Walidacja sinhalskiej i tamilskiej wersji kwestionariusza do badania przesiewowego epilepsji na Sri Lance

BACKGROUND: Epilepsy imposes a significant public health burden in low- and middle-income countries, including Sri Lanka, where reliable epidemiological data are limited. Validated, culturally adapted screening tools are essential for accurate case identification in large-scale community surveys. METHODS: A 9-item screening questionnaire previously validated in several languages was used. The English version of the questionnaire was translated into Sinhala and Tamil, and both versions underwent forward-backward translation and cultural adaptation using a modified Delphi process. Judgmental validity (face, content, and consensual validity) was evaluated by expert panels. Criterion validity was assessed using independent clinical diagnoses by consultant neurologists, who were blinded to screening outcomes. Diagnostic accuracy metrics-sensitivity, specificity, predictive values, likelihood ratios, Cohen's kappa, and internal consistency-were estimated in people with epilepsy and healthy controls for the Sinhala and Tamil versions. RESULTS: The Sinhala validation included 143 participants (70 cases, 73 controls). Sensitivity was 100%, specificity 65.8%, PPV 73.7%, NPV 100%, accuracy 82.52%, LR⁺ 2.92, LR⁻ 0.00, kappa 0.65, and Cronbach's alpha 0.78. The Tamil validation included 165 participants (87 cases, 78 controls). Sensitivity was 92%, specificity 92.2%, PPV 93.1%, NPV 91.0%, accuracy 92.1%, LR⁺ 10.3, LR⁻ 0.08, kappa 0.84, and Cronbach's alpha 0.858. DISCUSSION: The Sinhala and Tamil versions demonstrated strong diagnostic performance and cultural appropriateness. The Sinhala version prioritized maximal sensitivity, while the Tamil version showed a balance between high sensitivity and specificity. These validated tools are suitable for national community-based studies of epilepsy prevalence and treatment gaps.