Chapter 46 Intensive Treatments for Eating disorders

Print: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

1 of 15

10/05/2009 17:33

Print Close Window

Marion P. Olmsted, D. Blake Woodside, Jacqueline C. Carter, Traci L. McFarlane, Randolf A. Staab, Patricia A. Colton,

Lynda L. Molleken: Chapter 46. Intensive Treatments, in Gabbard’s Treatments of Psychiatric Disorders, 4th Edition.

Edited by Glen O. Gabbard. Copyright ©2009 American Psychiatric Publishing, Inc. DOI:

10.1176/appi.books.9781585622986.261789. Printed 5/10/2009 from www.psychiatryonline.com

Gabbard’s Treatments of Psychiatric Disorders > Part IX. Eating Disorders >

Chapter 46. Intensive Treatments

INTRODUCTION

Inpatient and day hospital treatments are the most common intensive interventions for eating

disorders and are appropriate for the most severely ill patients. Most programs are multifaceted

and based on a biopsychosocial model; because of their complexity, the specific components of

treatment and their relative emphasis will likely vary across settings. There is evidence that

intensive therapies have good short-term effectiveness for patients who are willing to accept this

type of treatment. However, there are no comparative studies indicating that one specific approach

is better than any other.

In this chapter we review the literature related to the efficacy of intensive treatment and describe

suitable patients, criteria for hospital admission, issues related to hospital discharge, the treatment

of comorbidity, and the structure and goals of intensive treatment programs. Some

cognitive-behavioral aspects of treatment and group therapy features are described, along with

strategies for enhancing motivation in an intensive group environment. Staff issues related to

working closely with eating disorder patients in an intensive environment are also considered.

EVIDENCE FOR THE EFFICACY OF INTENSIVE TREATMENT

There is a striking paucity of research on the effectiveness of intensive treatments in eating

disorders. The relatively high cost of intensive treatment is both an indication for and an obstacle to

the conduct of well-controlled effectiveness trials. Most of the available literature consists of small

uncontrolled studies based on variable definitions of outcome.

Engagement With Treatment

A significant number of individuals with severe eating disorders refuse participation in intensive

treatment, and others initiate treatment but leave prematurely. Reported dropout rates for

inpatient treatment range from 20% (Surgenor et al. 2004) to 51% (Woodside et al. 2004). A

number of predictors of dropout have been identified, including later age at onset of anorexia

nervosa (AN), older age at admission, longer duration of illness, and lower socioeconomic status

(Vandereycken and Pierloot 1983). Two studies found that patients with the binge-purge subtype of

AN were more likely to drop out of inpatient treatment (Kahn and Pike 2001; Woodside et al. 2004).

Other predictors of dropout were higher admission weight, higher levels of depression at

admission, more severe eating-disorder psychopathology at admission, and higher levels of

maturity fears (Zeeck et al. 2004).

Although the evidence is limited, reported dropout rates for day hospital treatment are not as high,

possibly because participants retain more freedom. Olmsted et al. (2003) found dropout rates of

18.8% and 13.5%, respectively, for a 4-day and a 5-day program in Toronto. Franzen et al. (2004)

reported a dropout rate of 15.2% for a day hospital in Munich, and dropout was associated with

more severe bulimic symptoms, higher levels of aggression and extraversion, and lower levels of

inhibition. Patients who drop out of intensive treatments appear to be among the more severely ill,

indicating that current intensive treatment programs are not a panacea.

Treatment Effectiveness for Participants

Results of case series studies of inpatient treatment for AN have consistently shown that inpatient

treatment is effective at achieving weight restoration, at least in the short term (Bowers andPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

2 of 15

10/05/2009 17:33

Anderson 1994; Steinhausen 1985; Touyz et al. 1987). However, few studies report follow-up data,

and most do not report outcome data for binge eating and purging symptoms. Engel and Wilfarth

(1988) reported good maintenance of change at 2-year follow-up for a series of 39 female AN

patients who were weight-restored with inpatient treatment. Lowe et al. (2003) studied a case

series of 279 AN patients who received inpatient treatment. Follow-up data, available for only 35%

of the sample, indicated that changes in weight were maintained at follow-up.

