Introduction
Over the last years, the treatment of the most severe eating disorders (anorexia nervosa and bulimia nervosa) has significantly improved, and today we have available various therapies, whose efficacy has been demonstrated by rigorously controlled and randomised trials. In bulimia nervosa, over 30 controlled and randomised trials have demonstrated that patients treated with the Cognitive-Behavioural Therapy (CBT), and facing the maintenance factors which specifically relate to this disorder (worry about body weight and shape, strict diet, binges and compensation behaviours), reach complete recovery in 50% of cases and a significant improvement of their symptoms in 80% of cases.

Two controlled trials have shown that the Interpersonal Therapy (IPT), even though not as quick as the CBT in generating an improvement in symptoms, at one year’s distance from completion reaches similar effectiveness values. In addition to psychotherapy, a number of studies have shown that one category of psychotropic drugs, that is antidepressants, can cause an interruption of binges and compensation behaviours in 20% of patients treated. Unfortunately the effects of antidepressants do not appear to last in time and most patients experience a relapse after 4-5 months’ treatment. There are few controlled trials in anorexia nervosa and the results produced are insubstantial. The choice of the type of treatment still broadly depends on the therapist’s preferences and on the resources available on the territory.
Anyway, evidence exists that in patients under 18 a family form of therapy, set up by a group of researchers of the Moudsley Hospital in London, is more effective than individual psychodynamic therapy. Furthermore, a great number of case reports and clinical experiences indicate that the extended CBT, a treatment which, besides tackling the maintenance factors specific for the disorder, also addresses problems such as low self-esteem, emotional regulation, and interpersonal and family problems, proves effective with many out-patients. Unfortunately, a broad sub-group of patients (approximately 50%) does not respond to the out-patient treatment (psychotherapy and/or drugs) and requires more intensive treatments. In Italy, unlike other European countries and the USA, where out-patient treatment non-responders and patients displaying severe medical complications are admitted in internal medicine and psychiatric departments, a new treatment has been recently developed; this is called “intensive hospital rehabilitation”.
This article describes the theoretic bases, the practical aspects and the results obtained by the intensive hospital rehabilitation programme implemented at the Nursing Home Villa Garda. The treatment, which was formally adopted in 1988, was the first specialised residential programme implemented in Italy for treating the most severe forms of eating disorders.

The Theoretical Bases of the Programme
A good treatment is usually supported by a good theory. Over the last years in the field of eating disorders the validity of the cognitive-behavioural theory has been demonstrated by cognitive science studies, by clinical trials that have shown the effectiveness of the CBT and by studies which, by using sophisticated statistical models, (cross-sectional structural equation modelling), have analysed the most significant psychopathologic relations in eating disorders. The cognitive-behavioural theory assumes that the central maintenance mechanism of eating disorders is a dysfunctional personal value assessment scheme (see picture 1). According to this theory, at the heart of this psychopathology is the individual’s tendency to exclusively or largely judge his own value depending on the control he has on his eating habits, as well as on his body weight and shape. Again according to the theory, the dysfunctional self-assessment scheme develops as a result of a multiple and complex interaction of various (socio-cultural, individual and family) risk factors, which are deemed to be triggered by certain precipitating factors.

