Psychometric properties of the Czech versions of the Impact of Pediatric Epilepsy Scale (IPES) and Quality of Life in Epilepsy Inventory for Adolescents (QOLIE-AD-48)

Objective: In order to introduce a complex system of monitoring and evaluation of the heath-related quality of life (HRQoL) of children and adolescents with epilepsy (CWE) and their families in the Czech Republic, we aimed to validate the Czech versions of two appropriate instruments - the Impact of Pediatric Epilepsy Scale (IPES) and the Quality of Life in Epilepsy Inventory for Adolescents (QOLIE-AD-48).

Methods: The verification of the 11-item IPES was carried out in the group of parents of 248 CWE aged 2-18 years. One hundred and thirty-five CWE from the given group aged 11-18 years then completed the 48-item QOLIE-AD-48. Internal consistency, test-retest reliability (with a three-month interval) and the factorial structure of the Czech versions were determined and compared with the original instruments.

Results: We found that the Czech version of the IPES exhibited very good psychometric properties including high internal consistency (Cronbach's alpha, α, of 0.93), high test-retest reliability (intraclass correlation coefficient, ICC, of 0.76) and the same 3-factor structure as the original instrument. The superiority of this 3-factor solution over the alternate 2-factor model proposed for some language versions of the IPES was determined using confirmatory factor analysis. We found 8 items in the Czech version of the QOLIE-AD-48 belonging to original Attitudes towards epilepsy and Social support subscales that do not fit well with the Czech version due to their low correlation with the total score and insufficient test-retest reliability and should be omitted. For the remaining 40 items, we have determined high internal consistency (α = 0.95) and test-retest reliability (ICC = 0.82). Confirmatory factor analysis revealed that the 6-factor solution derived from the original instrument (without two removed subscales) was appropriate for the Czech version. The individual subscales exhibited high internal consistency with α = 0.61-0.91. The external validation of both instruments was confirmed based on a significant correlation between test results and physicians' reports of the characteristics of the child's epilepsy.

Conclusions: The Czech versions of both instruments studied are reliable and valid, and can be used in the next research focusing on the effect of different treatment approaches on the HRQoL of CWE and their families.