Bulimia nervosa (BN) is an eating disorder characterized by binge eating, purging or fasting, and concerns associated with body shape and weight [1]. Patients with BN often have a distorted body image despite being at standard levels with regard to shape and weight [2]. Consequently, they perform frequent purging or fasting or excessive exercise on the basis of a subjectively abnormal perception of their body shape and weight [3]. Body image distortion involves perceptual disturbances and body dissatisfaction [4]. Repeated episodes of binge eating and inappropriate compensatory behaviors can cause serious physical and mental health concerns. Vomiting after overeating may result in thickened skin on the knuckles, broken teeth, and thyroid dysfunction due to problems with metabolic rate and caloric intake [5,6]. In terms of mental health, patients with BN frequently show low self-esteem, self-harming behaviors, suicidal tendencies, depression, anxiety disorders, sleep disorders, and fatigue or exhaustion [7-12].

Epidemiological studies confirm that eating disorders are highly prevalent, particularly among women: 2.58% of women in Western countries [13]. The overall observed prevalence of eating disorders was 3.5% between 2000 and 2006 and increased to 7.8% between 2013 and 2018 [14]. A recent meta-analysis estimated that the lifetime and 12-month prevalence of eating disorders was 0.91% and 0.43%, respectively, while the lifetime prevalence of BN was 0.63% [13]. Face-to-face cognitive behavioral therapy (CBT) has been shown to be effective for eating disorders [15], but poor access to treatment remains a problem [16]. In fact, only 19%-36% of people with eating disorders have access to treatment within 1 year [17-19], and 35%-40% of these patients receive standard treatment for eating disorders [20,21]. Furthermore, treatment for eating disorders is sought an average of 5-15 years after the onset of the disorder [16,22]. Access to these appropriate treatments is often hampered by physical barriers, lack of practitioners, and stigma. These problems are particularly associated with face-to-face delivery. Web-mediated interventions may be effective in initiating appropriate early treatment [23,24]. CBT via the internet can dramatically improve access to treatment, especially in developed countries, in which the internet infrastructure is well-developed and information communication devices are widely used.

Enhanced CBT has been shown to be effective in treating eating disorders, including BN. Enhanced CBT encourages the early establishment of healthy and safe eating habits and facilitates behavioral changes and cognitive modification through psychoeducation and self-monitoring. In addition, accumulating evidence from many clinical studies is being integrated to establish CBT techniques that have shown promise for the treatment of eating disorders. In a network analysis, fear of weight gain was suggested to be central to the psychopathology of BN, and hypersensitivity to physical sensations was shown to bridge BN with anxiety or depression [25]. Based on the evidence, this network meta-analysis argued that focusing on the fear of weight gain during exposure therapy and interventions focused on interoceptive exposure is effective in treating BN. However, several other CBTs may also be effective in treating BN, including attentional bias correction training to correct the direction of excessive attention to the body, weight, and food [26], relaxation, and mindfulness to reduce physiological responses such as anxiety and tension [27], traumatic memory care for experiences of being criticized for one’s body type and appearance [28], and impulse control, such as cue exposure [29]. A therapist’s manual for CBT of BN that systematically summarizes these CBT components has been published by Hamatani and Matsumoto [30]. Our research group recently developed a web-based program based on this CBT therapist’s manual by checking the cultural adaptability in Japan [31].

As mentioned earlier, we developed an internet-based cognitive behavioral therapy (ICBT) program for BN that is adapted to Japanese culture and based on Japanese food culture [30,31]. This paper describes the study protocol for a randomized controlled trial (RCT) involving female patients with BN and is designed to evaluate its clinical effects. We will include an intervention group receiving ICBT in addition to treatment as usual (TAU) and a control group receiving TAU alone.

This study is designed as a multicenter, prospective, randomized, assessor-blinded clinical trial. The RCT and follow-up study are planned from August 2022 to October 2026 (UMIN000048732). It will be conducted at the following 7 institutions (6 university hospitals and a national medical center) in Japan: University of Fukui Hospital, Kagoshima University Hospital, Chiba University Hospital, Tokushima University Hospital, Dokkyo Medical University, Tohoku University Hospital, and National Center for Neurology and Psychiatry.

