**3. Comparison of the German S3-Guideline with International Evidence-Based Clinical Treatment Guidelines**

#### *3.1. International Evidenced-Based Eating Disorders Guidelines*

There are currently several additional evidence-based guidelines available, which provide recommendations regarding the diagnosis and treatment of eating disorders. Most of the guidelines were written by multidisciplinary groups (comprising health care professionals and researchers), and most were designed solely for use by health specialists involved in the treatment of eating disorders. The most recent of these guidelines are the Dutch [37] and the revised British guidelines [38], both published in 2017. The British guideline [38], published by the National Institute for Health and Care Excellence (NICE), addresses all age groups (children, adolescents and adults), and all eating disorder categories (AN, BN, BED and Other Specified Feeding or Eating Disorders (OSFED)). Several lay members of the community were involved in the development of this guideline. The Dutch guideline addresses AN, BN and BED [39]. This guideline, designed to be used by both specialists and population members, is only available in Dutch [39]. Healthcare professionals collaborated with patients and relatives, as well as health insurance representatives, during the developmental stages of the guideline [39].

The next most recent guideline, published in 2016, is the Danish guideline [40]. This 'quick guide', provides a brief overview, designed solely for the treatment of AN. The guideline is available in English, and it addresses all age groups. The full-length version of this guideline is only available in Danish. The Australian and New Zealand guideline [41] was published in 2014 by the Royal Australian and New Zealand College of Psychiatrists. Community members and stakeholders collaborated with healthcare professionals and academics in the development of the guideline. This guideline contains two sections separately addressing AN in children and adolescents, and in adults. BN and BED, as well as avoidant/restrictive food intake disorder, are also addressed.

In 2012, the American Psychiatric Association (APA) released a guideline watch [42], reviewing new evidence published since the last APA guideline in 2006, but gives no explicit recommendations [43]. This guideline addresses AN, BN and BED, and also makes reference to EDNOS. The guideline is designed primarily for the treatment of adults, but also briefly addresses the treatment of children and adolescents. The French guideline [44], published in 2010, is written specifically for AN. It addresses all age groups and is available in English. In 2009, the Spanish guideline [45] for eating disorders was published. This guideline, which concerns eating disorder patients over 8 years of age, is written not only for healthcare specialists, but also for the population and educational professionals. It addresses AN, BN, BED and EDNOS.

In addition to these national guidelines, several more specific evidence-based guidelines also exist. A guideline, developed specifically for the Canadian province of British Columbia, was released in 2010 [46]. This guideline addresses AN, BN and EDNOS (except BED), and advisesthe on treatment of all age groups. In 2011, the World Federation of Societies of Biological Psychiatry (WFSBP) released a guideline specifically addressing the pharmacological treatment of eating disorders [47]. This guideline, written in English, addresses the pharmacological treatment of AN, BN and BED. In 2014, the Management of Really Sick Patients with Anorexia Nervosa (MARSIPAN and Junior

MARSIPAN) guideline [48] was published, a guideline which specifically addresses the treatment of children, adolescents and adult patients with 'severe' AN.

In line with an evidence-based approach, most of the guidelines explicitly state that the development of the guideline involved a systematic literature review, a rating of the identified literature, and a complex consensus process, involving collaboration and review by numerous experts [14,38,41,43–48]. Only the MARSIPAN [48] and WFSBP guidelines [47] do not explicitly refer to a complex consensus process, and the British Columbia guideline [46] does not mention a rating system. The Danish 'quick guide' [40] has a complete absence of information on the methodological process. However, the inclusion of evidence levels in the guide implies that the developmental process was rigorous. A detailed review of the evidence upon which the recommendations are based is only available in the British guideline and the Danish full-length guideline. A more detailed comparison of the methods employed in developing the guidelines goes beyond the scope of this review article. All of the guidelines are available online. The Australian and New Zealand, MARSIPAN, WFSBP and APA guidelines are published in online scientific journals and partly in print versions, and the remainder of the guidelines are available on the relevant publishing society's website.

