Validating CRAVED: a brief screening instrument for ultra-processed food addiction.

BACKGROUND

Addictive-like eating behaviours related to ultra-processed foods have gained increasing research and clinical attention. The Yale Food Addiction Scale (YFAS) is the most widely used measure of food addiction. It was primarily designed as research tool, so its length and scoring complexity may limit routine clinical use. CRAVED is a brief six-item screening tool developed to assess core addiction-related eating behaviours using a simple yes/no response format.

OBJECTIVE

This study aimed to provide a preliminary psychometric evaluation of CRAVED.

METHODS

A cross-sectional secondary analysis was conducted using anonymised baseline data from 117 adults enrolled in a community-based intervention for addictive-like eating behaviours. Participants completed CRAVED, YFAS 2.0, and the Binge Eating Scale (BES). Analyses examined internal consistency, construct validity, logistic regression, receiver operating characteristic (ROC) performance, and agreement with YFAS diagnostic classification.

RESULTS

CRAVED scores were positively associated with YFAS symptom scores (r = 0.55, p < 0.001) and BES scores (r = 0.56, p < 0.001). Higher CRAVED scores significantly predicted YFAS-defined food addiction. ROC analysis demonstrated acceptable discrimination (AUC = 0.75, 95% CI 0.64-0.85). A threshold of ≥3 showed high sensitivity for identifying YFAS-positive cases. CRAVED scores were significantly higher among participants meeting YFAS criteria and increased across binge eating severity groups. Internal consistency was modest (Cronbach's α = 0.58, McDonald's ω = 0.68), consistent with the brief multidomain binary structure of the tool.

CONCLUSION

These findings provide preliminary support for CRAVED as a brief and clinically feasible screening tool for addictive-like eating behaviours. CRAVED may be useful for rapid identification in healthcare and community settings, although it should not be considered a standalone diagnostic instrument. Further validation in larger and more diverse samples is warranted.