• Anorexia nervosa;
  • eating disorder;
  • hospitalisation;
  • outcome;
  • adolescence


  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References


A CAMHS Eating Disorder (ED) service is presented together with cases referred and treated within a 3-year period. An outcome evaluation was conducted for restrictive eating disorders.


Case files were analysed to investigate service throughput as well as clinical outcomes for discharged cases (n = 39).


Highly positive End of Treatment (EOT) outcomes were achieved for both anorexia nervosa (AN) and Eating Disorders Not Otherwise Specified (EDNOS). Morgan Russell scores, drop-out and admission rates compared favourably with previous trials. User satisfaction ratings were high.


The study offers insight into real-life clinical practice and as outpatient treatment is usually much less expensive than inpatient treatment, this approach has the potential to be cost-effective.

Key Practitioner Message
  • Only about 70% of adolescents with AN recover. Outpatient family-based treatment is the most effective reported treatment
  • Despite the evidence, many clinicians in Europe report using individual therapy only. There are few studies on ‘real-world’ treatment, or on service development
  • This study evaluated a CAMHS ED service that used family-based treatment. Service development, operational processes and clinical outcomes were reported for restrictive ED
  • Outcomes showed better results than published trials, with a yearly admission rate for AN of 14%, low drop-out rates and a mean treatment time of about 18 months. 100% had good or intermediate Morgan Russell outcomes. A high level of client satisfaction was expressed


  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Evidence on treatment effects for AN has accumulated with findings that family-based intervention is the treatment of choice for adolescent AN (Eisler, 2005; Lock et al., 2010; National Institute for Clinical Excellence, 2004). Studies suggest that inpatient treatment tends to be ineffective in general (Crisp et al., 1991) and specifically for adolescents, with outpatient services achieving superior outcomes (Eisler, 2005; Gowers, Weetman, Shore, Hossain & Elvins, 2000).

Studies in the adult population suggest that only 50% of patients with AN recover (Steinhausen, 2002), whereas in adolescents, approximately 70% recover in the longer term (Herpertz-Dahlmann et al., 2001; Strober, Freeman & Morrell, 1997). The reason for this is unknown, but may be due to the support provided by families. Time scale for recovery was examined in the Treatment Outcome for Child and Adolescent Anorexia Nervosa (TOuCAN) trial by Gowers et al. (2007), who reported that across settings, fewer than one in five fully recovered within a year and a third by 2 years. Evidence for the effectiveness of AN treatment has reported EOT results in categories of good, intermediate and poor (Morgan & Russell, 1975). In Eisler's review of outcome studies (2005), good outcomes ranged from 9% to 60%, intermediate from 6% to 50% and poor from 10% to 31%. More recently, Lock et al. (2010) reported full remission after 12 months of treatment in 42% of their family-based treatment group, and partial remission (intermediate or better) in 89%. Drop-out rate was 14% at EOT.

Hospital admission rates for adolescents with AN vary between 10% and 40% (Eisler, 2005; Lock et al., 2010), with 32% reported in TOuCAN for generic CAMHS during 1 year of treatment. Research has highlighted outpatient treatment as about a tenth of the cost of inpatient treatment (Byford et al., 2007; Eisler, 2005).

Limited information exists to indicate what actually happens in ‘real-world’ clinical care (Richard, 2005), and various authors (e.g. Baker et al., 2010) have raised concerns that important research findings are not transferring into clinical practice.

Gowers et al. (2002) examined service provision for adolescents across 12 countries and found a great variation in intervention methods and length of stay varying at least four-fold between services.

In their survey of clinical practice in outpatient CAMHS, Kleve and Savile (2011) found that most UK clinicians ensured regular weighing of clients (78.7%), 18.1% used individual therapy only, 14.5% family therapy only and 30% a combination of the two. In a Norwegian survey, 79.6% reported individual therapy to be the most common intervention for AN for adults and adolescents (Petterson & Rosenvinge, 2006).

There is also a surprising lack of practice-based evidence and knowledge base to guide the development of effective ED treatment services (Gowers et al., 2007; Richard, 2005). In the absence of a clear strategy for service delivery, the organisation, delivery and monitoring of care provision is mostly left to individual clinicians and teams. Increasingly it is acknowledged that to improve our understanding and ability to help clients, practising clinicians and researchers need to share perspectives.

Among the few studies concerning consumer satisfaction with ED treatment, a perception of limited professional expertise within generic CAMHS has been highlighted (Newton, 2001). The need for clients to be motivated and practitioners to challenge in nonconfrontational ways, treating clients as individuals and to consider their psychosocial and physical functioning was stressed by Tierney (2008).

