Aim: To assess the dimensionality and psychometric properties of an Italian version of the Yale Food Addiction Scale (YFAS) in a sample of obese/overweight patients attending low-energy diet therapy.
Methods: Participants were 300 overweight and obese patients who were admitted to a private medical center in Rome, Italy. Controls were 300 (231 women and 69 men) adults from the general population. All of the participants were administered the YFAS and the binge eating scale (BES).
Results: The one-factor model of the YFAS reported in previous studies did not fit the data χ(2)(209) = 466.69, p < 0.001, root mean square error of approximation (RMSEA) = 0.07; 90% CI: 0.06/0.08; comparative fit index (CFI) = 0.91; weighted root mean square residual (WRMR) = 1.40]. Through item analysis, it was suggested that five items (items #10, #11, #22, #24, and #25) with low item-total correlations should be removed from the measure. A 16-item one-factor model revealed a better fit to the data (χ(2)(104) = 174.56; p < 0.001; RMSEA = 0.05; 90% CI: 0.04/0.07; CFI = 0.96), although the WRMR was slightly higher than that suggested as an indicator of good fit (WRMR = 1.01). The YFAS-16 had satisfactory internal consistency; it was able to discriminate obese patients from controls and strongly correlated with BES scores.
Conclusion: The YFAS-16 assesses all of the "symptoms" represented in the original version and has satisfactory psychometric properties, although the percentage of food addiction diagnoses according to the YFAS-16 is lower than the percentage of diagnoses according to the original version of the questionnaire.