Reliability and validity of the Japanese version of the Ocular Surface Disease Index for dry eye disease

Abstract

Objectives: The Ocular Surface Disease Index (OSDI) questionnaire is widely used to evaluate subjective symptoms of dry eye disease (DED) as a primary diagnostic criterion. This study aimed to develop a Japanese version of the OSDI (J-OSDI) and assess its reliability and validity.

Design and setting: Hospital-based cross-sectional observational study.

Participants: A total of 209 patients recruited from the Department of Ophthalmology at Juntendo University Hospital.

Methods: We translated and culturally adapted the OSDI into Japanese. The J-OSDI was then assessed for internal consistency, reliability and validity. We also evaluated the optimal cut-off value to suspect DED using an area under the receiver operating characteristic curve (AUC) analysis.

Primary outcome measures: Internal consistency, test-retest reliability and discriminant validity of the J-OSDI as well as the optimal cut-off value to suspect DED.

Results: Of the participants, 152 had DED and 57 did not. The J-OSDI total score showed good internal consistency (Cronbach's alpha=0.884), test-retest reliability (interclass correlation coefficient=0.910) and discriminant validity by known-group comparisons (non-DED, 19.4±16.0; DED, 37.7±22.2; p<0.001). Factor validity was used to confirm three subscales within the J-OSDI according to the original version of the questionnaire. Concurrent validity was assessed by Pearson correlation analysis, and the J-OSDI total score showed a strong positive correlation with the Dry Eye-Related Quality-of-Life Score (γ=0.829). The optimal cut-off value of the J-OSDI total score was 36.3 (AUC=0.744).

Conclusions: The J-OSDI was developed and validated in terms of reliability and validity as an effective tool for DED assessment and monitoring in the Japanese population.

Keywords: OSDI; dry eye disease; ocular surface disease index; reliability; validity.

Figure 1

Three subscales of the J-OSDI as determined by factor analysis. The existence of three clusters that were used as subscales are shown. These were in accordance with the subscales that are used in the original version of the OSDI: vision-related function (components 1–5), ocular symptoms (components 6–9) and environmental triggers (components 10–12).

Figure 2

Clinical utility of the J-OSDI for evaluating DED. (A) The proportion of patients in each DED severity category as determined by the J-OSDI total score. (B) The proportion of patients who were clinically diagnosed with DED by category of severity according to the J-OSDI total score. (C) The receiver operator characteristic (ROC) curve for the diagnosis of DED determined by the Asia Dry Eye Society 2016 criteria using the J-OSDI. The area under the ROC curve (AUC) is 0.744.