Development and Validation of the Psychometric Properties of the FitMIND Foundation Sweets Addiction Scale—A Pilot Study

Background: The rising consumption of ultra-processed foods, especially those high in added sugars, poses a growing public health concern. Although several tools exist to assess food addiction, there is a lack of validated instruments specifically designed to measure addiction-like behaviors related to sweet food intake. Objectives: This study evaluates the psychometric properties of the FitMIND Foundation Sweets Addiction Scale (FFSAS), adapted from the Yale Food Addiction Scale 2.0 (YFAS 2.0), using data from Polish adults recruited through the FitMIND Foundation. Methods: The FFSAS was evaluated by 11 expert judges on four criteria: clarity, content validity, linguistic appropriateness, and construct representativeness. Afterwards, 344 adult volunteers (mean age 40.6 ± 10.7 years, 78% female, mean body mass index (BMI) 27.86 kg/m2) completed online FFSAS and provided demographic data, BMI, and self-reported sweets consumption. Internal consistency was assessed with Cronbach’s alpha and external validity was examined through Spearman’s correlations. Moreover, we conducted Exploratory and Confirmatory Factor Analyses (EFA and CFA). Results: Content validity of the FFSAS was supported by expert validation. The scale demonstrated good overall internal consistency (α = 0.85), with specific criteria such as tolerance (α = 0.916) and withdrawal (α = 0.914) showing particularly high reliability. The FFSAS total score was moderately correlated with sweets consumption frequency (ρ = 0.39, p < 0.05) and feelings of guilt (ρ = 0.35, p < 0.05). Exploratory factor analysis (EFA) revealed a robust three-factor structure, explaining 68.6% of the variance; the individual factors (subscales) derived from this structure demonstrated excellent internal consistency (Cronbach’s α ranging from 0.951 to 0.962). Sampling adequacy was high based on Kaiser–Meyer–Olkin measure (KMO = 0.956). Confirmatory factor analysis (CFA) indicated suboptimal model fit (Comparative Fit Index (CFI) = 0.74, Tucker–Lewis Index (TLI) = 0.69, Root Mean Square Error of Approximation (RMSEA) = 0.14), with a significant chi-square test (χ2 = 3761.76, p < 0.001). Conclusions: This pilot study demonstrated that the FFSAS may be a promising tool for assessing sweet food addiction in adults. Future research should focus on assessing the FFSAS’ suitability on more diverse populations in other countries for further validation.