Validity of Ocular Surface Disease Index for the Diagnosis of Dry Eye Syndrome at Kenyatta National Hospital

Abstract

Background: Dry eye syndrome (DES) is a collection of tear film disorders characterized by instability or decreased tear production, ocular pain and/or visual abnormalities and ocular surface inflammation. DES is a common disorder and can be a symptom of serious underlying illness. The diagnosis is made by subjective self evaluation, using standardized tools like Ocular Surface Disease Index (OSDI), or objective clinical evaluations e.g. Tear Break-Up Time (TBUT) which is the gold standard. The OSDI is a composite score that evaluates ocular irritation, its impact on vision and environmental triggers. It is widely used to assess DES symptoms and severity, with benefits including standardizing diagnosis of DES, enabling research and providing clinicians with a rapid tool to diagnose DES. While the OSDI has been widely adopted internationally, there remains a critical validation gap regarding its applicability and reliability in our local context at Kenyatta National Hospital (KNH). Objective: To evaluate the validity of the Ocular Surface Disease Index questionnaire in diagnosing DES. Methods: This was a cross-sectional study conducted at the KNH Outpatient Eye Clinic. Patients aged 18 years and older were randomly selected. Each participant completed the OSDI questionnaire, followed by TBUT assessment. SPSS v28 was used for data analysis. An ROC curve was utilized to evaluate the best cutoff values of sensitivity and specificity. The validity of the OSDI was analyzed using the McNemar test. In this study, criterion validity was assessed, defined as “the extent of the agreement between two measures taken at the same time”. Significance was set at p<0.05. Results: A total of 94 participants were recruited in the study. TBUT exhibited a strong positive correlation between the right and left eyes. Moreover, a negative correlation was observed between TBUT and OSDI scores. The diagnostic accuracy analysis demonstrated moderate sensitivity and specificity (0.712 and 0.857 respectively) of the OSDI test in detecting ocular surface disease. The overall predictive accuracy of the OSDI was 73.4%. The McNemar test yielded an asymptotic significance value of <0.001. Conclusion: The diagnostic accuracy of the OSDI test put forward its utility in identifying individuals with ocular surface disease. Future research can validate these findings in different populations.