Early results were promising, but dropouts were a problem.

Response to treatment for BN is modest, but an important factor is that relapse is common (Psychosom Med. 2011; 73:270). To combat this, a group in Germany recently designed a 9-month manualized CBT Web-based aftercare program for women with BN (J Med Internet Res 2017; 19:e321, published online before print). This is one of the first studies designed to evaluate the efficacy of programs targeting maintenance or improvement of treatment gains achieved during inpatient treatment for BN.

Web-based interventions have several advantages over face-to-face interventions, according to Dr. Corinna Jacobi and her colleagues. Patients’ access to the Internet reduces barriers such as cost, availability of services, waiting time, transportation challenges, and stigma about having an eating disorder. The increased anonymity may also encourage individuals to seek help.

Dr. Jacobi and her fellow researchers conducted a randomized controlled trial of 253 women with DSM-IV diagnoses of BN. Over more than 4 years, patients were screened and recruited from 13 hospitals throughout Germany that offer inpatient treatment for eating disorders. Patients were eligible for the study if they were at least 17 years old and had reduced their binge-eating and compensatory behaviors by at least 50% compared to their levels at hospital admission. Participants were then randomized to treatment with the Internet program, “IN@,” or to treatment as usual.

The Web-based intervention, IN@, included information on eating behaviors and core bulimic symptoms, and advice about healthy exercise, body image, and self-esteem. In addition, the IN@ program included a monitoring log for bulimic symptoms and a personal diary. Three clinical psychologists specially trained in behavior therapy for EDs also provided individualized email feedback and up to 9 monthly real-time individual personal chats of about 1 hour per participant. Women in the treatment-as-usual group were also assessed at all follow-up points, and had the option to take part in any outpatient treatment programs offered by the individual hospitals. Assessments were made at admission, at discharge from the hospital, 9 months after admission, 9 months after the intervention, and 9 months after this. Because the study hospitals were scattered throughout Germany, and the patients returned home to different parts of the country, all interview assessments were done by telephone by trained assessors blinded to patient group assignments.

The authors reported that 15% of the women in the intervention program never logged onto the Internet site after being discharged. The remaining 107 women accessed at least half of the intervention content. Forty-seven women (37%) participated in at least one live chat.

Symptoms and program adherence at study’s end

In general, results were numerically better in the IN@ group, but many othercomparisons were not statistically significant. Binge-eating episodes increased significantly in both groups after discharge from the hospital. However, at the post-intervention follow-up, binge-eating episodes were 27% lower in the intervention group than in the treatment-as-usual group. At follow-up there was no difference between the two groups. At the post-intervention assessment, about 1 in 5 patients reported abstinence from binge eating and lack of compensatory behaviors during the prior 2 months, but the difference between the treatment groups was not significant.

As for the incidence of vomiting at the post-intervention point, it was 46% lower in the intervention group, a statistically significant difference. At follow-up, the frequency of episodes of all compensatory behaviors was 41% lower but the difference was no longer significant. The authors note that even though the intervention did not significantly affect abstinence, the results may have important clinical implications, namely that rates of vomiting in the intervention group were almost half of those among the treatment-as-usual group.

What could improve adherence to a Web-based CBT program? The authors suggest that future programs be directed at adapting interventions to help increase adherence and reduce dropouts because use of the website and live chats was modest.