The aim of this study was to provide a practice-based perspective, describing the organisation, outcome evaluation and user perspective of an eating disorder service based in a community CAMH service.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

North Somerset CAMHS ED service

The service has been established without any targeted funding. Covering a population of 205,000 and coordinated by a clinical psychologist, it comprises a core team including specialist nurse therapist, two clinical psychologists, psychiatrist and systemic family psychotherapist. Clinicians hold generic as well as specialist case loads. The extended team includes acute paediatricians, GPs and dieticians. A development plan has helped to guide and focus the service and an ED protocol provides consistency in management. The primary therapeutic model used is Family Based Intervention (Lock et al, 2001) for AN and EDNOS (restrictive type). Clinicians are also trained in Extended CBT (Fairburn, 1995) and Interpersonal Psychotherapy (Mufson, Noreau, Weissman & Klerman, 1993), relevant for Bulimia Nervosa (BN). Clients are assessed within 2 weeks of referral or sooner if necessary. Team members meet 6-weekly to review practice and relevant literature and for peer supervision. Local policy is to encourage early referral.


The study examined 70 referrals of eating problems during the calendar years 2007–9 and 16 cases with diagnosed ED referred in previous years, open at the start of 2007. Diagnoses were made according to DSM IV criteria.


Data were collected from case files at assessment and EOT. As a service innovation, this work did not require ethical approval. Informed consent was given to treatment. Audit sheets required completion of demographic details, family composition, family involvement in treatment, referrer, medication administered, treatment length, use of inpatient services and whether menstruation and/or vomiting were/was present. Height, weight, BMI, percentage of expected body weight for height and age (%EWH) were recorded, and after January 2007, the Global Eating Disorders Examination Questionnaire (EDE-Q) (Fairburn & Beglin, 1994) was used. To obtain user feedback, the Commission for Health Improvement Experience of Service Questionnaire (CHI ESQ) (Attride-Stirling, 2002) child and parent versions were sent to all completed cases.


Descriptive analysis was performed in relation to a range of variables. Statistical analyses on pre- and post-treatment measures, using independent variables t-testing on SPSS v 11.5 software (SPSS Inc., Chicago, IL, USA) were performed on completed cases and Morgan-Russell categories (Morgan & Russell, 1975) were assigned.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Service activity

Table 1 shows that 37 (53%) of the eating problem referrals in 2007–9 were judged to have diagnosable eating disorders and started treatment. Together with the 16 existing cases open at year end 2006, they made up the sample of 53 cases in Table 2. Drop-out rate from treatment was low, at 1.9% per year during the 3 years studied. Also low was the rate of inpatient admissions at 5.5% of all open cases with ED per year.

Table 1. Details of service throughput and inpatient admissions
Open ED files at year start161819(mean = 17.7)
All new referrals with eating problems18322070
DNA at first appointment0123
Referrals with mild eating problems615930
Diagnosed with ED (AN, EDNOS, BN) and treatment started1216937
Total ED cases worked in year (new plus open cases at year start)283428(mean = 30.0)
Dropped out during year1023
Referred on to adult services0101
Discharged following treatment9141336
Year end open files181913(mean = 16.7)
Inpatient admissions during year2215
Table 2. Diagnosis of ED cases treated
DiagnosisOpen files at beginning 20072007–9 referralsTotal cases worked during 3-year period
AN6 (38%)17 (46%)23 (43%)
EDNOS Restrictive type9 (56%)13 (35%)22 (42%)
BN0 (0%)6 (16%)6 (11%)
EDNOS Bulimic type0 (0%)1 (3%)1 (2%)
Food Avoidance Emotional Disorder (FAED)1 (6%)0 (0%)1 (2%)
Total1637 (100%)53 (100%)

Diagnosis at assessment

See table 2. From now on, description will focus only on AN and restrictive EDNOS (N = 45).

Restrictive eating disorders – Client characteristics

All referrals were from health, with 71% from GPs. Of the 45 cases worked, 42 (93%) were female. Mean age at assessment was 14 years and 8 months, range 10 years 4 months to 17 years 7 months, with EDNOS 4 months younger (14 years 6 months) than AN. Ethnicity was only recorded for 25 cases; all were White British, reflecting local trends. 36 (80%) were from an intact two-parent family. Mean BMI at assessment was 16.7, range: 13.6–21.4, SD: 1.72. Mean %EWH, calculated using UK growth charts (Child Growth Foundation, 1996) was 84.3%, range: 71–108%, SD: 8.8.

