Validation d’une version francophone de l’Anorexia Nervosa Stages of Change Questionnaire

Abstract

BackgroundNumerous authors regard motivation to change as a key feature in the treatment of individuals with anorexia nervosa. Although several questionnaires are available to assess motivation to change, most of them were not intended for adolescents nor individuals with anorexia nervosa. The Anorexia Nervosa Stages of Change Questionnaire (ANSOCQ) was developed to assess motivation to change specifically in individuals with anorexia nervosa. This study aimed to validate a French translation of the ANSOCQ based on the three-factor model identified elsewhere. MethodologyEighty adolescents suffering from anorexia nervosa and two from an eating disorder not otherwise specified were recruited at their arrival in a specialized eating disorder inpatient treatment program. Fifty-two adolescents participated again at the time of their discharge. Respondents completed the ANSOCQ, the University of Rhode Island Change Assessment Scale, a single item derived from the Motivation Stages of Change for Adolescents Recovering from an Eating Disorder, and the Eating Disorder Inventory-3. Cronbach's Alpha, Pearson's and Spearman's correlations, ANOVA and multiple regressions were performed. ResultsThe global scale and subscales of the ANSOCQ demonstrated excellent internal consistency (α≥0.82). The global scale was significantly correlated with both the URICA (r=0.46) and the MSCARED (rs=0.68). Participants’ scores on the three subscales of the ANSOCQ were significantly higher at the end of their hospitalization (ps<0.05), especially for the Ego-Alien Aspects subscale. Youths’ scores on the Weight Gain subscale were significantly lower than their scores on both Eating, Shape, and Weight Concerns and Ego-Alien Aspects subscales (ps<0.05). The ANSOCQ subscales significantly explained 52% and 41% of variance on the Eating Disorder Risk Composite Scale and the General Psychological Maladjustment Composite Scale, respectively. Furthermore, while the Weight Gain subscale was uniquely related to the former scale (β=−0.38), the Eating, Shape, and Weight Concerns subscale was uniquely associated with the latter scale (β=−0.39). ConclusionThese results support the psychometric qualities of this French translation of the ANSOCQ. Given the differences observed in the present study, the use of the ANSOCQ subscales also seems promising. It is possible that adolescents report higher motivation to change aspects of their difficulties that they recognize more, that are associated with fewer secondary benefits, and that entails greater unwanted repercussions. However, further studies are needed to confirm the factorial structure of the ANSOCQ.