The Mediation Effect of Body Disapproval on Eating Disordered Behavior Among Adjudicated Male Youth

Background: Empirical studies have identified increasing rates of eating disordered behavior (EDB) among male youth. Negative consequences of EDB among males include substance abuse, mental health problems, and suicidal ideation. Previous research has suggested that adjudicated male youth have higher rates of EDB compared to their non-adjudicated peers. However, research has not yet examined the risk mechanisms of EDB in this population. Failure to understand the risk mechanisms of EDB among adjudicated male youth prevents the development of effective treatments. This study explores paths linking potential risk factors (including body disapproval, substance abuse, childhood sexual abuse, and exposure to pornography) to EDB among adjudicated male youth.

Methods: The study has a cross-sectional design. Data were collected from 800 adjudicated male youth living at seven residential facilities in a Midwestern state and an Eastern state. All male youth living at the facilities were invited to participate the study. Of the 800 participants, 71.6% were adjudicated for sexual offenses, and 29.4% were adjudicated for non-sexual crimes. Measures included the Millon Adolescent Clinical Inventory and the Childhood Trauma Questionnaire. Participants also provided information about their demographic characteristics and pornography exposure.

Path analysis within a structural equation modelling (SEM) framework was used. Two competing models were analysed and compared: 1) a model with full mediation of the effects of predictor variables (e.g., substance abuse, childhood sexual abuse, pornography exposure) on EDB through body disapproval; and 2) a model with partial mediation of the effects of the same predictor variables on EDB through body disapproval. We compared fit of the two models using the χ² difference test.

Results: The fully mediated model [χ² (4, N = 800) = 8.586, p = 0.072; RMSEA = 0.038; CFI = 0.996; and TLI = 0.991] and the partially mediated model fit the data well [χ² (1, N = 800) = 0.636, p = 0.425; RMSEA = 0.000; CFI = 1.000; and TLI = 1.003]. Based on the model fit comparison (Δχ² (3) = 7.95, p = 0.047), we retained the partially mediated model. Substance abuse (b = .044, p <.05) and body disapproval (b = .888, p <.05) were directly and statistically significantly related to EDB. In addition, substance abuse had an indirect effect on EDB through its relationship to body disapproval (b = .222, p < .001). Childhood sexual abuse (b = .125, p < .001) and pornography exposure (b = .143, p < .001) had only indirect effects on EDB through their relationships to body disapproval. Overall, the model explained 19.3% of the variation in body disapproval and 83.1% of the variation in EDB.

Conclusion and Implications: Findings suggest that body disapproval mediated the effects of substance abuse, childhood sexual abuse, and porn exposure on EDB among adjudicated male youth. The study suggests body image may be an important link in the effects of early risk factors on EDB. Future research on the role of body disapproval on EDB and other risk behaviors in this population is needed. Implications for research and practice will be discussed.