Eligible participants will be women aged 13-65 years (1) diagnosed with BN according to the Diagnostic and Statistical Manual of Mental Disorders criteria during a clinical interview [1]; (2) having a BMI over 17.5 kg/m²; (3) using computers, tablets, smartphones, etc, on a daily basis, with access to the internet and the minimum necessary information and communications technology skills; and (4) with no history of CBT in the last 2 years. Exclusion criteria include (1) serious mental disorders such as organic brain disorders, psychotic disorders, and drug dependence; (2) imminent risk of suicide; (3) repeated engagement in antisocial behavior; (4) serious progressive physical disease; and (5) difficulty in exposure to feared objects due to severe stress reactions or dissociation symptoms due to acute stress disorder or posttraumatic stress disorder. The purpose of the study is to investigate the effects of an ICBT program that has been adapted based on the culture of Japanese women. Men and women differ in their caloric needs, and most studies regarding eating disorders include female patients. This intervention program is based on the findings of previous studies; therefore, only women will be recruited. In addition, information and communication technology literacy varies considerably by patient age in Japan, with older adults having lower information and communication technology literacy. The intervention in this study is mainly self-help CBT using a web-based platform. Therefore, patients aged 65 years and older will be excluded from the study.

From August 2022 to July 2024, we will recruit 60 out-participants with female BN through posters, flyers, web or app advertisements (Google Ads, Twitter, and Facebook), newspaper advertisements, etc, posted at medical institutions throughout Japan. Since all participants will continue to be treated, they will be required to obtain clearance from their primary care physician prior to their study enrollment. The participants pay for TAU, but guided ICBT is free. Participants will be rewarded US $82.51 after intervention or waiting completion, regardless of group allocation. These include the amount of time patients spent on the trial and the difficulty of recruiting patients. To participate in this study, some hospitals will require referrals, which may cost the participant money.

Participants will be directed to a website for this study by the various media mentioned earlier. There they will access the study website to apply for enrollment in the study. The clinical trial office at the University of Fukui that receives this application will coordinate the informed consent and eligibility test. Written informed consent will be solicited after the participants are provided research briefing either face-to-face or in a video call. Similarly, in the case of a minor, written informed consent from a legal representative is required. Next, they will complete a screening survey and a structured interview using Mini International Neuropsychiatric Interview version 5 [32,33]. Finally, they will be asked to complete a baseline data questionnaire on the web. The decision regarding the participant’s exclusion or inclusion in the study will be made after they have completed the questionnaire. See Figure 1 for the flow of participants in this RCT.

The required sample size was calculated using statistical analysis software (G*Power, version 3.1; Heinrich-Heine-Universitaet Duesseldorf). An effect size of Cohen d=0.90 from a previous study was used as a reference [64]. The significance level was set at P values <.05 for 2-sided tests. The power was set to 80%. The required sample size (assigned number of participants) will be 60 when considering a 30% noncompletion rate based on a previous study [65]. Completers in this study are defined as those who complete at least 80% of the module [65]. In addition, dropouts in this study are defined as those who drop out by not providing postevaluation data or those who discontinue their participation due to adverse events.

Participants confirmed to be eligible and enrolled will be randomized in a 1:1 ratio to either the intervention (TAU added guided ICBT) or TAU alone groups by the truncated binomial design using UMIN’s (University Hospital Medical Information Network) computer program (UMIN Center). The truncated binomial design is that complete randomization will be performed until the number of randomizations to any group reaches n/2, after which all subjects are allocated to the other group to prevent an imbalance in the number of people.

Statistical analysis will be performed using SPSS Statistics software (version 29; IBM Corp). Statistical analysis will be performed in accordance with the CONSORT (Consolidated Standards of Reporting Trials) guidelines and based on the intention-to-treat principle (Multimedia Appendix 1). Missing data will be handled by multiple imputations. Unpaired t test and Fisher exact test will be used to investigate the difference between the 2 groups at baseline. To compare treatment effects, we will use analysis of covariance for primary and all secondary outcomes. As a covariate, we plan to include a scale that is significantly different between the 2 groups at baseline. We also will perform a per-protocol analysis by excluding patients who deviate from the protocol. Analysis of covariance will also be performed on the follow-up data. However, if there are many missing values, we would consider generalized linear mixed models. All P values are 2-tailed, and P values <.05 will be considered statistically significant.

