Abstract
BACKGROUND: There is paucity of measures to conduct epidemiological studies related to disasters in Sri Lanka. This study validates a Sinhalese translation of the Impact of Event Scale- 8 items version (IES-8) for use in Sri Lanka. METHODS: This cross-sectional validation study was conducted in the densely populated rural area of Tangalle in the Southern province of Sri Lanka. The English version of the IES-8 after translation procedures in to Sinhalese was administered by trained raters to a community sample of 30 survivors of tsunami aged 13 years and above. Diagnostic accuracy, reproducibility and validity of the translated IES was assessed in terms of sensitivity, specificity, predictive values, likelihood ratios, diagnostic odds ratio, inter-rater reliability, internal consistency, criterion validity and construct validity. RESULTS: The cut-off score of 15 gave a fair sensitivity (77%) for screening along with other components of diagnostic accuracy. The inter-rater reliability was high (0.89). The internal consistency for the whole scale was high (0.78) with a high face and content validity. The criterion validity was high (0.83) and the construct validity demonstrated the two factor structure documented in the literature. CONCLUSION: This study demonstrates that this Sinhalese version of the Impact of Event Scale has sound diagnostic accuracy as well as psychometric properties and makes it an ideal measure for epidemiological studies related to natural and man made disasters in Sri Lanka.