At our own center, 183 patients with AN were admitted to the inpatient program during the period

between 2000 and 2005. In terms of weight outcomes, 71 patients (39.3%) achieved a body mass

index (BMI) of at least 20, 35 (19.1%) achieved a BMI between 18.5 and 20, and 77 (42.1%) had a

BMI below 18.5 at discharge. Among the 106 patients (57.9%) who achieved a BMI of at least 18.5,

the mean length of stay was 14.6 weeks, the mean weight gain was 13.4 kg, and the mean rate of

weight gain was 0.92 kg per week. Significant improvements were also seen on measures of both

eating disorder psychopathology and general psychopathology, including depression, anxiety, and

self-esteem. In terms of bulimic symptoms, 92.5% were free from binge eating and 88.7% were

free from purging symptoms over the 4 weeks preceding the end of inpatient treatment.

To our knowledge, there is only one published randomized controlled trial of inpatient treatment for

AN (Gowers et al. 1989, 1994). This study compared inpatient with two types of outpatient

treatment for AN. There was a higher rate of refusal to participate following randomization among

patients assigned to inpatient treatment and refusals were apparently replaced (Crisp et al. 1991).

Among participants, there were no significant differences between inpatient and outpatient

treatments in terms of weight gain or Morgan-Russell outcome indices (Morgan and Russell 1975)

at follow-up. Lower weight at presentation and the presence of vomiting were associated with a

poorer outcome.

Inpatient treatment for bulimia nervosa (BN) is not common. Williamson et al. (1989) compared

outcomes of 27 BN patients who received 5 weeks of inpatient treatment with those of 22 BN

patients who were treated on an outpatient basis for 15 weeks. Patients who had received inpatient

treatment showed a more rapid improvement in binge-eating and purging symptoms, but they were

more likely to relapse at 6-month follow-up. Gleaves et al. (1993) studied a case series of 452 BN

patients who received inpatient treatment. Over a 4-year follow-up period, approximately one-third

remained free from binge eating and purging, one-third reported subthreshold symptoms, and the

rest continued to meet diagnostic criteria for BN. Zeeck et al. (2004) found that inpatient and day

hospital treatment produced similar outcomes for 36 patients with BN. At 18-month follow-up, 28%

of day patients and 33% of inpatients were abstinent from binge eating and vomiting.

Although inpatient treatment is more common for AN than for BN, day hospital treatment programs

typically accommodate patients with either diagnosis and outcomes are reported for both. Day

hospital programs that are multimodal and that incorporate a cognitive-behavioral focus on

symptom change consistently report significant decreases in bingeing and purging behaviors in BN

patients, weight gain in AN patients, and improvements in eating-disordered attitudes and

depressive symptoms in both adults (Gerlinghoff et al. 1998; Kaplan et al. 1997; Olmsted et al.

1994, 1996; Williamson et al. 2001; Zipfel et al. 2002) and adolescents (Dancyger et al. 2003;

Heinberg et al. 2003).

In contrast to the more common cognitive-behaviorally based day hospital programs, Thornton et

1. (2002) described a 4-day program that was based on the principles of supportive and

interpersonal group therapy. The goals of the program did not include behavioral change,

normalization of eating, or weight gain. An evaluation of the first 23 patients with AN indicated that

95% lost weight during the program and 64% required admission to an inpatient unit. The failure

of this approach to induce weight gain highlights the need for day programs to maintain behavioral

goals and expectations as a priority.

More recently, these researchers (Thornton et al. 2004) have described a 5-day and a 3-day

intervention, in which the 5-day program involves less motivated patients (and therefore more

emphasis on motivational enhancement) and the 3-day program is consistent with the more typicalPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

3 of 15

10/05/2009 17:33

action-based programs found at other centers. There was a significant increase in weight for AN

patients in the 3 day program but not the 5-day program, which was predicted given the differences

in levels of motivation. Preliminary data for the 3-day program also showed significant decreases in

vomiting. Although data collection is still in the initial stages, matching day hospital treatment

approach to the patient’s level of motivation is innovative and warrants future investigation.