The clinical trial protocol was approved by the Research Ethics Committee of the University of Fukui on August 15, 2022 (20220054), and was registered in the UMIN Domestic Clinical Trial Database (UMIN000048732). Each participant will then be informed that all participants will receive TAU from the general practitioner and that half of the participants will receive ICBT in addition to TAU. All adverse events will be reported, and serious adverse events will be reported immediately to the institutional review board. An adverse event is defined as any symptom or illness occurring during a clinical trial, whether related to the ICBT program or not. The results of the trial will be published in the appropriate journal, regardless of the outcomes. The trial will be implemented and reported in accordance with the CONSORT recommendations.

This paper presents the research protocol for the first RCT designed to investigate the effectiveness of therapist-guided ICBT for female Japanese patients with BN. Notably, a recent RCT demonstrated the efficacy of ICBT for binge eating disorder and other specified feeding or eating disorders [66]. However, although that study, conducted in the Netherlands, showed promising results for guided ICBT in these eating disorders, it involved participants with varying BMI of 19.5 or 40.0 kg/m² and did not include patients specifically diagnosed with BN. In contrast, this study protocol uses specific eligibility criteria, focusing solely on individuals with BN as the primary diagnosis while excluding other eating disorders, such as binge eating disorder. Consequently, the patient cohort to be included in our RCT is anticipated to be homogenous, thereby bolstering confidence in the treatment’s effectiveness. Meanwhile, a systematic review and meta-analysis examining e-therapy’s impact on eating disorders reported inconclusive evidence regarding the effect on overeating frequency in patients with BN [67]. Although this meta-analysis found improvements in various aspects, such as binge eating, vomiting, and laxative misuse, and in the discontinuation rate of overeating symptoms at the intervention’s conclusion, the estimates were imprecise due to the heterogeneity in the quality of studies included in the analysis. Therefore, the authors concluded that e-therapy, including guided ICBT, could be effective for the management of eating disorders; however, further research is needed. A recent single-arm ICBT study targeting patients with eating disorders, including BN, reported positive outcomes [65]. In evidence-based medicine, individual RCT holds the highest level of evidence among single clinical trials [68]. Nevertheless, the outcomes of single-arm clinical trials lacking a control group should be interpreted with caution. In this context, our RCT may expand the understanding of the efficacy of guided ICBT for BN. To our knowledge, even recent RCTs have not provided effect estimates for patients with BN alone, as they also included other eating disorders [69,70]. Therefore, this study, by adhering closely to the RCT protocol described herein, will be well-positioned to offer valuable insights into the efficacy of guided ICBT on the frequency of overeating and purging along with any secondary symptoms in patients with BN. Furthermore, to the best of our knowledge, this is the first study designed to examine the effects of ICBT on behavioral changes related to eating styles in Japan. This is especially significant given Japan’s distinct food culture and medical system that sets it apart from Western countries. The investigation into the effects of ICBT for BN within the context of Japan carries particular importance, as it navigates cultural nuances and differences in the medical system. Therefore, a multicenter study design is more likely to reduce bias and yield findings with broader generalizability.

One limitation of the study is that it would not allow elucidation of the specific effects of the ICBT program because it does not use a psychological placebo group to control for nonspecific factors.

This RCT aims to validate and prove evidence for the efficacy of a Japanese culture–adapted ICBT program for female patients with BN. This will allow discussion of efficacy beyond the preliminary results obtained to date. The low availability of CBT is an international problem, and the implementation rate of CBT in Japanese psychiatric clinics is extremely low at 6.2% [71]. If positive results are obtained from this RCT, more patients with BN could receive early treatment, which could lead to early improvement in eating disorders that tend to be prolonged and chronic.