Emotional Eating in Adults: The Role of Sociodemographics, Lifestyle Behaviors, and Self-Regulation-Findings from a U.S. National Study

Abstract

Background: Emotional eating, the tendency to overeat in response to negative emotions, has been linked to weight gain. However, scant evidence exists examining the prevalence and correlates of emotional eating among large samples of adults in the United States (U.S.). Hence, we examine the relationship among individual and socioeconomic factors, health behaviors, and self-regulation with emotional eating patterns among U.S. adults.

Methods: Cross-sectional analysis of 5863 Family Health Habits Survey participants. Multivariable, ordered, logistic regression was employed to examine the relationship between the frequency of the desire to eat when emotionally upset (never, rarely, sometimes, often, and very often) and the independent variables.

Results: Analysis reveals that 20.5% of the sample tended to emotionally eat often or very often. Being female, non-Hispanic White, and of younger age were all related to a higher likelihood of emotional eating. Additionally, inability to delay gratification (impatience) was related to an 18% increased likelihood (95% confidence interval (CI) 1.05-1.33) for emotional eating. Finally, emotional eating was significantly related to more frequent fast-food consumption.

Conclusions: Program planners might need to develop targeted interventions aimed at enhancing emotional regulation skills while addressing these less healthful behaviors (e.g., fast-food intake) with the goal of obesity and chronic disease prevention.

Keywords: emotional eating; lifestyle behaviors; self-regulation; sociodemographics.

Figure 1

Socioeconomic variables, Self-regulation, and emotional eating ^a: ordered logistic regression ^b. ^a Emotional eating is based on a survey item pertaining to the desire to eat when emotionally upset or stressed. ^b The circles represent odds ratios, and the range bars are the 95% confidence interval. When the horizontal line crosses the vertical line, the relationship is not statistically significant.

Figure 2

Self-rated health, obesity, lifestyle behaviors, and emotional eating: ordered logistic regression ^a,b. ^a The circles represent odds ratios, and the range bars are the 95% confidence interval. When the horizontal line crosses the vertical line, the relationship is not statistically significant. ^b The model adjusts for age, sex, race/ethnicity, marital status, annual household income, education, household size, and self-regulation.