The intensity of day hospital treatment required for effectiveness is unknown. Programs with

published outcome data have ranged from 4 or 5 days per week (Olmsted et al. 2003; Williamson et

1. 2001) to 7 days weekly (Gerlinghoff et al. 1998), with no clear trends related to intensity.

Olmsted et al. (2003) compared a 5 day program with a 4-day program in a sequential cohort

design. The two programs were equally effective in promoting weight gain in patients with AN.

However, the 5-day program was associated with higher abstinence rates for bingeing and vomiting

in BN patients and with better psychological functioning at the end of treatment for both diagnostic

subgroups.

Common clinical practice involves treating the most severely ill, usually low-weight patients, with

inpatient treatment, at least initially, and less emaciated AN patients and most BN patients with day

hospital treatment. Within this guideline, there are patients who could start intensive treatment in

either type of program, but little evidence is available to guide this decision. Freeman (1992)

reported on a small unpublished trial in which AN patients were randomly allocated to inpatient or

day hospital treatment. Although the inpatient group gained weight more quickly, there were no

significant differences in outcome at 20-month follow-up. Williamson et al. (2001) compared

outcomes for a mixed sample of eating disorder patients treated initially in either an inpatient or a

day hospital program and found no differences. However, patients were streamed toward inpatient

or day hospital treatment based on their clinical presentation, leading to the conclusion that day

hospital treatment is an efficacious and cost-effective alternative for a subgroup of patients who

might otherwise require inpatient treatment. In another study, Guarda and Heinberg (1999)

presented data from the Johns Hopkins inpatient and day hospital programs, which are sequenced

so that transfer to the day hospital follows an admission to the inpatient unit. Both intensive

treatments were associated with good weight gain, but the rate at which this occurred was faster

with inpatient treatment. Howard et al. (1999) identified prognostic factors associated with the

transfer from inpatient to day hospital treatment. In this study of 59 AN patients treated initially in

an inpatient program, patients who had a longer duration of illness, lower BMI at admission to

inpatient treatment, or lower BMI at transfer to day hospital treatment were less likely to do well

with day hospital treatment. Unfortunately, there was no control or comparison group to clarify

whether these predictors were specific to day hospital treatment or were general indicators of poor

outcome.

Relapse Following Intensive Treatment

There have been eight studies of relapse in AN following inpatient treatment. Based on different

definitions of relapse and various lengths of follow-up, relapse rates ranging from 9% (Strober et

1. 1997) to 42% (Eckert et al. 1995) have been reported. Studies that have examined the timing of

relapse have generally found that the highest risk is during the first year postdischarge (D. B.

Herzog et al. 1999; Isager et al. 1985). Five of the studies followed a series of adult patients (Deter

and Herzog 1994; Eckert et al. 1995; D. B. Herzog et al. 1999; Isager et al. 1985; Touyz and

Beumont 1984). The duration of follow-up ranged from 1 to 11 years, and most had a relatively

small sample size. The average reported rate of relapse across the five studies was 30%. In

addition, there have been two studies of adolescent patients (Martin 1985; Strober et al. 1997).

Both reported a relapse rate of 9%, suggesting that early intervention is associated with a lower

relapse rate. In terms of predictors of relapse, Deter and Herzog (1994) reported that younger age

at presentation, shorter duration of illness, and more severe psychiatric symptoms were associated

with relapse. Strober et al. (1997) reported that excessive exercisers were more likely to relapse.

Carter et al. (2004) studied 51 consecutive first-admission AN patients who were weight-restored

following inpatient treatment. At a median follow-up of 15 months postdischarge, the rate of

relapse (defined as meeting diagnostic criteria for AN) was 35%. Several significant predictors ofPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

4 of 15

10/05/2009 17:33

relapse were identified: history of a suicide attempt, previous specialized treatment for an eating

disorder, severity of obsessive-compulsive symptoms at presentation, excessive exercise

immediately after discharge, and residual concern about shape and weight at discharge.

