Supporting someone with an eating disorder: what the latest research says

A 2018 review of 16 studies estimated that 31% of people treated for anorexia relapsed, with the highest risk in the first year after discharge [1]. This does not mean risk ends after that year. An earlier review of 27 studies found estimates from 9% to 52%, partly because relapse definitions and follow-up periods differed [2]. These figures describe study groups, not what will happen to one person. The end of treatment is a time to keep support going, even when things seem settled.

In a US clinical cohort, researchers re-interviewed 77% of 228 surviving participants about 22 years after they entered the study, and estimated missing outcomes. About 63% of women with anorexia and 68% with bulimia met its recovery criteria. At nine years, the figures had been 31% and 68% [3]. The repeated bulimia percentage does not mean the same women stayed continuously well.

A Swedish community study of 51 people who developed anorexia as teenagers found 64% met its full recovery criteria at 30 years, while 19% had an eating disorder [4]. Some had never received psychiatric treatment; this was not simply a follow-up of hospital patients. These are small, older cohorts using specific definitions of recovery. They show that recovery can happen after many years, without setting a limit on anyone's chances.

A 2023 review of 35 studies linked relapse with factors including more intensive care, other mental health problems and more severe eating disorder symptoms. More complete recovery by the end of treatment and greater motivation to change were among the factors linked with less relapse [5]. These associations do not show that intensive treatment causes relapse, or that recovery depends on willpower.

A six-year study of 117 women with bulimia or related eating disorders, recruited through a study of personality disorders, estimated a 43% chance of relapse after remission. Stress at work or in social life was associated with relapse [6]. That selected group cannot represent everyone with an eating disorder. If a stressful change is coming, ask what extra support the person would find useful.

NICE recommends planning discharge and community care from admission, involving family or carers where appropriate. Within a month, the team should review whether to continue inpatient care or move to a less intensive setting; this is not a deadline to leave hospital [7].

A German trial involved 172 girls aged 11 - 18 with anorexia on their first admission. After three weeks in hospital, they were allocated to continued inpatient care or an intensive day programme. The day programme was no worse on the trial's main measure, BMI at 12 months [8]. This does not establish that every outcome or every patient's needs are the same, or that ordinary outpatient care is equivalent. Ask what treatment and follow-up will be available after discharge, and what help family or friends are being asked to provide.

The ECHO trial allocated 178 patients with severe anorexia and 268 carers across 15 UK inpatient units to usual care or usual care plus a book, DVDs and carer coaching. It found no clear improvement in its main outcomes: time to patient relapse and carer distress at 12 months. There were small benefits in patient symptoms and quality of life at six months, and fewer hospital days during months 7 - 12 [9]. These secondary findings are encouraging but do not establish that coaching prevents relapse.

TRIANGLE allocated 371 patient-and-carer pairs across 31 UK units to usual aftercare or aftercare plus ECHOMANTRA, an online programme with videos and group chats. It found no clear benefit for its main outcome, patient distress at 12 months, and costs were higher. Only about one in five pairs met the planned minimum forum attendance [10]. This does not show that all online support is ineffective. If coaching is offered, ask what it involves, whether it suits you both and what personal follow-up is planned.

NICE recommends considering eating-disorder-focused family therapy for children and young people with anorexia or bulimia. In anorexia treatment, parents may initially take a central role in helping with eating, with later phases supporting age-appropriate independence and planning for treatment ending [7]. This is not a general instruction for partners or friends to take control of an adult's eating.

A Cochrane review found low-quality evidence that family therapy might improve anorexia remission compared with usual care in the short term. There was insufficient evidence to establish an advantage over other psychological therapies [11]. Discuss roles, consent and the pace of change with the person and their therapist. Ask how the people supporting them can contribute to the plan after treatment.

A 2018 review linked parental criticism with poorer treatment outcomes, particularly in anorexia [12]. A 2023 study of 203 adolescents with anorexia in intensive treatment found that criticism did not predict outcomes at discharge [13]. Different settings and measures may help explain mixed findings; neither study means families caused the illness.

A 2024 review of 36 studies examined accommodation: changing routines or going along with the eating disorder's demands. This was associated with carer distress and often reduced during interventions, but links with the patient's outcome were mixed and no intervention was clearly best [14]. If meals or conversations have become difficult, ask the team for practical help with communication and boundaries. Family and friends need support themselves, not responsibility for controlling every outcome.

The Royal College of Psychiatrists' MEED guidance warns that appearance and normal blood tests can give false reassurance about serious medical risk. Information from parents and carers matters to assessment [15]. You do not have to judge physical safety yourself.

Beat suggests a written plan covering warning signs, triggers, helpful actions and emergency contacts, with input from people close to the person if they want it [16]. NICE also recommends assessing carers' own support needs, including emergency planning [7]. Agree what to do if eating, activity, mood or physical health changes, and which clinician to contact. Friends can contribute what they notice even if they do not live together. Ask the team how any home monitoring should work; it does not replace medical assessment.

FREED is an early-intervention pathway for 16 - 25-year-olds with an eating disorder lasting less than three years. An evaluation across four English services compared 278 people offered FREED with 224 previous patients. Waiting times were shorter and treatment uptake was higher [17]. It was not randomised, so changes in services or the people referred could also contribute to the results.

The Maudsley Biomedical Research Centre describes a national rollout in England [18]. This does not guarantee that every local service offers the full pathway. Ask the GP or eating disorder service about local eligibility and referral. People outside FREED's criteria can still seek eating disorder assessment and treatment.

Next check due: 3 November 2026. Details to check: Age limits and local availability.

Beat’s free eight-week workshops for family and friends (formerly Developing Dolphins 2) run on Zoom, led by an eating disorder clinician, with an online workbook. It's for carers aged 18 or over, including siblings, friends and partners, supporting someone aged 10 or over who's had an eating disorder for less than five years. No formal diagnosis is needed. Beat currently lists funded self-referral in Essex, Scotland, Sussex, Wales and York, and service-referral routes in Kent, Suffolk, Sussex and Gloucestershire; the service will confirm eligibility. Elsewhere, Beat points carers to its online Eating Disorder Support Hub for Families and Friends [19].

Next check due: 3 November 2026. Details to check: Funded areas and eligibility.

The Charlie Waller Trust runs free online workshops for parents and carers: five two-hour sessions covering motivational language, responding to difficult behaviour, coaching someone to make their own changes, and planning for a possible relapse [20]. An Italian study of New Maudsley workshops found lower carer anxiety and accommodation afterwards, but only 189 of 447 participants completed both assessments and there was no comparison group [21]. It was not a trial of the Charlie Waller service and cannot establish that workshops caused the changes.

Next check due: 3 November 2026. Details to check: Workshop dates and booking.