The activation of the self-assessment scheme leads the individual to develop specific thoughts and worries regarding his eating behaviour, body weight and shape, and certain typical behaviours such as for instance strict diet, excessive physical exercise, binges, compensation behaviours, body checking (compulsive control of body weight and shape) and the avoidance behaviours related to the body. The positive and/or negative reinforcements that the individual perceives when he manages to control what he eats, the body weight and shape, the low weight and the fasting syndrome, typical eating disorder behaviours, contribute in turn in keeping the dysfunctional self-assessment scheme constantly activated, which results in the fact that the eating disorder, once triggered, tends to be self-perpetuated. It has recently been found that a subgroup of patients displays certain additional maintenance factors which, depending on the specific situation, may be perfectionism, low nuclear self-esteem, intolerance to emotions and interpersonal problems (including family problems). Based on this theory, a new transdiagnostic cognitive-behavioural treatment was devised; this can be applied with a minimum number of changes in various therapy settings (out-patient, day-hospital or intensive hospital rehabilitation therapy). ‘Cognitive-behavioural’ means that the treatment follows the theory and praxis of the cognitive-behavioural theory (see above) and refers to the theoretic and clinical principles of the most famous authors in the field of cognitive-behavioural theory and therapy. ‘Transdiagnostic’ means that the treatment has been studied and designed to be applied with a minimum number of changes in all eating disorders (anorexia nervosa, bulimia nervosa and atypical eating disorders). The choice to organise a transdiagnostic and non-specific therapy for each disorder is the result of the observation that anorexia nervosa, bulimia nervosa and atypical eating disorders share distinctive clinical features and that patients move in time from one disorder to the other. The patient’s specific eating disorder is not relevant for the treatment. On the other hand, the contents of the therapy depend on the specific existing psychopathologic features and on the processes which allow their maintenance.

Indications for Hospitalisation
Although no universally accepted guidelines exist to advise the admission to hospital of patients suffering from eating disorders, the following hospitalisation criteria, which are essentially based on the clinical experience of the various authors, are widely accepted by the international scientific community: · severe or rapid weigh loss associated with psychological and social medical complications requiring hospital treatment;

· non-response to out-patient or day-hospital treatment;

· presence of significant psychiatric comorbidity factors hindering out-patient treatment;

· presence of severe medical complications; · need to separate the patient from his family.

Place
Rehabilitation hospitalisation has to be carried out in units (the Nursing Home Villa Garda handles 25-27 patients) specialising in the treatment of eating disorders, where a multi-disciplinary team with a long experience in the treatment of these disorders operates. These units, which can be psychiatric or non-psychiatric (such as the one of the Nursing Home Villa Garda), should be situated in small rehabilitation facilities, which are not burdened with the high costs of hospitals for acute cases.

My experience as a supervisor of psychiatric and non-psychiatric facilities has led me to the conclusion that non-psychiatric departments offer significant advantages compared to psychiatric units. Indeed, patients suffering from eating disorders, with the exception of an extremely limited number of cases, do not require closed or highly restrictive facilities, but a comfortable, quiet and safe environment, suitable for their age, which empathetically encourages them to face the difficult path that leads to recovery. A too restrictive approach, such as the one that inevitably occurs in psychiatric facilities, encourages patients’ psychological regression and the development of impulsive or excessive dependence behaviours.

The Hospital Team and the Cognitive-Behavioural Milieu
The treatment developed at Villa Garda is not a wide-ranging or integrated programme in which each professional makes his specific competence available but, although administered by various professional figures (physicians, dieticians, psychologists/psychotherapists and professional nurses), it is based on a single theory (see above) and is applied in a context that can be described as a “cognitive-behavioural milieu”. This term is used to refer to hospital programmes based on three essential features: a) the therapeutic work is chiefly based on the CBT theoretical assumptions and methods; b) the cognitive-behavioural theory is employed to design and implement the treatment; c) the CBT is associated with other medical treatments in an integrated performance model. The team members in the department, whilst having specific and separate tasks, have all received training in CBT, they use the same language and have common treatment objectives.

The Therapeutic Programme
The cognitive-behavioural therapy follows an “experimental” model, in which each step is carefully planned and the decision to proceed to the next step is evaluated with great care. The treatment can be regarded as some sort of “experiment” to assess if the therapy can supply more effective and satisfactory solutions compared to those obtained through the eating disorder. If the patient does not prove satisfied of the results obtained with the therapy, he will always be free to resume the diet and the other means of weight control required to tackle his problems. The therapeutic programme follows two parallel courses: the first one handles the low body weight, the diet and the other weight and binge control behaviours; the second one handles psychological issues, such as the dysfunctional self-assessment scheme and, if applicable, other maintenance factors (perfectionism, low nuclear self-esteem, interpersonal and family problems and intolerance to emotions). The treatment provides for three phases: the hospital phase, the residential day-hospital phase and the relapse prevention phase.

