Comparison of the reliability and validity between the Chinese version of SANDE and OSDI dry eye questionnaires

Objective

To assess the applicability of the Chinese version of the symptom assessment in dry eye (SANDE) questionnaire refer to the ocular surface disease index (OSDI) questionnaire.

Methods

A cross-sectional study was conducted.Three hundred and twenty-three students from a senior high school in Hongkou District in Shanghai were enrolled in June 2020.The Chinese version of the SANDE and the OSDI questionnaires were answered by the students to assess the dry eye symptom and were collected on-site.Scores of the two questionnaires were calculated.According to the OSDI score, the students were divided into the normal control group (n=87), mild dry eye group (n=82), moderate dry eye group (n=87) and severe dry eye group (n=67). Cronbach α was obtained to evaluate the internal consistency.Kruskal-Wallis test was used to analyze the difference in SANDE scores among dry eye groups with different severities and evaluate the discriminative validity.Correlation between the total scores of the two questionnaires was analyzed by using Spearman rank correlation analysis to evaluate the criterion validity.Factor analysis was used to evaluate the construct validity.Receiver operating characteristic curve (ROC) was drawn to obtain the area under the ROC curve (AUC) and cut-off point to evaluate the diagnostic threshold for dry eye.This study protocol adhered to the Declaration of Helsinki and was approved by an Ethics Committee of Shanghai General Hospital (No.2020KY026). Written informed consent was obtained from guardians of each subject.

Results

The Cronbach α of the SANDE and OSDI questionnaires were 0.856 and 0.829, respectively.SANDE score of the four groups classified according to the OSDI score was 7.0 (2.5, 16.9) for the normal control group, 17.0 (8.4, 31.0) for the mild dry eye group, 29.2 (14.6, 49.4) for the moderate dry eye group and 49.1 (24.4, 60.7) for the severe dry eye group, respectively, which was increased in turn.There was a statistically significant overall difference in the SANDE score among the four groups (H=113.213, P<0.001), and statistically significant pairuise differences were found (all at P<0.05). The SANDE score was moderately positively associated with the OSDI score (r_s=0.615, P<0.001). The factor analysis revealed that for the SANDE questionnaire, factor loadings of its two items on the common factors were higher than 0.5, among which the frequency of dry eye symptoms was 0.936, and the severity of dry eye symptoms was 0.936.The AUC of the SANDE questionnaire was 0.815 (P<0.001, 95% CI: 0.770-0.860). When using 23 as the diagnostic threshold for dry eye, the sensitivity and specificity of the SANDE questionnaire were 60.59% and 88.51%, respectively.

Conclusions

The Chinese version of the SANDE questionnaire shows good reliability, validity, discrimination ability and slightly poor sensitivity in detecting dry eye.As a simple assessment questionnaire, it is applicable to the Chinese population and is of good clinical application value.

Dry eye syndromes; Symptom assessment; Surveys and questionnaires; Reliability; Validity