Treatment Delivery Strategies for Eating Disorders

Paulo P.P. Machado; Tânia F. Rodrigues


Curr Opin Psychiatry. 2019;32(6):498-503. 

In This Article

Internet Delivered Interventions

Recent data from the statistical office of the European Union (EUROSTAT, showed that the percentage of individuals, across 28 European Union Countries, using mobile devices to access the internet on the move has risen from 36% in 2012 to 69% in 2018. The average number of individuals aged 16–24 years old accessing the internet, using mobile devices, rose to 92%, in 2018. It is not surprising that the internet and mobile devices have become the most recent media to be explored for self-help, and guided self-help interventions.

Aardoom and colleagues[10] reviewed 21 studies addressing internet-based treatment of eating disorders. Two of the studies focused on the use of e-mail as an adjunct to therapy, three addressed the delivery of CBT over e-mail, one study investigated a cognitive behavioral self-help program with internet-based guidance and another study investigated an open trial of internet-based unguided self-help. The remaining 14 reported on six different internet-based CBT programs, but only 12 studies focused on the effectiveness of the intervention. In their review, the authors concluded that internet-based treatment was more effective for BED than for bulimia nervosa, more effective for individuals with less comorbid psychopathology, and with an eating disorder characterized by binge eating as opposed to restriction. In addition, initial face-to-face contact and assessment seemed to improve participants compliance, and higher levels of compliance were related to more symptomatic improvement.

In a more recent review,[11] the authors identified three additional large randomized controlled trials (RCT) investigating internet-based CBT, consisting of a structured CBT program, and weekly therapist support via e-mail. One study found internet-based CBT to be more effective than waitlist control (WLC) for individuals with bulimia nervosa, BED, and OSFED. The other two, investigated the efficacy of these interventions amongst individuals with bulimia nervosa, and found it to be superior to WLC. However, they identified a lack of studies comparing face-to-face and internet-based treatment.

Since then, three RCT of internet-based interventions were published. The first one, investigated the effectiveness of an internet-based intervention for individuals with eating disorder symptoms, called 'Featback'.[12] In addition, the added value of different intensities of therapist support was investigated. Participants were randomized to: Featback, consisting of psychoeducation and a fully automated self-monitoring and feedback system, Featback supplemented with low-intensity (weekly) digital therapist support, Featback supplemented with high-intensity (three times a week) digital therapist support, and a WLC. The three Featback conditions were superior to a WLC in reducing bulimic psychopathology. The fully automated Internet-based self-monitoring and feedback intervention, Featback, was effective in reducing eating disorder and comorbid psychopathology. Supplemental therapist-support enhanced satisfaction with the intervention but did not increase its effectiveness.

The second, CBT4BN,[13] compared the use of an internet-based manualized version of CBT group therapy for bulimia nervosa, conducted via a therapist moderated chat group, to the same treatment manual conducted via a traditional face-to-face group therapy. The study, a two-site, randomized, controlled noninferiority trial, showed that the internet-based version of the treatment was inferior to the face-to-face version in producing abstinence at end-of-treatment, but by the 12-month follow-up internet-based treatment was noninferior on most measures.

The third, the Internet and binge eating disorder (INTERBED) study[14] compared the efficacy of Internet-based GSH with face-to-face CBT for individuals with BED. The study showed that face-to-face CBT was more efficacious than internet-based GSH in reducing binges, promoting abstinence from binge eating, and in reducing eating-related psychopathology at the end-of-treatment (4 month) and at 6-month follow-up. However, exploratory analysis in a smaller sample revealed that these differences disappeared in the long run (i.e., at 1.5-year follow-up).

An additional RCT,[15] soon to publish its results, found that face-to-face and e-mail GSH were both superior to waitlist in reducing eating disorder psychopathology. There were no differences between the two active treatments (although they weren't powered enough for this), and attrition (drop-out) was higher in the e-mail condition (although there might have been nontreatment-related reasons for this) (Personal communication, May 24, 2019).

Two of the previously reported trials[13,14] were subsequently subject of cost-effective analysis. The first concurrent study[16] evaluated the cost-effectiveness of Internet-based cognitive-behavioral therapy for bulimia nervosa (CBT-bulimia nervosa) compared to face-to-face delivery of CBT-bulimia nervosa, as a secondary analysis of a RCT.[13] The authors concluded that face-to-face and Internet-based CBT-bulimia nervosa had similar cost-effectiveness, measured as the cost per abstinent patient. Cost-utility, measured as the cost per gain in Quality Adjusted Life Years was also comparable across intervention arms. Patient out-of-pocket costs were significantly lower in internet-based treatment because of the absence of travel-related costs.

The second study[17] determined the cost-effectiveness of individual face-to-face CBT compared to therapist-guided internet-based self-help in overweight or obese adults with BED (INTERBED trial[14]). The authors found no clear evidence for one of the treatments being more cost-effective. Face-to-face CBT tended to be more effective but also more costly; and concluded that if societal willingness to pay for an additional binge free day is low, then results suggest that internet-based GSH should be adopted.

Although less research exists on the use of self-help for anorexia nervosa (AN), Albano and colleagues,[18] recently performed a systematic review on interventions targeting either the individual affected by AN or their carers, and a meta-analysis of studies using randomized controlled designs evaluating the outcome of such interventions. The review included 22 articles, 15 of which discussed the use of self-help or GSH for patients and 7 for carers. The interventions were diverse in term of theoretical models, used different media (books and digital materials), and were guided by individuals with a variable experience and expertise (e.g., former patients, graduate students, or clinically trained professionals). The control group often consisted of patients on WLC or receiving standard treatment as usual. Seven of these studies used a randomized-controlled design and were included in the meta-analyses. The results of the meta-analysis indicated a reduced drop-out rate at end-of-treatment in the guided self-help/self-help group compared to the control group. However, guided self-help/self-help interventions did not produce greater clinical improvements in BMI, depression, anxiety or quality-of-life when compared to a control group. The authors are currently conducting two large RCTs[19,20] aimed at testing GSH for patients with AN.

In a timely and very recent to be published article, Barakat and colleagues[21] reviewed the literature and conducted a meta-analysis to determine which components of internet-based interventions were associated with higher retention and improved outcome in eating disorder treatment. Each of the 23 studies reviewed was coded on four dimensions: channel used (e.g., text, audio, and video); degree of interactivity with the program (e.g., self-monitoring and exercises); level of automated feedback (e.g., reminders, personalized feedback, and reinforcement messages); and, device used (e.g., smartphone, tablet, and computer). The authors concluded that e-Therapies, especially those aimed at binge eating and bulimic type disorders, produced significant decreases in eating disorder symptomatology, with small to medium effect sizes. Most importantly, results suggest that two technical characteristics of the intervention, multimedia and feedback, were moderators of outcome. In summary, the more the range of multimedia channels the greater symptom improvement; and studies which employed more frequent and personalized feedback achieved greater results at end-of-treatment when compared to the control condition.