Few studies of relapse following day hospital treatment have been reported. One 2-year follow-up

study documented a relapse rate of 31% for BN patients. Relapse was predicted by younger age,

higher vomiting frequency, and higher scores on the Eating Attitudes Test at admission to

treatment (Olmsted et al. 1994). Patients who responded rapidly to day hospital treatment and

achieved control of their symptoms in the first 4 weeks of treatment had much lower relapse rates

(16%) than patients who achieved symptom control only toward the end of day hospital treatment

(57%) (Olmsted et al. 1996).

PATIENTS SUITABLE FOR INTENSIVE TREATMENT

Intensive treatments are generally reserved for patients who are low weight, who binge and/or

purge with high frequency (e.g., at least daily), or who have not responded to less intensive forms

of treatment. In practice, a significant portion of these patients have other comorbid conditions,

with anxiety, posttraumatic stress disorder, depression, substance abuse, and personality

disturbance among the most common. Choice of a specific intensive therapy may relate to

geographic and financial access issues, which vary across regions and countries. A continuum of

care is ideal, as it allows patients to start treatment at an intensity suited to their condition and

taper down through progressively less intensive treatments. Patients who start off in a hospital

may graduate to a day hospital partway through treatment, while those in day hospital treatment

may graduate to intensive outpatient follow-up treatment. Intensive treatments are viewed as a

stage of treatment, and subsequent maintenance therapies appear to be critically important.

In comparison with outpatient treatment, intensive treatment provides repeated opportunities for

“in vivo” work on eating, symptom control, affect expression, dysfunctional thoughts,

assertiveness, and interpersonal risk taking. Any acute psychiatric problems such as suicidal intent

or medication problems that may arise can be attended to promptly. With a focus on multimodal

treatment as opposed to supervision or filling time, intensive treatment can provide 30–40 hours of

therapy each week and can help very sick patients make significant progress.

Indications for Full Admission to Hospital

There are three general indications for full admission to a hospital:

Low weight. Patients with AN and very low weight will usually require a period of inpatient treatment.

The American Psychiatric Association Practice Guidelines (American Psychiatric Association 2000)

recommend that inpatient treatment be considered for patients who weigh 75% or less of their healthy

weight and that additional considerations include recent changes in weight or eating, the presence of

additional stressors, and psychiatric problems that warrant hospitalization. Very emaciated individuals

who are eating only small quantities of a restricted range of foods may require 6–8 weeks before they

are able to eat reliably outside of a contained environment.

1.  

Intractable and severe symptoms. Several symptom profiles indicate that 24-hour daily containment

and support may be needed to help the patient change her habitual patterns. These include AN patients

who compulsively exercise, BN patients with uncontrollable purging, and patients with poorly controlled

type 1 diabetes mellitus. The role of a hospital admission may be to interrupt the cycle of symptoms and

allow the patient a fresh start while engaging in intensive treatment. The admission to a hospital may

last 2 or 3 weeks and should be integrated with a full course of intensive treatment, which may be

provided in a day hospital. Brief hospitalization alone is not expected to have long-term efficacy for the

interruption of bingeing, vomiting, or restricted eating. However, there is some evidence for the efficacy

of brief hospitalization as treatment for laxative dependence (Colton et al. 1999).

2.  

Medical complications. Many individuals require hospitalization for medical complications of their eating

disturbance. Such hospitalizations are not generally to an eating disorder program but rather to general

medical or pediatric services. The durations of and specific interventions provided in such admissions

vary by jurisdiction and by the nature of the medical complication. Most such admissions are brief and

3.Print: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

5 of 15

10/05/2009 17:33

are focused on correcting the specific abnormality that prompted the admission rather than on treating

the underlying eating disturbance.

Involuntary Admission

There are occasions where patients are at acute risk of death and yet are unwilling to take any

steps to alter their condition. In such cases, involuntary admission to provide a lifesaving

intervention may be warranted. The precise circumstances under which this is allowed vary from

jurisdiction to jurisdiction. There is some debate as to the efficacy and ethics of involuntary feeding

(Beumont and Carney 2004; Russell 2001).

Involuntary admission should be viewed as a last resort rather than a routine step. It should be

recognized that there is a severe power imbalance in the setting of an involuntary admission and

that the patient must be treated with the maximum of respect and kindness, consistent with the

medical goals of the admission. Clear goals should be set for the involuntary portion of the

treatment, and the involuntary status should be terminated as soon as these goals are met. For

many such patients, an involuntary admission represents management of a life-threatening

emergency rather than treatment that would be provided in an eating disorder treatment program.

