Relapse Prevention via Videoconference for Anorexia Nervosa - Findings from the RESTART Pilot Study

Background: Anorexia nervosa (AN) is characterized by high rates of relapse and chronicity [1]. Inpatient care is the treatment of choice for patients with severe AN and mostly results in weight gain [2,3]. However, up to half of weight-restored inpatients relapse within the first year after discharge [4]. A subsequent outpatient therapy to ensure weight gain maintenance is recommended [2]; however, in some health care systems and areas, the transition to outpatient therapy has been identified as a critical point in AN care due to long waiting times [5,6]. Hence, a strong need for innovative intervention strategies to ensure postinpatient continuity of care and to prevent relapse in AN has been emphasized [2,3,5,6,7]. Technology-enhanced interventions have the potential for targeting poor continuity of care, through bridging different therapy settings, improving access to evidence-based treatments and shortening waiting times [8]. In the present phase II pilot study, we used videoconferencing as a tool for delivering a relapse prevention intervention in AN. We provided a manualized outpatient relapse prevention programme which is based on the cognitive-interpersonal Maudsley Model of Anorexia Nervosa Treatment [9]. This intervention was offered to adult AN patients who completed inpatient or day patient care. We conducted a pilot evaluation of this novel intervention strategy for AN patients by investigating (a) the subjective need for a relapse prevention intervention by AN patients at discharge from inpatient or day patient care, (b) the feasibility and (c) the acceptance of the intervention and the delivery via videoconference. Taking an exploratory perspective, we looked at weight change, relapse rates, eating disorder pathology and the transition to outpatient care. Methods: Eligible patients fulfilled a diagnosis of full-syndrome or atypical AN according to ICD-10 [10] at admission to acute treatment and had completed inpatient/day patient treatment. Applying DSM-5 criteria for AN would lead to a comparable patient sample. Exclusion criteria comprised a severe somatic disorder, current substance dependence, psychotic disorder, suicidal crisis and pregnancy. The relapse prevention programme is based on the Maudsley Model of Anorexia Nervosa Treatment by Schmidt et al. [9]. Patients received 10 sessions of this intervention over 4 months. The first and last sessions took place face to face, the other 8 sessions were delivered via videoconference. Patients were encouraged to arrange subsequent outpatient therapy during the relapse prevention. Five therapists who were at least in their third year of training delivered the intervention. They received regular supervision. Diagnostic assessments were conducted before and after treatment and comprised assessment of body weight and height, SCID-I and Eating Disorder Examination interviews and self-report instruments assessing eating disorder pathology and general psychopathology. Patients and therapists were asked to fill in a questionnaire assessing their views on the treatment programme. We used a videoconference system by Cisco (San José, Calif., USA; www.cisco.com) which ensures high data safety by encrypted data transfer. Patients received a laptop with webcam, microphone and specific videoconference software. They were introduced into handling of the laptop and received technical support. As in our recent randomized controlled trial on outpatient psychotherapy in AN [11], we took steps to ensure patient safety, including regular visits to their general practitioner and the possibility of a short-term inpatient crisis intervention. We developed an emergency plan in case of patients' unavailability at scheduled videoconference sessions, including patients' contact details and contact details of close others and the general practitioner. This study was approved by the Ethics Committee of the Medical Faculty Tübingen (518/2011BO2). All participants gave their written informed consent.