Psychometric validation of the French version of the adverse childhood experiences international questionnaire (ACE-IQ)

Abstract

ObjectiveSince Dr. Vincent Felitti conducted his seminal work on adverse childhood experiences (ACEs) in the 1980′s, a large body of literature has emerged concerning the relation between early life adversity and physical and psychological health outcomes in adulthood. The objective of this study was to assess the psychometric properties of the French version of the Adverse Childhood Experiences International Questionnaire (ACE-IQ). A validated translation of the ACE-IQ will facilitate comparing data in French-speaking countries where this tool is used to measure a dose–response relationship between adverse childhood experiences (ACEs) and health decline in adulthood. MethodThe ACE-IQ is a 29-item measure designed to assess exposure to ACEs categorized into three groups: maltreatment (all forms of physical and emotional ill-treatment, sexual abuse, neglect), family dysfunction and violence outside the home. The questionnaire was translated and back-translated according to guidelines and methodology defined by the World Health Organization, 2011 (WHO). A total of 367 participants were recruited to take part in this study. After analyzing descriptive statistics, we assessed the questionnaire’s internal structure using principal component factor analysis. Based on classical test theory we calculated internal consistency and test–retest reliability. We then performed an item response theory (IRT) analysis using a partial credit model (PCM) for polytomous data, which was applied to each of the three dimensions of the ACE-IQ. ResultsTaken together, our results suggest that while the original three-factor structure was not fully supported, the psychometric properties were good. A high degree of overall internal consistency was demonstrated with an alpha coefficient of 0.80. For the three dimensions of the questionnaire Cronbach’s alphas ranged between 0.77 and 0.41. In terms of temporal stability, we observed a good intra-class correlation of 0.79 for the entire questionnaire, with temporal stability for the three dimensions ranging between 0.78 and 0.61. Finally, our IRT analysis revealed discrepancies on certain items for each of the three dimensions, which confirm, and complete findings previously obtained through classical analysis. ConclusionThe French translation of the ACE-IQ has good psychometric properties which make it a reliable and valid measure for assessing ACEs in French-speaking populations, both for research purposes and clinical practice.Further validation will be necessary, however, to account for different cultural characteristics and perspectives between French speaking countries.