1. Hospital Phase (Duration: 13 Weeks) The hospital phase is aimed at eroding the chief psychological and behavioural factors which aid development and maintenance of eating disorders (see Tab. 1), by means of three chief procedures:

a) diagnostic assessment,

b) nutritional rehabilitation and

c) intensive psychotherapy.

A. Diagnostic Classification During the first two days the patient is subject to a multidimensional diagnostic assessment including medical examination, nutritional anamnesis, psychological evaluation and various tests (both biohumoral and instrumental), as well as psychodiagnostic tests (see table 1). The patient is also asked to sign a “therapeutic contract” in which he confirms his acceptance of the programme rules. B. Nutritional Rehabilitation The programme provides for two separate phases:

Phase 1. Assisted mechanical nutrition;

Phase 2. Planning meals and responsible mechanical nutrition.

Phase 1. Assisted mechanical nutrition. This phase is carried out for the first two months and until the patient reaches a Body Mass Index or BMI (weight in kg/height raised to the second power in metres) above 18.5 kg/m2, an adequate behaviour during meals and a remission of the chief symptoms characterising the eating disorder. In this phase patients have their meals is a room shared with other patients suffering from eating disorders. The dietician always attends meals: his role is to teach the patients cognitive-behavioural techniques in order to learn to eat “mechanically”.

By eating mechanically the patient learns to ascribe his dysfunctional thoughts (on food, body weight and shape) and physical sensations to the symptoms of his disorder and not to use them to handle his eating behaviour (this is called re-attribution cognitive-behavioural technique). The patient is encouraged to view food as a “medicine” and to fully trust the nutritional plan prescribed by the team. This eating routine has to be continued until eating is no longer influenced by dysfunctional worries and thoughts on food, body weight and shape. The diet provides for three chief meals, plus an afternoon snack. During the first two weeks the diet scheme is a low-calory one (approximately 1500 Kcal) to avoid potential complications associated with re-nourishment. The diet calory content is subsequently increased by 500 Kcal per week until a level of 2,500 Kcal/day is reached. Calory adjustments are subsequently introduced according to the weight increase speed. The programme is aimed at allowing patients to reach a BMI of at least 20, with a wait increase ranging between 1 and 1.5 kg a week; if the patient’s weight increases by less than 1 kg a week, the diet is increased by 500 Kcal/day; if it increases by more than 1.5 kg a week, the diet is reduced by 250 Kcal/day. The body weight is checked twice a week and notified to the patients. After all meals, to help patients not to resort to compensation behaviours (self-induced vomiting), bathrooms remain locked for one hour. Patients who have difficulties is overcoming the symptoms of their disorder (such as hyperactivity, for instance) are provided constant individual assistance during the day.

Phase 2. Planning meals and responsible mechanical nutrition. Planning meals involves the following procedures:

a) having meals without the assistance of the dietician in a different room from the one used for assisted mechanical nutrition;

b) weekly planning what to eat from the menu of the nursing home;

c) planning in advance the time, nature and amount of food to be eaten during meals outside the nursing home;

d) taking part in two weekly group meetings with the dieticians: a theoretical one for the planning of the practical cooking group, and a practical one during which patients cook and have their meals together with other patients in the group;

e) doing specific exercises involving exposure to high-risk situations, such as for instance getting one’s own helping, eating in problematic situations (at the restaurant, at a fast-food, at a pizzeria, with one’s parents or with friends), or eating food that one usually avoids;

f) continuing to autonomously use the cognitive-behavioural techniques to eat mechanically (responsible mechanical nutrition);

g) autonomously managing alternative activities (it is no longer compulsory to remain in the alternative activity room after meals).

