The Portuguese version of the Epilepsy Surgery Inventory (ESI-55): Cross-cultural adaptation and evaluation of psychometric properties

• Published in: Epilepsy & behavior Vol. 9; no. 1; pp. 126 - 132
• Main Authors: Alonso, Neide Barreira, Ciconelli, Rozana Mesquita, da Silva, Tatiana Indelicato, Westphal-Guitti, Ana Carolina, Azevedo, Auro Mauro, da Silva Noffs, Maria Helena, Sales Ferreira Caboclo, Luís Otávio, Sakamoto, Américo Ceiki, Targas Yacubian, Elza Márcia
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01-08-2006
• Subjects: Adult; Brazil; Cross-Cultural Comparison; Cultural Characteristics; Epilepsy Surgery Inventory (ESI-55); Epilepsy, Temporal Lobe - psychology; Epilepsy, Temporal Lobe - surgery; Female; Health Status Indicators; Health-related quality of life; Humans; Language; Male; Mesial temporal sclerosis; Portuguese version; Preoperative Care; Psychometric properties; Psychometrics - instrumentation; Quality of Life; Reproducibility of Results; Surgical treatment; Surveys and Questionnaires; Temporal lobe epilepsy; Brazil; Health-related quality of life; Surgical treatment; Temporal lobe epilepsy; Portuguese version; Mesial temporal sclerosis; Epilepsy Surgery Inventory (ESI-55); Psychometric properties
• ISSN: 1525-5050; 1525-5069
• DOI: 10.1016/j.yebeh.2006.04.009

The purpose of this study was to develop a Portuguese version of the Epilepsy Surgery Inventory (ESI-55) and to assess its psychometric properties. Sixty patients with temporal lobe epilepsy related to unilateral mesial temporal sclerosis who underwent presurgical evaluation at the Universidade Federal de São Paulo (UNIFESP) formed the sample for this study. The psychometric properties of the ESI-55 included: reliability, validity, and responsiveness. Internal consistency was high in all domains (Cronbach’s
α ranging from 0.76 for Social Function to 0.88 for Physical Function) except Overall Quality of Life (
α
=
0.45). Test–retest reliability after 1 week was good, with the intraclass correlation coefficient ranging from 0.79 (Energy/Fatigue) to 0.92 (Role Limitations due to Emotional Problems). Interrater reliability ranged from 0.84 (Cognitive Function) to 0.94 (Role Limitations due to Physical Problems). For construct validity, we verified a high correlation between the ESI-55 and Health Assessment Questionnaire-8 for the Physical Function domain (Pearson linear correlation
=
−0.84), and a moderate correlation for the Pain domain (
P
=
−0.58), but for the other subscales no correlation was detected. Beck Depression Inventory and ESI-55 domains were highly statistically correlated (ANOVA:
P
<
0.005), but there was no association of the Cognitive Function and Role Limitations due to Memory Problems subscales with neuropsychological evaluation (Pearson coefficient:
P
>
0.05). With respect to demographic characteristics, a statistically significant correlation was observed for the variable educational level (Student
t,
P
<
0.005) and ESI-55 scores. There was a high correlation between seizure frequency and ESI-55 domains for clinical variables (ANOVA,
P
<
0.005). Surgical treatment in this series improved health-related quality of life in the seizure-free group in three domains—Health Perception (1.24), Emotional Well-Being (1.32), and Energy/Fatigue (1.48)—as reflected by the standard response mean and the effect size of the sample. Our results support the psychometric properties of the Portuguese version of the ESI-55 as a measure of health-related quality of life.