Measuring Digital Health Literacy in Marginalised Populations: A Randomised Comparison of Layered Comprehension Support and Standard Survey Delivery, Embedded in the Validation of the Burmese eHealth Literacy Questionnaire (eHLQ)
Year: 2026
Type: study-registration
Abstract: Validated questionnaires are often given to people who read with difficulty or have little experience of the ideas the items ask about. When a respondent does not understand an item, data collectors usually explain it in their own words, which is neither standardised nor recorded, or studies exclude people who cannot read. Both weaken the validity and fairness of the scores. CLEAR (Comprehension-Layered, Equitable, Auditable, Reproducible) is a framework and open-source tool for standardised help on request. Items are shown and read aloud exactly as validated. Respondents who need help can open pre-approved clarifications in layers: a definition, then an example, with pictures and audio. A typed question is matched by keyword to the approved clarifications; no generative AI is used. Every help event is logged. This study tests CLEAR within the validation of the Burmese eHealth Literacy Questionnaire (eHLQ; 35 items, seven scales) among adults attending primary healthcare services at Mae Tao Clinic (MTC), Mae Sot, Thailand. Participants are randomised 1:1 to help on or help off. Both arms receive the same items, audio, picture-story introduction and practice items on the same device; only the help layer differs. The primary outcome is comprehension accuracy: blinded coders score participants’ own explanations of two randomly chosen items. We also test whether help benefits weaker readers more (differential boost), whether the eHLQ measures the same construct in both arms (measurement invariance), whether help reduces satisficing, and whether logged help requests identify the items that caused problems in cognitive interviews. The comparison is embedded in the approved validation study (CEAB-2026-028), following the logic of a Study Within A Trial. This registration also fixes, before any data are collected, how the validation analyses will use the two arms.
Source: Open Science Framework
Authors: Hein Thu, Aung Thura Htoo, Moe Oo Thant
Language: English
Subfield: General Health Professions
Field: Health Professions
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7214907394