C. Intensive Psychotherapy During the hospitalisation period, after a few initial diagnostic interviews, the patient is granted an intensive psychotherapy treatment both individually (one session a week) and in a group (seven days out of seven). The group sessions use the most modern cognitive-behavioural procedures to improve the patients’ dysfunctional self-assessment schemes. The techniques employed are aimed at: · training the patient on the eating disorder development and maintenance cognitive-behavioural model;

- widening and modifying the dysfunctional self-assessment scheme based on the excessive value set on the eating behaviour, on the body weight and shape and their control;

- if applicable, tackling any additional maintenance factors, such as perfectionism, low nuclear self-esteem, intolerance to emotions and interpersonal problems. Weekly group sessions include: 1 group session to learn to handle the symptoms, 2 psychoeducational group sessions, 3 group sessions to improve self-assessment, 1 group session to improve the body image and, for patients who require it, 1 group session to learn to modulate their emotions without using dysfunctional behaviours (“Emotion Regulation” module). Patients under 18 take part together with their family in the “Family Therapy” module, which includes six family therapy sessions and ten group meetings with other parents.

2. Day-Hospital Phase (Duration: 7 Weeks) The day-hospital phase has been included in the programme to enable a gradual transition to out-patient therapy. In this phase, the patient increasingly experiences meal planning and responsible mechanical nutrition. He gradually gets used to having his meals outside the rehabilitation unit (first for breakfast, next for his afternoon snack, followed by dinner and lunch) up to the point of having all meals outside the facility during the last week of stay. During the day-hospital phase, patients spend all weekends at home and return to the clinic on Mondays. Throughout the day-hospital period the therapies that have been started during the hospitalisation phase continue. Towards the end of the day-hospital phase, sessions involving the whole family are scheduled both with the dietician and with the physician, in order to prepare the return home. These meetings are aimed at discussing the behaviours regarding all the meal planning and mechanical nutrition, school or work, the place where the patient will live and the type of out-patient therapy that will be carried out.

3. Relapse Prevention Phase (1 or more Years) After release, the patients are invited to take part in an out-patient cognitive-behavioural therapy programme and to free group meetings organised at the Nursing Home Villa Garda, every two weeks for six months, aimed at preventing relapses. The group sessions are open and provide for a maxim number of eight participants; they are organised according to a guided self-help pattern (patients help each other to avoid a relapse, but are supported in this difficult task by a psychologist/psychotherapist from the department).

Results Obtained with the Treatment
Since 1999, 135 patients suffering from anorexia nervosa have been consecutively admitted. The patients’ average age was 22.7 years (SD 5.6). The treatment produced a significant increase of the BMI (intention to treat analysis), which passed from 14.1 (SD 1.7) upon admission to 19.0 (SD 1.7) upon release (p<.0001). 113 patients out of 135 (84%) completed the whole treatment (13 weeks hospitalisation plus 7 weeks residential day-hospital therapy) and normalised their body weight. The normalisation of the BMI was associated with a significant improvement of many psychological symptoms associated with the eating disorders (depression, anxiety, low self-esteem, social isolation, worries about food, body weight and shape). For 74 patients consecutively released in 1999 and 2000, the data relating to the BMI and the menstrual cycle at one year’s distance from release are available. The BMI, which upon admission was 14.4 (SD 1.6), after a one- year follow-up period was 17.7 (SD 2.4), close to the 18.5 value which represents the minimum normal weight threshold. Out of this sample of patients, 43% related the presence of at least one menstrual cycle during the three months before the assessment.

Conclusions
This article has presented a cognitive-behavioural intensive hospital rehabilitation treatment developed in a non-psychiatric department of the Nursing Home Villa Garda. The programme has proved capable of normalising body weight and of improving many psychological symptoms associated with eating disorders in 84% of hospitalised patients. At one year’s distance from release, even though one subgroup of patients had not maintained the results achieved with the rehabilitation treatment, the average BMI was significantly above the value detected upon admission and was close to the minimum normal weight threshold. The results obtained through the intensive rehabilitation programme, even though not yet completely satisfactory, offer tangible hope of treatment and recovery for many patients suffering from severe eating disorders which do not respond to the out-patient therapy. Translated by interpres sas

Riccardo Dalle Grave
Responsabile Unità di Riabilitazione Nutrizionale della casa di cura Villa Garda Responsabile scientifico dell’Associazione Italiana Disturbi dell’Alimentazione e del Peso (AIDAP)

Arianna Banderali
Dirigente Medico presso l'Unità di Riabilitazione Nutrizionale della Casa di Cura Villa Garda
Presidente AIDAP