Reliability of Chinese web-based ocular surface disease index questionnaire in dry eye patients: a randomized, crossover study

Aim To assess the reliability of web-based version of ocular surface disease index in Chinese (C-OSDI) on clinically diagnosed dry eye disease (DE) patients. Methods A total of 254 Chinese participants (51% male, 129/254; mean age: 27.90±9.06y) with DED completed paper- and web-based versions of C-OSDI questionnaires in a randomized crossover design. Ophthalmology examination and DED diagnosis were performed prior to the participants being invited to join the study. Participants were randomly designated to either group A (paper-based first and web-based second) or group B (web-based first and paper-based second). Final data analysis included participants that had successfully completed both versions of the C-OSDI. Demographic characteristics, test-retest reliability, and agreement of individual items, subscales, and total score were evaluated with intraclass correlation coefficients (ICC), Spearman rank correlation, Wilcoxon test and Rasch analysis. Results Reliability indexes were adequate, Pearson correlation was greater than 0.8 and ICCs range was 0.827 to 0.982; total C-OSDI score was not statistically different between the two versions. The values of mean-squares fit statistics were very low compared to 1, indicating that the responses to the items by the model had a high degree of predictability. While comparing the favorability 72% (182/254) of the participants preferred web-based assessment. Conclusion Web-based C-OSDI is reliable in assessing DED and correlation with the paper-based version is significant in all subscales and overall total score. Web-based C-OSDI can be administered to assess individuals with DED as participants predominantly favored online assessment.