EVALUATING THE CONCEPTUAL EQUIVALENCE BETWEEN PAPER AND THREE ELECTRONIC DATA COLLECTION MODES OF THE EQ-5D-5L HEALTH STATUS INSTRUMENT
Author(s)
Crescioni M1, Flood E2, Lundy J3, Coons SJ1, Byrom B4, Eremenco S1
1Critical Path Institute, Tucson, AZ, USA, 2ICON Clinical Research, Gaithersburg, MD, USA, 3Outcometrix, Tucson, AZ, USA, 4ICON Clinical Research, Marlow, UK
OBJECTIVES: The Electronic Patient-Reported Outcome (ePRO) Consortium and the EuroQol Research Foundation commissioned a study to assess the measurement equivalence of data collected using the EQ-5D-5L across various data collection modes. One study component examined equivalence of paper, handheld, web and interactive voice response (IVR) modes of the EQ-5D-5L using qualitative methods, in accordance with the ISPOR Task Force Report on Measurement Equivalence. METHODS: A single-visit, cognitive interview and usability testing (CI/UT) study was conducted. After providing informed consent, participants completed the paper format and one of three electronic modes (handheld, IVR, or web) of the EQ-5D-5L and afterwards participated in an interview. Following a semi-structured interview guide, participants were asked about their interpretation of the content between the modes and issues completing the measure using the assigned mode. Interviews were audio-recorded, transcribed, and analyzed qualitatively. RESULTS: Thirty UK participants, aged 20-70, participated in the CI/UT study, with 10 participants completing each electronic mode (handheld, IVR, or web). Most participants (n=23; 77%) had a chronic health condition. Overall, participants in all mode groups interpreted the measure content consistently and appropriately for each item of the EQ-5D-5L. When asked whether layout differences would impact their answers between paper and electronic formats, most participants (n=23) indicated there would be no difference. Reported potential discrepancies were mostly related to the 0-100 visual analogue scale (VAS), where four participants noted answer discrepancies due to difficulty selecting a precise response on the handheld and web versions. CONCLUSIONS: Consistent interpretation of items in the EQ-5D-5L, by participants using paper, handheld, IVR, or web, supports conceptual equivalence of the items across modes. Minor differences in presentation did not appear to undermine conceptual understanding of these items; however, usability issues potentially affected measurement equivalence for the VAS item, highlighting the importance of usability testing to address such issues prior to implementation.
Conference/Value in Health Info
2017-11, ISPOR Europe 2017, Glasgow, Scotland
Value in Health, Vol. 20, No. 9 (October 2017)
Code
PHP159
Topic
Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes, Stated Preference & Patient Satisfaction
Disease
Multiple Diseases