RESUMEN
OBJECTIVE: The objective was to evaluate the psychometric properties of the Mindful Eating Questionnaire (MEQ) in Brazilian subjects with type 2 diabetes mellitus (T2DM) and validate a Brazilian version of the MEQ for adults with T2DM (MEQ-DM). METHODOLOGY: Baseline data from the multicentre Nutritional Strategy for Glycaemic Control in Patients with Type 2 Diabetes Mellitus (NUGLIC) trial were used. Construct validity was assessed using exploratory factor analysis (EFA). The root mean square error of approximation (RMSEA), comparative fit index (CFI) and TuckerâLewis index (TLI) fit indices indicated the adequacy of the model. The reliability of the questionnaire was evaluated considering the different factor loadings. Criterion validity was tested by correlating the MEQ-DM with sociodemographic variables, body mass index (BMI) and physical activity levels. RESULTS: A total of 370 participants were included, who were mostly female (60.8 %) and had a median age of 61 (54-67) years. The EFA results supported the two-factor structure of the 25-item MEQ-DM: disinhibition and awareness. The results of the fit indices (RMSEA = 0.04; CFI = 0.95 and TLI = 0.94) and composite reliability (disinhibition = 0.84 and awareness = 0.81) were consistent. The criterion validity analysis indicated a significant association between MEQ-DM scores and age, sex, civil status, education level, BMI and physical activity (p < 0.05). CONCLUSION: When explored with Brazilian adults with T2DM, the MEQ-DM presented a factorial model with two dimensions: disinhibition and awareness. This model must be confirmed in future studies with Brazilians with T2DM.