The reward-based eating drive scale: a self-report index of reward-based eating

E.S. Epel*, A.J. Tomiyama, A.E. Mason, B.A. Laraia, W. Hartman, K. Ready, M. Acree, Tanja C. Adam, S. St Jeor, D. Kessler

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Why are some individuals more vulnerable to persistent weight gain and obesity than are others? Some obese individuals report factors that drive overeating, including lack of control, lack of satiation, and preoccupation with food, which may stem from reward-related neural circuitry. These are normative and common symptoms and not the sole focus of any existing measures. Many eating scales capture these common behaviors, but are confounded with aspects of dysregulated eating such as binge eating or emotional overeating. Across five studies, we developed items that capture this reward-based eating drive (RED). Study 1 developed the items in lean to obese individuals (n = 327) and examined changes in weight over eight years. In Study 2, the scale was further developed and expert raters evaluated the set of items. Study 3 tested psychometric properties of the final 9 items in 400 participants. Study 4 examined psychometric properties and race invariance (n = 80 women). Study 5 examined psychometric properties and age/gender invariance (n = 381). Results showed that RED scores correlated with BMI and predicted earlier onset of obesity, greater weight fluctuations, and greater overall weight gain over eight years. Expert ratings of RED scale items indicated that the items reflected characteristics of reward-based eating. The RED scale evidenced high internal consistency and invariance across demographic factors. The RED scale, designed to tap vulnerability to reward-based eating behavior, appears to be a useful brief tool for identifying those at higher risk of weight gain over time. Given the heterogeneity of obesity, unique brief profiling of the reward-based aspect of obesity using a self-report instrument such as the RED scale may be critical for customizing effective treatments in the general population.
Original languageEnglish
Article numbere101350
JournalPLOS ONE
Volume9
Issue number6
DOIs
Publication statusPublished - 1 Jan 2014

Cite this