Author(s)
Chuang L1, Cohen AT2, Agnelli G3, Gumbs PD4, Bauersachs R5, Kroep S1, Mismetti P6, Gitt AK7, Monreal M8, Willich SN9, van Hout B10
1Pharmerit International, Rotterdam, The Netherlands, 2Guy's and St Thomas' NHS Foundation Trust, London, UK, 3University of Perugia, Italy, Perugia, Italy, 4Daiichi-Sankyo Europe GmbH, Munich, Germany, 5Department of Vascular Medicine, Klinikum Darmstadt, Darmstadt, Germany, 6Saint Etienne University Hospital, Saint Etienne, France, 7Herzzentrum Ludwigshafen, Ludwigshafen, Germany, 8Hospital Universitari Germans Trias I Pujol, Barcelona, Spain, 9Charité - Universitätsmedizin Berlin, Berlin, Germany, 10University of Sheffield, Sheffield, UK
OBJECTIVES: EQ-5D-5L has been developed from its parental measure, EQ-5D-3L, moving from 3 to 5 levels, aiming to improve sensitivity, discriminative power and limit ceiling effects. The use of 5L has grown rapidly in recent years. However, the use of EQ-5D-5L in pulmonary embolism (PE) and deep vein thrombosis (DVT) populations have not been validated. Furthermore, there is a lack of performance comparisons for EQ-5D-5L against disease-specific quality of life measures. The aim of this study was therefore to compare the measurement properties of EQ-5D-5L against PEmb-QoL and VEINS-QOL/Sym (disease-specific), and PACT-Q2 (treatment-specific) in both PE and DVT populations. METHODS: PREFER in VTE,a non-interventional disease registry, was conducted between 2013 and 2014 in primary and secondary care across seven European countries. Consecutive patients with acute PE/DVT were documented and followed up over 12 months. Patients who completed all above-mentioned questionnaires at baseline were included in the study sample. The psychometric properties examined included acceptability (missing, ceiling and floor effects), validity (convergent, discriminant, and known-groups validity), and responsiveness. Known group validity and responsiveness were assessed using both effect size (Cohen’s d) and relative efficiency (F-statistic). RESULTS: A total of 1,054 and 1,537 complete cases for PE and DVT were identified. A ceiling effect in EQ-5D-5L was observed (14% and 10% in PE and DVT, respectively), while a higher proportion of missings was observed in PACT-Q2 (19% and 18%). EQ-5D-5L was low to moderately correlated with other measures (r<0.5). Low correlations (r<0.2) between each QOL measure against age and gender were observed, supporting the discrimination validity. EQ-5D-5L was associated with larger effect size/relative efficiency in most of known group comparisons in both groups. Similar results were observed for responsiveness. CONCLUSIONS: The results suggest that EQ-5D-5L is similar to PEmb-QoL, VENS-QOL/Sym and PACT-Q2 in terms of acceptability, validity and responsiveness in both PE and DVT populations.
Conference/Value in Health Info
2017-11, ISPOR Europe 2017, Glasgow, Scotland
Value in Health, Vol. 20, No. 9 (October 2017)
Code
PCV124
Topic
Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes, Stated Preference & Patient Satisfaction
Disease
Cardiovascular Disorders
