(PS7-15) Classifying Emotional Eating by Valence, Amount Eaten, and Contextual Factors: A Latent Class Analysis

Introduction: The validity of self-reported emotional eating scores has been called into question recently. This is in part due to limitations of existing emotional eating questionnaires that do not capture the nuances within emotions (e.g., positive, negative) and eating behaviors (e.g., undereating, overeating) that may be involved in emotional eating. In the current study, latent class analysis was utilized to better characterize subgroups that may exist among self-reported “emotional eaters.”

Method: This study recruited an online sample of healthy adults (N = 389; Mage = 37.42; SD = 12.84) through Prolific Academic. Participants responded to self-report measures of emotional eating and contextual factors, as well as measures of distress intolerance, psychological flexibility, disordered eating, loss of control (LoC) eating, emotional dysregulation, and psychosocial impairment. Latent class analyses were conducted in Mplus and ANOVAs were conducted in SPSS.

Results: A four-class solution emerged, i.e., Class 1) overeating in response to negative emotions (e.g., scared, anxious) and undereating in response to positive emotions (e.g., optimistic, confident; 20.8%), 2) overeating in response to positive emotions and undereating in response to negative emotions (39.3%), 3) undereating in response to both negative and positive emotions (22.6%), and 4) moderate eating (not over- or under-eating) in response to both negative and positive emotions (17.2%). Class 1 had higher eating disorder psychosocial impairment (p < .001, η2 = .09), psychological inflexibility (p < .001, η2 = .04), distress intolerance (p = .001, η2 = .04), and difficulties with emotion regulation (p = .002, η2 = .04) compared to other classes. Class 4 had higher disordered eating symptoms (p < .001, η2 = .15) and loss of control eating (p < .001, η2 = .17) compared to other classes. Both these classes endorsed similar contextual factors prompting emotional over- or under-eating, regardless of emotional valence (e.g., being alone, hungry, at a celebration/party, and end of day). 



Discussion: Findings advance the classification of self-reported emotional eating via the assessment of valence (e.g., negative and positive) and amount eaten (e.g., undereating and overeating), and they further support evidence that negative and positive emotional eating are distinct constructs. Consistent with prior research, the emotional eating class most closely associated with psychopathology (e.g., disordered eating impairment) was Class 1 (negative over- and -positive under emotional eating). Given positive emotional eating and contextual factors of emotional eating are historically ignored, interventions that target these aspects of emotional eating may prevent the development of clinical eating disorders and obesity.