Testing the original and the extended dual-pathway model of lack of control over eating in adolescent girls. A two-year longitudinal study

• We prospectively tested and compared the original and extended dual pathway model.
• Both models fit the data well and all proposed mediational pathways were supported.
• The extended model accounted for a greater proportion of variance in binge eating.
• A bi-directional relation between dietary restriction and negative affect was found.

Stice's (1994, 2001) dual pathway model proposed a mediational sequence that links body dissatisfaction to lack of control over eating through dieting and negative affect. Van Strien et al. (2005) extended the negative affect pathway of the original dual pathway model by adding two additional intervening variables: interoceptive deficits and emotional eating. The purpose of this study was to test and compare the original and extended model using prospective data. Both types of loss of control over eating (i.e., subjective and objective binge eating) were evaluated. Data collected from 361 adolescent girls, who were interviewed and completed self-report measures annually over a 2-year period, were analysed using structural equation modeling. Although both models provided a good fit to the data, the extended model fit the adolescent girls' sample data better and accounted for a greater proportion of variance in binge eating than the original model. All proposed mediational pathways of both models were supported and all indirect effects examined through bootstrap procedure were significant. Although our results confirmed the validity of both models and extended previous findings to an early- to middle adolescent group, the bi-directional relationship between dietary restriction and negative affect suggests that the association between these key risk factors for binge eating are more complex than outlined in both the original and extended dual-pathway models.