Both patients and their families may require help to understand this distinction. Although regular

refeeding with food is always preferred, nasogastric (NG) tube feeding or total parenteral nutrition

(TPN) may be necessary in this context. NG tube feeding and TPN require experienced staff, as they

can feel very invasive to patients, and expertise is needed to administer and manage the medical

problems associated with these methods. NG tubes can cause gastric fullness, diarrhea, and

electrolyte abnormalities. TPN is a much more complicated procedure associated with infections

and metabolic abnormalities. Both require care to monitor for complications of refeeding.

MANAGEMENT OF DISCHARGE FROM FULL HOSPITALIZATION

Gradual Decreases in Containment in Inpatient Programs

Programs vary on how to manage the transition between living in the hospital and living outside

the hospital. This partly depends on models for funding treatment and the availability of inpatient

and day hospital treatment. Ideally, patients should be allowed to experience the maximum number

of opportunities to practice their eating outside the hospital before being discharged. This may

include passes away from the hospital on weekends or weekday evenings, where this is possible for

the patient. If the home environment is stable and supportive, passes to home should be part of any

inpatient program. If the home situation is not stable or if the patient lives at a long distance from

the hospital, alternative arrangements need to be considered. There is limited empirical research

comparing the effectiveness of “strict,” or externally contained, and “lenient,” or less externally

contained, treatment approaches. Touyz et al. (1984) compared “strict” and “lenient” operant

conditioning programs in the inpatient treatment of 65 consecutive patients with AN. There was no

difference between the two programs in rate of weight gain (approximately 1 kg/week), and a

similar proportion of patients in each group reached their target weight. No follow-up data were

reported. Current practice is more consistent with a lenient approach, with a significant emphasis

on patients taking responsibility for behavior change and challenging themselves with practice

opportunities outside the hospital.

Timing of Discharge From Inpatient Treatment

Timing of discharge is a critical but poorly researched area in the treatment of AN. The point at

which AN patients may be safely transferred to a day hospital program is a matter of clinical debate

(Howard et al. 1999). In some settings, patients are partially weight-restored in an inpatient

setting and then transferred to a day hospital setting partway through their treatment. In other

settings patients remain in inpatient treatment (with passes) until they are weight-restored. In any

event, the need for consistency in treatment approach and, if possible, staffing appears to be

important for many patients with AN during the process of weight restoration.

RESIDENTIAL TREATMENTPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

6 of 15

10/05/2009 17:33

Residential treatment programs are common in the United States but have received little empirical

investigation. Residential treatment is an intermediate step between day hospital and inpatient

treatment in terms of level of care (American Psychiatric Association 2000). Patients who do not

require medical intervention or frequent laboratory tests can be provided with the highly structured

environment and 24-hour support that may be necessary for those who are less motivated to

change. As with inpatient treatment, a gradual decrease in the level of containment may be

important for the maintenance of benefits.

TREATMENT OF COMORBIDITY

Most individuals in intensive treatment will experience psychiatric comorbidity of one type or

another. Depression, anxiety disorders (especially obsessive-compulsive disorder and

posttraumatic stress disorder), and substance use are all very common areas of comorbidity

affecting individuals with eating disorders. In a general way, it is appropriate to provide the usual

treatment for such comorbid conditions. For individuals who are underweight or who have very

unstable eating, diagnosing some conditions, such as depression, may be complex, and the

response to treatments such as antidepressant medication may be attenuated. It is useful to

reassess both diagnosis and response to treatment once a patient has been eating fairly regularly

for about 6 weeks.

The management of comorbid substance use and comorbid posttraumatic stress disorder presents

special problems. Both conditions may become activated as the patient begins to eat more normally

and experience fewer eating symptoms. Ideally, treatment for these comorbid conditions should be

integrated into the treatment for the eating disorder. However, some treatment centers do not have

expert resources available to provide contemporaneous treatment, and in these cases treatment

may need to be either staggered or sequenced. This can produce significant difficulties when

patients are unable, for example, to achieve sufficient progress in their posttraumatic stress

disorder to tolerate eating more normally, which in itself may be a requirement to benefit from the

treatment for the posttraumatic stress disorder.