Treatment for restrictive eating disorders

Treatment did not differ between AN and EDNOS. Three cases dropped out and one was referred on to adult services. Treatment for the 41 clients who remained involved family members in 98% of cases, with 31 (76%) receiving both individual therapy and family work, nine (22%) family work only, and one (2%) individual sessions only, at her request.

Five AN clients were admitted during the evaluation period. The mean rate of admission per year of 1.7 cases represented 14% of open AN cases per year and 7% of open AN and EDNOS together.

Thirty-one clients were discharged during 2007–9. Of the 10 remaining open cases, seven were discharged in 2010, one was discharged in 2011 and two dropped out and were lost to follow-up leaving 39 with before and after data and an overall drop-out rate for all restrictive cases of 11% spread over 5 years. Of the 39 discharged cases, 17 had AN and 22 had EDNOS. Mean duration of treatment was 17.7 months, range: 3–47 months, SD: 10.8, and was similar for AN and EDNOS.

Eight discharged clients, (21%) were prescribed psychotropic medication, six with AN and two with EDNOS. Most frequently prescribed was fluoxetine (six cases) followed by amitryptaline (two cases). Six were prescribed an anti-depressant alone, one an anti-depressant and an anti-psychotic and one an anti-depressant, two anti-psychotics and a benzodiazepine. Of those who had been inpatients, 50% had been prescribed medication compared with 15% of outpatients.

Clinical outcomes for restrictive eating disorders


Of the 39 discharged cases, 29 had regular menstrual periods, five had not started or restarted, three were male and two were under 13 and had not menstruated before developing EDs.


Table 3 shows that for discharged clients, those with AN started with a lower BMI and %EWH than those with EDNOS, but put on more weight during treatment. Two-tailed t-tests confirmed highly significant changes in EWH and BMI following treatment (p < .0001).

Table 3. BMI and %EWH before and after treatment for discharged cases with restrictive eating disorders
Diagnosis N Mean initial %EWH SD Mean Final %EWH SD Difference between before and after %EWHMean Initial BMI SD Mean Final BMI SD Difference between before and after BMI
Diff in meansp-valueDiff in meansp-value
AN and EDNOS Together3985.0%8.894.6%9.49.6<.000116.71.819.01.92.4<.0001

The EDEQ was only used after 2007. Missing data due to client noncompliance, therapist omission or incomplete scripts meant that there were insufficient scores to compute differences.

Morgan Russell categories

Please see Table 4

Table 4. Morgan Russell categorical outcomes (1975)
  1. Good: Body weight maintained within 15% of average body weight and regular menstrual cycles.

  2. Intermediate: Body weight risen to within 15% of average body weight, but not constantly sustained; and/or continuing menstrual disturbances.

  3. Poor: Body weight so low that it never approached average body weight minus 15% level; Menstruation absent or virtually absent.

AN13 (77%)4 (23%)0 (0%)17
EDNOS18 (82%)4 (18%)0 (0%)22
Total31 (79%)8 (11%)0 (0%)39
User feedback

The CHI Experience of Service Questionnaires were completed by 20 young people (51%) and 22 parents (56%). Table 5 shows very high user satisfaction levels with the service.

Table 5. Experience of Service Questionnaire (CHI ESQ)
CHI ESQ questionnaire% young people answering true or partially true (N = 20)% parents answering true or partially true (N = 22)
I feel that the people who saw me/my child listened to me(17 + 1) 90%(20 + 2) 100%
It was easy to talk to the people who saw me/my child(12 + 8) 100%(20 + 2) 100%
I was treated well by the people who saw me/my child(18 + 2) 100%(20 + 2) 100%
My views and worries were taken seriously(14 + 6) 100%(20 + 2) 100%
I feel that the people here know how to help me(16 + 3) 95%(20 + 2) 100%
I have been given enough explanation about the help available here(16 + 1) 85%(20 + 2) 100%
I feel that the people who have seen me/my child are working together to help me(17 + 3) 100%(17 + 5) 100%
The facilities here are comfortable(14 + 6) 100%(15 + 7) 100%
The appointments are usually at a convenient time(13 + 6) 95%(13 + 6) 85%
It is quite easy to get to the place where the appointments are(15 + 5) 100%(16 + 5) 95%
If a friend needed similar help, I would recommend that he or she come here(17 + 1) 90%(20 + 1) 95%
Overall, the help I have received here is good(17 + 1) 90%(21 + 1) 100%

Mean percentage of young people who responded true or partially true to all questions was 95%, and for parents this figure was 98%.

