Abstract
PurposeWe aim to compare the psychometric properties of the EQ-5D-5L questionnaire with the EQ-5D-3L version and EQ VAS, based on a survey conducted in a sample representing the general adult population of Poland.MethodsThe survey comprised health-related quality of life (HRQoL) questionnaires: EQ-5D-5L, EQ VAS, SF-12 and EQ-5D-3L, together with demographic and socio-economic characteristics items. The EQ-5D index values were estimated based on a directly measured value set for Poland. The following psychometric properties were analysed: feasibility, distribution of responses, redistribution from EQ-5D-3L to EQ-5D-5L, inconsistencies, ceiling effects, informativity power and construct validity. We proposed a novel approach to the construct validity assessment, based on the use of a machine learning technique known as the random forest algorithm.ResultsFrom March to June 2014, 3978 subjects (aged 18–87, 53.2% female) were surveyed. The EQ-5D-5L questionnaire had a lower ceiling effect compared to EQ-5D-3L (38.0% vs 46.6%). Redistribution from EQ-5D-3L to EQ-5D-5L was similar for each dimension, and the mean inconsistency did not exceed 5%. The results of known-groups validation confirmed the hypothesis concerning the relationship between the EQ-5D index values and age, sex and occurrence of diabetes.ConclusionsThe EQ-5D-5L, in comparison with its EQ-5D-3L equivalent, showed similar or better psychometric properties within the general population of a country. We assessed the construct validity of the questionnaire with a novel approach that was based on a machine learning technique known as the random forest algorithm.
Highlights
Assessment of the health-related quality of life (HRQoL) is an essential step in the process of inferring the cost-effectiveness of medical technology
According to our best knowledge, this is the first study reporting the validity of the EQ-5D-5L in comparison to EQ5D-3L, measured on the same sample of the general population coming from a European country
As a part of the construct validity assessment, we proposed a novel approach based on the theoretical model for HRQoL assessment in the Polish population, designed using the random forest algorithm for regression
Summary
Assessment of the health-related quality of life (HRQoL) is an essential step in the process of inferring the cost-effectiveness of medical technology. To ensure the correctness of pharmacoeconomic inference, the HRQoL estimates should tend to the true value. This is difficult because the HRQoL is a latent variable and cannot be measured and interpreted directly. The assessment of psychometric properties, i.e. validation, includes determining if there are grounds to believe that the questionnaire used to assess HRQoL. The most commonly assessed and pointed in the COSMIN checklist (COnsensus-based Standards for the selection of health status Measurement INstruments) psychometric properties comprise validity, reliability and sensitivity [2]. Validity determines whether the questionnaire measures what is intended to measure and that it is useful for its intended purpose. Reliability assesses whether the results of the questionnaire specifying the same feature will be repeatable for subsequent measurements. Sensitivity is the ability to measure differences in the HRQoL among patients or patient groups [1, 3]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