#### *3.2. Commonalities and Differences*

#### 3.2.1. Treatment Setting

For adults: Similar to the German guideline [14], all remaining guidelines (excluding the Danish [40] and WFSBP guidelines [47]) recommend outpatient treatment as a first treatment option, suggesting day patient or inpatient treatment as a more intensive treatment option if outpatient treatment proves ineffective [38,39,41,43,44,46,48]. The German guideline states, however, that in some cases this 'stepped-care' approach may not be appropriate.

Inpatient treatment is recommended in cases with a BMI <15 kg/m2, rapid or continuing weight loss (>20% over 6 months), high physical risk, severe co-morbid conditions or denial of illness. If these criteria are met, an inpatient setting may be necessary for initial treatment. Likewise, all remaining guidelines (excluding the Danish and WFSBP guidelines) also suggest more intense treatment settings from the outset in cases of severe medical instability. All of these guidelines provide information regarding hospital admission criteria with varying degrees of detail, but agree on the necessity to judge the need for hospitalisation on an individual and multifactorial basis. Further, they state that compulsory treatment is possible in the case of extreme medical complications. The Danish and WFSBP guidelines do not make reference to treatment setting. For an overview of indicators of high medical risk and the handling of medical complications see the review of Zipfel and colleagues [4].

For children and adolescents: Corresponding to the treatment recommendations for adults, outpatient treatment is proposed as the first line treatment for children and young people by the German [14] and most other guidelines [38,41,43,44] if the patient is in a stable medical state. If more intensive care is needed, several guidelines suggest a graduated procedure from inpatient to partial and finally to outpatient treatment programs [40,44,45]. Only the German guideline [14] gives a special recommendation for a referral to day patient treatment. Interestingly, the British and accordingly the Spanish guidelines advise admitting children and young people to a setting with age-appropriate facilities, which are near to their home and have the capacity to provide appropriate educational activities [38,45].

Regarding medical risk and necessity for inpatient treatment, the Australian and New Zealand, British Columbia, British, APA and French guidelines [38,41,44,46,49] provide exact criteria, such as a BMI below the 3rd percentile or an expected body weight (EBW) below 75%, an abnormally low heart rate or blood pressure, electrolyte disturbances, etc. However, the exact values vary between countries. As for adults, these guidelines also indicate psychiatric risk factors, such as suicidality or severe self-injurious behaviour. The German and Spanish guidelines [14,45] are more unspecific to indicate hospitalisation (see above). The German and French guidelines [14,44] also refer to psychosocial risks, such as social isolation and family crisis, to consider inpatient treatment.

#### 3.2.2. Psychotherapy

For adults: All guidelines except for the Danish [40] and WFSBP [47] address the efficacy of specific psychological interventions. No guideline recommends one single superior treatment option. The German [14], British [38] and Dutch guidelines [39] conclude that cognitive-behavioural therapy (CBT or CBT-E respectively), MANTRA, and SSCM are equally effective treatment options, and so, all treatments are recommended as first-line options. Additionally, the German guideline recommends FPT as another first-line treatment option. The remaining guidelines all review evidence for CBT, as well as a variety of other treatments including SSCM [41], psychodynamic therapy [43–46], interpersonal therapy [43,45,46], behaviour therapy [45] and 'systematic and strategic therapies' [44]. These guidelines all conclude that psychological interventions are effective, however, state that there is insufficient evidence to identify which is the most efficacious. The French [44], Dutch [39] and APA guidelines [43] also suggest that psychological interventions may not be as effective in severely malnourished patients.

The Danish guideline [40] also recommends the use of psychotherapeutic treatments, however, does not make any recommendations regarding specific interventions. This guideline provides a 'weak recommendation' that both group and individual psychotherapeutic treatment be considered as first-line treatment options, based on 'very low evidence' which suggests the approaches are equally effective. Recommendations for the inclusion of alternative elements, such as meal support and supervised physical activity, during the treatment phase are mentioned. Other guidelines make specific recommendations against alternative treatments; for example, the German [14] and the Australian and New Zealand guidelines [41] state that nutritional counselling alone should not be used as the sole treatment, and the British guideline [38] recommends against the use of alternative physical therapies, such as yoga, warming therapy, transcranial magnetic stimulation and acupuncture. The Spanish guideline [45] also advises against the use of excessively rigid behavioural programs for inpatients.

