Abstract
BACKGROUND Study participants and patients often perceive (long) questionnaires as burdensome. In addition, paper-based questionnaires are prone to errors such as (unintentionally) skipping questions or filling in a wrong type of answer. With the emergence of mobile questionnaire applications, such errors can be prevented. OBJECTIVE The objective of the present study was to validate an innovative way to measure quality of life using a mobile application (mobile app), based on the EQ-5D-5L questionnaire. This validation study compared the EQ-5D-5L questionnaire requested by a mobile app to the gold standard, a paper-based version of the EQ-5D-5L. METHODS The study was designed as a randomized, cross-over, and open study. Each participant received a digital version of the EQ-5D-5L questionnaire requested via a mobile app and the EQ-5D-5L paper-based questionnaire sent by postal mail. In the mobile app participants received, spread over five consecutive days, one question in the morning and one question in the afternoon and all questions were asked twice (at timepoint 1 (App T1) and timepoint 2 (App T2)). Primary outcomes were the correlations between the answers (scores) of each EQ-5D-5L question requested via the mobile app compared with the paper-based questionnaire to assess convergent validity. RESULTS In total 255 participants (healthy at own discretion), of which 117 men and 138 women in the age range of 18 – 64 years, completed the study. All questions showed a high correlation (0.64 - 0.92) for the paper-based EQ-5D-5L questionnaire versus the mobile app-based questions at App T1 and App T2. The scores and their variance remain similar over questionnaires, indicating no clear difference in answer tendency. Also, the correlation between the two app-based questionnaires was high (>0.73), illustrating a high test-retest reliability, and a reliable replacement for the paper-based questionnaire. CONCLUSIONS The present study indicates that the mobile app is a valid tool for measuring quality of life and is as reliable as the paper-based version of the EQ-5D-5L, while reducing response burden.

This publication has 38 references indexed in Scilit: