-5D-5L.
Results: The mean PROPr (0.48, 0.53, 0.48; P<.01) and EQ-5D-5L scores (0.82, 0.85, 0.83; P<.01) showed significant differences of similar magnitudes (d = 0.34; d = 0.32; d = 0.35; P<.01) across all samples. The differences were invariant to sex, income, occupation, education, and most conditions but not for age. The Pearson correlation coefficients between both scores were r = 0.74, r = 0.69, and r = 0.72. PROPr's ceiling and floor effects both were minor to moderate. The EQ-5D-5L's ceiling (floor) effects were major (negligible).
Conclusions: Both the EQ-5D-5L and the PROPr assessed by the PROMIS-29 show high validity. The PROPr yields considerably lower HSU values than the EQ-5D-5L. Consequences for quality-adjusted life-year measurements should be investigated in future research.
(Copyright © 2021. Published by Elsevier Inc.)
Databáze: MEDLINE
Externí odkaz: