Psychometric evaluation of the Serbian version of the Quality of Life in Epilepsy Inventory-31 (QOLIE-31)

Abstract

To evaluate the psychometric properties of the Serbian-language version of the Quality of Life in Epilepsy Inventory-31 (QOLIE-31). After undergoing a translation and cultural adaptation of its items in order to create a Serbian-language version of QOLIE-31, we assessed its psychometric properties-reliability, construct validity and criterion validity. The sample consisted of 203 adults with epilepsy. Reliability was tested both by assessing the internal consistency and by the test-retest method. Construct validity was assessed by factor analysis, multitrait-scaling analysis and method of known-groups validation. This was achieved by assessing the relationship between scales and external measures (socio-demographic characteristics, seizure severity and etiology of epilepsy). Criterion validity was assessed by correlation analysis between QOLIE-31 and Short form 36 health survey (SF-36) and Neurotoxicity scale-II. The domains showed high internal consistency (Cronbach's α 0.94). Test-retest reliability for Overall test score was 0.83 (Pearson's coefficient) indicating temporal stability. Seizure severity and etiology of epilepsy significantly influenced all QOLIE-31 domains except the Medication effect domain, with lowest scores in high seizure severity and symptomatic etiology groups. Employment status significantly influenced Overall quality of life, Emotional well-being, Social function and Overall score. Educational level was related to the Emotional well-being domain, with highest scores for students. The QOLIE-31 was highly positively correlated with SF-36 (rho=0.898) and strongly negatively correlated with Neurotoxicity scale-II (rho=-0.783). Serbian adaptation of the QOLIE-31 questionnaire is reliable and valid for assessing the quality of life in patients with epilepsy.