Prevalence and Characteristics of Binge Eating in an Adolescent Community Sample

Abstract

The objective of this article was to investigate the prevalence and psychological correlates of binge eating among adolescents. Self-report questionnaires were administered to a community sample of 708 adolescents (M _age = 14 years). Adolescents reporting loss of control over eating (17% of the sample) reported more eating pathology and more general maladjustment compared to those without loss of control. The combined presence of subjective and objective binge episodes was related to the most severe pattern of eating pathology. The results suggest that loss of control over eating is a relatively frequent experience among adolescents from the general population and it is related to eating pathology and maladjustment.

This work was supported by the Special Research Funds of Ghent University. Bart Soenens is a postdoctoral researcher at the Fund for Scientific Research Flanders (FWO).

Notes

¹Comparisons between participants who provided full LC data (n = 646) and those who did not (n = 62) showed that both samples did not significantly differ with regard to age, BMI, adjusted BMI, and the EDE–Q weight concern and shape concern subscales. We did obtain differences between both groups in terms of gender distribution, global self-worth, depressive symptoms, EDE–Q restraint and eating concern, and each of the EDI subscales. Boys were overrepresented in the group that did not provide LC data. Further, participants who did not provide LC data scored higher on each measure except for global self-worth. Given these findings, some caution is warranted in generalizing the results obtained in the sample that provided LC data to the full original sample.

Note: BMI = body mass index; M = male; F = female; LC = loss of control over eating; NoLC = no loss of control over eating.

∗p < .05.

Note: All correlation coefficients were significant at the p < .001 level.

Note: LC = loss of control over eating; NoLC = no loss of control over eating.

²In spite of the fact that the use of the one LC episode per month criterion is consistent with the overweight literature, we performed an additional set of analyses to explore whether results differ between those participants reporting LC less or more frequently. Specifically, we subdivided our sample into youth experiencing NoLC (n = 538), youth experiencing from 1–3 LC episodes per month (n = 78), youth experiencing at least one LC episode per week (n = 15; as proposed by Fairburn & Cooper, 1993), and finally youth experiencing at least 2 LC episodes per week (n = 15; as proposed by the DSM; APA, 2000). Post hoc comparisons between these subsamples indicate that with regard to the EDE–Q and EDI subscales, there is a clear tendency for the three LC groups to differ significantly (p < .05) from the NoLC group (except for the EDI Body Dissatisfaction scale), whereas the three LC groups generally do not differ from each other (see the EDI Bulimia scale for an exception). With regard to the CES–D and Self-Esteem scale, no significant differences were obtained between those with the highest frequencies of LC episodes and those with lower frequencies (p > .05). These results show that the main differences in terms of eating pathology and well-being occur between youth experiencing no LC episodes at all and youth experiencing at least one to three LC episodes per month. Further distinctions between youth experiencing LC episodes in terms of the frequency of their LC experiences do not seem to result in systematic differences in terms of eating pathology and adjustment.

Note: Means in the same row that do not share subscripts differ at p < .05 in the Tukey honestly significant difference comparison.

∗p < .05. ∗∗p < .01. ∗∗∗p < .001.