Some guidelines make recommendations regarding the required duration of treatment. The Australian and New Zealand guideline [41] states that a longer-term follow-up is necessary as relapse is common, and the Spanish guideline [45] states that duration of treatment should span at least six months for outpatients and twelve months for inpatients. The APA guideline [43] states that due to the enduring nature of the illness, psychotherapeutic treatment is usually required for at least one year, and the British guideline [38] makes specific recommendations regarding the time span of treatments, for example suggesting that CBT treatment for eating disorders should consist of 40 sessions over 40 weeks. The French guideline [44] recommends that treatment should last at least one year after significant clinical improvement, and the German guideline [14] states that after outpatient treatment, patients should regularly meet with their general practitioner (GP), or other care coordinator, for at least one year. The German guideline also recommends that the last phase of inpatient treatment before transfer to an outpatient setting should include a stabilisation period where patients demonstrate that they can maintain the achieved weight gain for a specified amount of time.

Some treatment guidelines make additional specific recommendations. The German [14], French [44], MARSIPAN [48] and Australian and New Zealand guidelines [41] all emphasise the importance of adopting a multi-disciplinary, collaborative approach to treatment. In a similar vein, the German [14], British Columbia [46] and APA guidelines [43] highlight the importance of effective communication between all involved health workers, and recommend identifying someone to act as the primary care coordinator, such as the patient's GP.

The MARSIPAN guideline [48] is specifically written regarding the treatment of patients who have a severe or enduring form of AN. The Australian and New Zealand [41] and British Columbia guidelines [46] also include comprehensive sections which address the treatment of such patients and suggest taking an alternative approach, focused on enhancing quality of life. The French [44], German [14] and APA guidelines [43] also briefly mention the treatment of patients with enduring AN. Other guidelines provide information regarding other additional elements related to AN. For example, both the Spanish [45] and French guidelines provide information regarding the care required for

pregnant patients. Additionally, the APA and British Columbia guidelines include recommendations for therapists and specialists regarding communicating with patients (for example addressing the therapeutic relationship, boundaries). The German guideline does not entail any recommendations, but devotes a separate chapter to this topic.

For children and adolescents: All guidelines strongly recommend the involvement of parents or near caregivers in all treatment settings. The Australian and New Zealand, Spanish, APA and German guidelines explicitly mention family-based treatment or therapy (FBT) [14,41,43,45]. However, the Australian and British guidelines also propose alternatives if FBT is not appropriate, such as other forms of family therapy [41], as well as individual treatment, such as adolescent-focused therapy (AFT) or CBT, in older adolescents [38,41]. No guideline gives an explicit advice whether conjoint or separate FBT should be conducted. The French guideline does not refer to FBT, but to family therapy in general [44]. The British guideline also does not specifically use the term FBT, but has its own terminology instead (anorexia nervosa-focused family therapy, FT-AN) [38]. Although many key features of this treatment resemble FBT, FT-AN also includes other approaches, such as multi-family therapy, conjoint or separate family therapy and exclusion or inclusion of a family meal, which is a core feature of FBT. The British guideline also requests therapists and staff to be aware of or address carers' needs [38].

A summary of guidelines' essential key recommendations regarding psychotherapy for AN is shown in Table 2.



France; GER, Germany; NETH, The Netherlands; SP, Spain; UK, United Kingdom; US, United States; CBT(-E), (Enhanced) Cognitive Behaviour Therapy; FPT, Focal Psychodynamic Therapy; MANTRA, Maudsley Model of Anorexia Nervosa Treatment for Adults; SSCM, Specialist Supportive Clinical Management; IPT, Interpersonal Therapy; FBT, Family-Based Treatment/Therapy; FT-AN, AN-focused Family Therapy; 1 and siblings; ?, ambiguous evidence.
