Testing the psychometric and conceptual validity of the IUM in the Italian context: preliminary findings

In Italy the one-year prevalence of Generalized Anxiety Disorder (GAD) is 0.5% and its lifetime prevalence is 1.9%; furthermore, many people do not meet the diagnostic criteria for GAD but experience worries in a subclinical form. The Intolerance of Uncertainty Model (IUM) is one of the most experimentally and clinically validated theoretical approaches targeting GAD. According to the IUM, four main components are associated with the development and maintenance of worry: Intolerance of Uncertainty (IU), Negative Problem Orientation (NPO), Positive Beliefs about Worry (PBW) and Cognitive Avoidance (CA). Several randomized control trials have demonstrated the effectiveness of therapeutic interventions based on the IUM in reducing GAD symptoms. Nonetheless, such a theoretical framework is not so well established in Italian clinical practice and to date worry and anxiety symptoms are managed through the implementation of traditional and more generic theoretical models of anxiety. The main aim of the present study was to provide preliminary validation of the IUM in the Italian context, by testing both the psychometric and conceptual validity of the model. One hundred and eighty-five undergraduate students (58.9% males) aged between 19 and 29 (M=21.94; SD=1.87) attending Psychology and Engineering courses were recruited. They completed the following self report measures, administered in a rotated sequence to control for order effects: Intolerance of Uncertainty Scale-12 (IUS- 12), Why Worry III (WW-III), Revised Cognitive Avoidance Questionnaire (R-CAQ), Negative Problem Orientation Questionnaire (NPOQ), Penn State Worry Questionnaire (PSWQ). Except for the PSWQ these self-report measures had not previously been validated in Italian so a forward-back translation procedure was employed. To assess the psychometric properties of the questionnaires internal consistency, correlations between measures and gender differences were tested. Then moderated mediation analysis was performed to assess the construct validity of the model. The four self-report measures preliminary showed good psychometric properties: analyses revealed high internal consistency (0.77<?s<0.94), small-medium range correlations between constructs (0.15<rs<0.47), and absence of gender differences. Mediation analysis revealed that NPO and PBW, but not CA, mediate the association between IU and worry. Moreover, moderated mediation analysis showed that IU moderates the effects of the indirect path through NPO and PBW on worry. These preliminary findings are the first to support the validity of the model in an Italian sample. Future studies aiming to validate such measures in the Italian context (e.g. community and clinical samples) are needed. Furthermore, present results also support the conceptual validity of the IUM and are in line with those previously found by Bottesi et al. (2012) among British undergraduates. The moderated mediation analysis highlights a specific mediational role played by NPO, PBW and CA in the path between IU and worry and it suggest that IU also acts as a moderator, thus outlining the key role played by current levels of IU on the degree of reported worry.