Parents and young people added a range of qualitative comments to three additional questions:

  1. What was really good about your care?
  2. Was there anything you didn't like or anything that needs improving?
  3. Is there anything else you want to tell us about the service you received?

For question one, young people left nine comments on the themes of being listened to and treated with respect, and parents left 12 comments emphasising care, respect and effective interventions. For question two, young people gave four positive comments and four suggestions for improvement, while parents gave five positive comments and five suggestions for improvement. Eighteen responses were given to the last question, 12 from parents and 6 from young people. These included comments about the accessibility of the service and the gratitude they felt.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This study presents a model and outcomes of ‘real-world’ community care for adolescents affected by ED. In all, 98% of cases with restrictive ED were treated with a family-based approach – a higher figure than reported previously in Europe. Medication was used in a minority of discharged cases and was more likely to have been prescribed for hospitalised clients. Significant improvements were demonstrated: 77% of AN and 82% of EDNOS achieved good Morgan Russell outcomes and none had poor outcomes. This compares favourably with previous treatment trials and with outcomes reported for CAMHS in the TOuCAN trial, where only 19% of treated AN after 1 year and 36% after 2 years were identified as having achieved a good outcome.

The 2% drop-out rate per year was low compared with the 14% reported by Lock et al. (2010). Admission rates (14% of AN per year and 7% of AN and EDNOS together) compare favourably with the TOuCAN trial, where 32% of treated young people with AN receiving community CAMHS treatment had been admitted during the first year of intervention. Service satisfaction ratings were high both from young people and their parents, indicating that they felt listened to, cared for and satisfied with the help they received.

There may be several explanations for the positive results. The AN sample was less underweight when referred than in the TOuCAN trial, although it was similar to Lock et al. (2010). More families were intact compared with the TOuCAN trial, which may have affected parental motivation and hence outcome. Lastly, the current study used end of intervention as the final point in contrast to the planned end points, including intention-to-treat analysis, in most controlled trials. The above points still do not adequately explain the reason for superior results to previous open follow-up studies [see Eisler (2005) for an overview of these].

This study has several methodological weaknesses. The absence of control and comparison groups, a small sample size and no measures of treatment fidelity are obvious limitations. The lack of a complete set of EDEQs also reduces its power. Such limitations are, however, frequent in real-life investigations and should not detract from the fact that it still serves as a clinically relevant supplement to the limited number of real-world studies available on treatment outcomes and community services for adolescent ED. The user perspective, which has hitherto often been neglected in ED research, is an additional and important component. That evidence-based interventions were adopted and positive results achieved within a generic service (although by highly qualified and experienced clinicians) should also be of interest to commissioners. A clear service development plan, evidence -based interventions underpinned by a clear protocol, peer supervision and early referral are likely to be key elements of the success. The combination of core and extended teams has also served to build up reliable and helpful care pathways.

