(PS7-8) Associations Between Food Insecurity and Restrictive Eating Pathology in a Large and Diverse College Sample

College students are particularly vulnerable to experiencing food insecurity (FI), due to high tuition rates, increasing grocery prices, and a weaker market for paid work. Evidence suggests that 20–50% of college students experience some level of FI. Moreover, FI is associated with a variety of negative psychological outcomes, including heightened risk for eating disorders (EDs). Numerous studies have identified a link between FI with binge eating and purging; however, few have examined whether FI heightens risk for weight-related restrictive eating. Further, little is known about the processes through which FI may increase risk for ED behaviors, and whether comorbid cognitive and affective experiences (e.g., depression, anxiety, overvaluation of weight) may influence relations between FI and EDs. We examined (1) cross-sectional and longitudinal associations between FI and anorexia nervosa (AN), atypical AN, and weight-control behaviors (e.g., fasting, driven exercise), and (2) the moderating effects of depression, anxiety, and weight valuation on these associations.

Methods: College students (N = 1727; 46.1% White, 49.8% food-secure) completed an online survey reporting demographics, socioeconomic status, and psychiatric symptoms during their freshman, sophomore, and junior years. Separate hierarchical logistic regressions examined associations with baseline and follow-up eating disorder outcomes from: Step 1) FI severity, Step 2) mood or weight valuation, and Step 3) the interaction between FI and mood or weight valuation.

Results: At Step 1, FI severity was cross-sectionally associated with increased odds of atypical AN (OR = 1.53, p = .002), fasting for weight control (OR = 1.13, p = .013), and driven exercise (OR = 1.12, p = .047), but not AN or purging (ps > .05). In the final model, FI remained significantly associated with atypical AN after accounting for depression (OR = 1.54, p = .008), anxiety (OR = 1.61, p = .002), and overvaluation of weight (OR = 1.64, p < .001). FI remained significantly associated with fasting after controlling for weight valuation (OR = 1.14, p = .009), but not depression or anxiety (ps > .05). FI was no longer associated with driven exercise after controlling for depression, anxiety, or weight valuation (ps > .05). Weight valuation was associated with a lesser relation between FI and atypical AN (OR = 0.87, p = .032). No other moderating effects of weight valuation, depression, or anxiety were detected.  Food insecurity did not longitudinally predict AN, atypical AN, or any weight control behavior (ps > .05). Discussion: FI is associated with increased odds of atypical AN, even after accounting for depression, anxiety, and weight valuation, suggesting an FI-specific mechanism of atypical AN. Moreover, associations among FI and weight-control behaviors were primarily explained through mood and overvaluation of weight. Therefore, weight-control behaviors may function to regulate emotions in response to heightened distress stemming from FI. In all, these findings further dispel the stereotype that restrictive EDs occur only among affluent individuals and emphasize the need to consider FI in the assessment, treatment, and modeling of restrictive eating pathology.