Evidence Base for Interpersonal Psychotherapy for Binge Eating Disorder
On the basis of IPT's initial success in the treatment of BN (Fairburn et al., 1991), Wilfley et al. (1993, 2000) developed IPT for BED in a group format. During this time, they worked with many clients who presented with chronically unfulfilling relationships, issues that were well suited to be addressed in a group format. Therefore, they adapted new strategies to specifically address such interpersonal deficits. For example, in the current format of group IPT for BED, group members with interpersonal deficits are strongly encouraged to view the group as a “live” social network. This social milieu is designed to decrease social isolation, support the formation of new social relationships, and serve as a model for initiating and sustaining social relationships outside of the therapeutic context (Wilfley et al., 1998). Clients with BED commonly experience shame and selfstigmatization, which potentially contribute to the maintenance of the disorder. Thus, group therapy inherently offers a radically altered social environment for clients with BED, who typically keep shameful eating behaviors hidden from others.
IPT has demonstrated effectiveness in the treatment of BED. As in the case of CBT for BN, CBT for BED has an extensive evidence base establishing specific and robust treatment effects (Devlin et al., 2005; Grilo, Masheb, & Wilson, 2005; Kenardy, Mensch, Bowen, Green, & Walton, 2002; Nauta, Hospers, Kok, & Jansen, 2000; Ricca et al., 2001; Telch, Agras, Rossiter, Wilfley, & Kenardy, 1990; Wilfley et al., 1993). Two randomized trials have compared IPT with CBT and found that IPT has similar effects to CBT in the treatment of BED. The first study (Wilfley et al., 1993) compared group CBT, group IPT, and a wait-list control group. The findings revealed that both treatments were more effective at reducing binge eating than the control group and resulted in similar significant reductions in binge eating in both the short and the long term. The second study (Wilfley et al., 2002) included a substantially larger sample size and demonstrated equivalent short- and long-term efficacy for CBT and IPT in reducing binge eating and associated specific and general psychopathology Approximately 60% of the clients remained abstinent from binge eating at 1 -year and 4-year follow-ups (Hilbert et al., 2012; Wilfley et al., 2002). In contrast to the literature on IPT for BN, CBT and IPT had identical time courses for almost all outcomes in the BED studies.
In a follow-up analysis of treatment predictors of long-term outcome in the Wilfley et al. (2002) study, clients with a greater extent of interpersonal problems at both baseline and midtreatment showed poorer treatment response to both CBT and IPT (Hilbert et al., 2007). An important caveat of this finding, however, is that those individuals with greater interpersonal problems were not surprisingly also those who had more Axis I and Axis II psychiatric disorders and lower self-esteem than those with less severe interpersonal problems. Such individuals are likely in need of more intensive or extended treatment. Supporting this assertion, Markowitz, Skodol, and Bleiberg (2006), in a study of IPT for individuals with borderline personality disorder and comorbid depression, suggested that extending IPT effectively improves the disorder. A preliminary examination of clients in the larger BED cohort indicated that those in IPT maintained reductions in binge eating and disordered eating cognitions at least 5 years posttreatment (Bishop, Stein, Hilbert, Swenson, & Wilfley, 2007). These data may suggest evidence for good maintenance of change for clients with BED who are treated with IPT.
Results from a more recent multisite trial comparing individual IPT, behavioral weight loss treatment, and CBT guided self-help (CBT-GSH) for the treatment of BED have pointed to the importance of making a clear connection between interpersonal problems and binge eating symptoms in the delivery of IPT. Similar to Wilfley et al.'s (2002) trial, in this multisite study the counselors linked interpersonal functioning to disordered eating symptoms throughout the course of IPT. In this study, IPT was rated as most acceptable to clients, and the dropout rate was significantly lower in IPT than in the other two interventions (Wilfley, Wilson, & Agras, 2008). At 2-year follow-up, IPT and CBT-GSH were significantly more effective than behavioral weight loss in eliminating binge eating. Furthermore, compared with the other two treatments, IPT also produced greater reduction in binge episodes for clients presenting with low self-esteem and more disordered eating behaviors and cognitions. CBT-GSH was generally effective only for those with low ED psychopathology. In this trial, compared with Wilfley et al. (2002; see also Hilbert et al., 2007), individuals with more psychopathology showed notably greater improvements in IPT than CBT-GSH. This finding was in concert with Hilbert et al.'s (2007) follow-up data, which suggested that greater disordered eating serves as a moderator in predicting poorer outcome in CBT.
In general, compared with Caucasian participants, individuals of other ethnic minorities demonstrated less retention in the multisite study (Wilfley et al., 2008). Although no treatment by ethnicity effects were found in this regard, attrition for minority participants in IPT was very low and dropout rates by minorities in CBT-GSH were very high. The low enrollment of minority participants across sites prevents us from drawing definitive conclusions. Nevertheless, this pattern aligns with the finding that IPT was particularly effective for African American participants in the previously described multisite study for individuals with BN (Chui et al., 2007). We posit that perhaps the personalized nature of IPT (e.g., problem areas and goals are developed on the basis of each individuals unique social environment) is modifiable to, and thus particularly acceptable to, people of various cultures and backgrounds.
Several recommendations may be drawn from the recent multisite study (Wilfley et al., 2008). CBT-GSH could be considered the first-line treatment for most individuals with BED, and IPT could be recommended for clients with low self-esteem and high ED psychopathology. Alternatively, IPT could be considered a first-line treatment for BED on the basis of several factors: IPT has been shown to be effective across multiple research sites, is associated with high retention across different client profiles (e.g., high negative affect, minority groups), and demonstrated superior outcomes to behavioral weight loss overall and to CBT-GSH among a subset of clients with high disordered eating psychopathology and low self-esteem. Counselors and clients should consider these alternatives when deciding the best approach to treatment. Finally, behavioral weight loss should not be considered a first-line treatment when treating individuals with BED. In summary, the literature suggests that IPT is an efficacious treatment alternative to CBT for BED.