by Michael D. Anestis, M.S.
In the current issue of Behaviour Research and Therapy, Natalie Pretorius and colleagues (2009) published a study that examined the degree to which an internet-based approach to cognitive behavioral therapy (CBT) might prove useful in the treatment of bulimia nervosa (BN) in adolescents. As we have discussed on PBB in prior articles, Division 12 of the American Psychological Association lists CBT as one of the gold standard treatments for BN, describing its research based, as well as that of interpersonal psychotherapy (IPT), as strong. Like most other mental illnesses, however, a significant number of individuals who meet criteria for BN do not seek treatment. This is true for a variety of reasons, including shame about symptoms, stigma about seeking treatment, and a lack of effective diagnostic procedures and treatment referrals in primary care environments. Finding ways to increase remote access to empirically supported treatments (ESTs) thus appears to be of potentially great value, as this would allow individuals to privately seek treatment in an environment of their choosing. Along these lines, James Mitchell and colleagues (2008) found that there was no statistically significant difference in outcomes for individuals who received CBT over the phone compared to those who received CBT face-to-face, thereby offering empirical support for the potential value of remote treatments. Such findings provide a strong rationale for examining other forms of remote treatments, including those based on the internet.
Pretorius and her colleagues (2009) thus attempted to examine the degree to which the apparent potential of web-based CBT would translate to actual results. To do so, they recruited a sample of 101 participants (98 females, 3 males) from eating disorder clinics across the United Kingdom as well as from BEAT (Beating Eating Disorders), a charity specializing in helping individuals with eating disorders find useful resources. All participants were diagnosed through a semi-structured clinical interview and met DSM-IV-TR criteria for either BN or eating disorder not otherwise specified with bulimic symptoms (EDNOS-BN). Individuals in the latter category either did not binge or purge often enough to meet BN criteria (on average twice per week for at least three months) or they used compensatory behaviors in the absence of objective binges. All participants were between the ages of 13 and 20.
The treatment itself consisted of eight interactive online sessions, each of which required 30-40 minutes to complete. Based upon their responses to interactive sections of each session, participants received personalized feedback. In addition, each participant was provided with workbooks, homework, and an audio session on controlling anxiety to accompany each session. Participants were able to access an online message board moderated by the researchers in order to ensure peer support throughout treatment. Additionally, participants received weekly email support from a therapist trained in CBT.
The results from this study, while promising to some degree, were fairly mixed. 17% of the participants did not complete a single web-based session and the average number of sessions completed was only 3 out of 8. In other words, on average, participants took part in less than half of the treatment protocol. Follow-up data was attained at three months and six months after the initial interview; however, only 51.5% of the sample provided information at the 3-month and only 62.3% provided information at the 6-month. So, in addition to the large number of individuals who did not complete much or any of the treatment, many individuals did complete treatment did not report on their symptoms throughout the course of the study.
Despite the low level of overall participation, the authors were able to examine the impact of treatment on BN symptoms. Participants showed improvement on their number of objective binge episodes, vomit episodes, laxative episodes, overall eating disorder symptoms levels, and body mass index (BMI). This is promising; however, a closer look at the numbers diminishes that promise quite a bit. At baseline, 1% of the sample who ended up completing treatment had been free of objective binges, vomiting, and laxative use for one month. At three months, this number increased to 10%. At six months, the number increased to 17%. When the authors looked at a less stringent outcome measure - percentage of participants who were either abstinent from or sub-clinical in binge episodes, vomiting, and laxative use - the numbers improved a bit. 9% met this criteria at baseline, 25% at three months, and 29% at six months. When the entire sample was considered, those numbers decreased.
So, 71% of the sample that completed treatment still utilized binge eating, vomiting, and/or laxatives at a clinically significant level and 83% reported that they were using these behaviors at least sub-clinically. These numbers, quite frankly, are unimpressive. As a comparison, face-to-face CBT tends to result in abstinence rates of 30-50% and a reduction of binge eating and purging behaviors of 80% (Wilson & Fairburn, 2002). In one of the better trials of CBT for BN, Wilson, Fairburn, Agras, Walsh, and Kraemer (2002) found that, on average, purging behaviors were reduced by 62% and binge eating was reduced by 72%. Clearly then, Pretorius and colleagues (2009) did not find results for internet-based CBT that compare to traditional face-to-face therapy. Additionally, because the authors did not have any form of control group, there is absolutely no way to determine the degree to which the internet-based treatment had any impact on BN symptoms. For all we know, the meager improvements were better accounted for by the mere passage of time.
Are these results consistent across other studies using adult populations? Not necessarily. Studies using adult samples have produced somewhat better results. For instance, Fernandez-Aranda and colleagues (2009) found that, while only 15% of participants receiving internet-based CBT for BN abstained from binge eating and purging after one month, the overall rates of abstinence for binge eating and purging after three months were 55% and 73% respectively. Similarly, Ljotsson and colleagues (2007) found a 64% decrease in objective binge episodes for individuals who received a copy of the book Overcoming Binge Eating
as well as email support and access to a moderated message board.
So, what do we know overall about the use of CBT to treatment BN? First of all, face-to-face CBT is an effective treatment that tends to produce strong results. Additionally, there is compelling evidence that CBT can be effectively administered over the phone and, for adults, the internet may also be a successful medium for treatment delivery. The Pretorius et al. (2009) study, however, indicates that internet-based CBT is not yet an effective method for the treatment of BN in adolescents. Alterations to the protocol need to be made in order to ensure a greater level of participation by the client as well as a significantly greater level of symptom reduction. So, if an adolescent is showing signs of BN, the best course of action remains seeking face-to-face help from a therapist trained in empirically supported treatments. If you would like to find such resources in your local area, please consult our EST clinics page.
If you would like to learn more about empirically supported treatments for bulimia, we recommend the following resources, all of which are available through our online store:
Overcoming Binge Eating
- Christopher Fairburn, DM, FMedSci, FRCPsych
Cognitive Behavior Therapy and Eating Disorders
- Christopher Fairburn, DM, FMedSci, FRCPsych
Crave: Why You Binge Eat and How to Stop
- Cynthia Bulik, Ph.D.
Dialectical Behavior Therapy for Binge Eating and Bulimia
- Debra Safer, M.D.
Mike Anestis is a doctoral candidate in the clinical psychology department at Florida State University.
Subscribe in a reader