It is recommended that a follow-up study is conducted to identify long-term clinical outcomes. A multi-site study comparing key components of care and outcomes across different outpatient services would further inform our knowledge of existing tier 2/3 service models and variables linked to clinical and financial success.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We gratefully acknowledge Amy Savile, Jaye Caffery and Stephanie Stafford-Smith for assistance with data preparation and statistical analysis, Louise Cooper, Dr Trisha Tallis and Paula Edgington for participating in data collection and Cambridge Eating Disorder Research Consortium for providing a stimulating forum for discussion. No funding was provided for this study; it was completed as part of clinicians' and psychology assistants' routine work. The authors have declared that they have no competing or potential conflicts of interest.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Attride-Stirling, J. (2002). Development of methods to capture users' views of child and adolescent mental health services in clinical governance reviews. London: Commission for Health Improvement.
  • Baker, R., Camosso-Stefinovic, J., Gillies, C., Shaw, E.J., Cheater, F., Flottorp, S., & Robertson, N. (2010). Tailored interventions to overcome identified barriers to change: Effects on professional practice and health care outcomes Cochrane Database of Systematic Reviews 1–63, Art.No. CD005470. DOI: 10.1002/14651858.CD005470.pub2
  • Byford, S., Barrett, B., Roberts, C., Clark, A., Edwards, V., Smethurst, N., & Gowers, S. (2007). Economic evaluation of a randomised controlled trial for anorexia nervosa in adolescents. British Journal of Psychiatry, 191, 436440.
  • Child Growth Foundation. (1996). Girls Growth Chart, London: Boys Growth Chart.
  • Crisp, A.H., Norton, K., Gowers, S., Halek, C., Levett, G., Yeldham, D., & Bhat, A. (1991). A controlled study of the effect of therapies aimed at adolescent and family psychopathology in anorexia nervosa. British Journal of Psychiatry, 159, 325333.
  • Eisler, I. (2005). The empirical and theoretical base of family therapy and multiple family day therapy for adolescent anorexia nervosa. Journal of Family Therapy, 27, 104131.
  • Fairburn, C. (1995). Overcoming Binge Eating. New York, London: The Guildford Press.
  • Fairburn, C., & Beglin, S. (1994). Assessment of eating disorders: Interview or self-report questionnaire? International Journal of Eating Disorders, 16, 363370.
  • Gowers, S.G., Clark, A., Roberts, C., Griffiths, A., Edwards, V., Bryan, C., & Barrelt, B. (2007). Clinical effectiveness of treatments for anorexia nervosa in adolescents. British Journal of Psychiatry, 191, 427435.
  • Gowers, S.G., Edwards, V.J., Fleminger, S., Massoubre, C., Wallin, U., Canalda, G., & Boyadjieva, S. (2002). Treatment aims and philosophy in the treatment of adolescent anorexia nervosa in Europe. European Eating Disorders Review, 10, 271280.
  • Gowers, S., Weetman, J., Shore, A., Hossain, F., & Elvins, R. (2000). Impact of hospitalisation on the outcome of adolescent anorexia nervosa. British Journal of Psychiatry, 176, 138141.
  • Herpertz-Dahlmann, B., Muller, B., Herpertz, S., Heussen, N., Hebebrand, J., & Remschmidt, H. (2001). Prospective 10-year follow-up in adolescent anorexia nervosa—course, outcome, psychiatric comorbidity, and psychosocial adaptation. Journal of Child Psychology and Psychiatry, 42, 60312.
    Direct Link:
  • Kleve, L., & Savile, A. (2011). Clinical practice with Eating disorders in outpatient CAMHS. A survey of UK clinicians. (unpublished manuscript).
  • Lock, J., Le Grange, D., Agras, W.S., & Dare, C. (2001). Treatment manual for Anorexia Nervosa; A Family-Based Approach. New York: The Guildford Press.
  • Lock, J., Le Grange, D., Agras, W., Moye, A., Bryson, S.W., & Jo, B. (2010). Randomized clinical trial comparing family based treatment to adolescent-focused individual therapy for adolescents with anorexia nervosa. Archives of General Psychiatry, 67, 10251032.
  • Morgan, H.G., & Russell, G.F.M. (1975). Value of family background and clinical features as predictors of long-term outcome in anorexia nervosa: Four-year follow-up study of 41 patients. Psychological Medicine, 5, 355371.
  • Mufson, L., Noreau, D., Weissman, M.M., & Klerman, G.L. (1993). Interpersonal psychotherapy for depressed adolescents. New York, London: The Guildford press.
  • National Institute for Clinical Excellence. (2004). Eating disorders: Core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders. Clinical Guideline 9. London: National Institute for Clinical Excellence.
  • Newton, T. (2001). Consumer involvement in the appraisal of treatment for people with Eating Disorders: A neglected area? European Eating Disorders Review, 9, 301308.
  • Petterson, G., & Rosenvinge, J.H. (2006). Tjenestetilbudet til pasienter med spiseforstyrrelser: En nasjonal kartlegging i primær- og spesialisthelsetjenesten. Tidsskrift for Norsk Psykologforening, 11, 11591168.
  • Richard, M. (2005). Care provision for patients with eating disorders in Europe: What patients get what treatment and where? European Eating Disorders Review, 13, 159168.
  • Steinhausen, H.C. (2002). The outcome of anorexia nervosa in the 20th century. American Journal of Psychiatry, 159, 128493.
  • Strober, M., Freeman, R., & Morrell, W. (1997). The long-term course of severe anorexia nervosa in adolescents: Survival analysis of recovery, relapse, and outcome predictors over 10–15 years in a prospective study. International Journal of Eating Disorders, 22, 33960.
  • Tierney, S. (2008). The individual within a condition: A qualitative study of young people's reflections on being treated for Anorexia Nervosa. Journal of the American Psychiatric Nurses Association, 13, 368375.