PATIENT MIX

Some intensive programs are devoted exclusively to AN patients, while others mix patients from

the diagnostic subtypes (i.e., AN, BN, binge-eating disorder, and not otherwise specified). The

composition of the patient group often relates to practical issues such as the mandate of the

treatment facility. In group therapy, cohesion is promoted by homogeneity in group members, but

the similarity in therapeutic issues across diagnostic subtypes appears to be sufficient, at least for

highly structured treatment groups. In mixed groups, the more expressive and impulsive BN

patients balance out the more quiet and restrained AN patients. Patients with less common features

(e.g., men or those with type 1 diabetes mellitus) tend to be in the treatment program one at a

time; their ability to integrate with other group members appears to relate more to their individual

personality style than to the less common aspect of their eating disorder. Women who are well

above average weight constitute the subgroup of patients most likely to feel misunderstood by

other group members. Heinberg et al. (2003) reported on a series of adult and adolescent patients

who were treated together in their inpatient and day hospital programs and suggested that mixing

across age groups is a viable option, especially given the paucity of specialized adolescent units. It

is important to note that the treatment provided to adolescents was tailored to meet their needs

and did include a family component. These authors reported that having contact with older, more

chronic patients motivated some of the adolescents to change.

ADVANTAGES OF GROUP THERAPY

Group therapy provides an atmosphere of safety and group support as group members work

together on shared goals and concerns. In intensive treatment, patients are together for 30 to 40

hours weekly, creating an intensive therapeutic arena well beyond the scope of individual therapy.

The therapy group takes on its own identity, and relationships with the therapists are more diffuse.

This provides two strong advantages for work with eating disorder patients. First, the potential for

the development of a power struggle with the therapist is offset by group pressure to work towardPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

7 of 15

10/05/2009 17:33

declared goals. Second, the isolation and shame experienced by many eating disorder patients

dissipate with the knowledge that they are similar to other group members whom they hold in

positive regard.

It is generally possible to include even very underweight individuals in groups as soon as they

enter intensive treatment. Contrary to common wisdom, the cognitive impairment of extreme

starvation tends to lift very rapidly once the patient starts to eat again. There is also often powerful

role modeling and instillation of hope provided by working with individuals further along their path

to recovery. In our highly structured treatment programs in Toronto, we routinely start patients

with BMIs as low as 10 or 11 in groups immediately.

STRUCTURE OF INTENSIVE TREATMENT

Intensive treatment programs require an operational model that provides structure and guidelines

for how time will be spent and how patients and staff will interact with one another. In recent

years, the contents of intensive treatments—the specific psychosocial and nutritional

interventions—have become increasingly similar in inpatient and day hospital settings, so that the

primary differences between the two include the number of treatment hours available and the level

of external containment provided (Zipfel et al. 2002). In this section, we approach the topic of

structuring treatment from the perspective that the content of the treatment is largely independent

of whether patients are living in the hospital or living at home.

Program structures should include a weekly schedule of activities as well as expectations about

attendance, punctuality, and length of stay. The goals of the treatment program and the methods

for working toward those goals should be clearly articulated and fully understood by patients and

staff members. While the goals may be straightforward, diverse methods can be employed,

depending on the underlying philosophy. One philosophical principle relates to setting

“appropriate” expectations; the challenge is to expect enough without expecting so much that

patients are too frightened or overwhelmed to begin. In one study of therapist expectations,

Mitchell et al. (1993) showed that BN patients who are expected to become abstinent early on in

treatment achieve better symptom control than patients who are expected to “do their best.”

However, some treatment providers may feel that it is unrealistic, unfair, or unnecessary to expect

this from patients. Another philosophical principle relates to the balance between program

requirements and patient responsibility. Some patients will aspire to only the minimum required

level of performance. Wherever this line is drawn, some patients may feel that they are being

coerced in an unhelpful manner, while others may (eventually) appreciate the pressure to face a

task that they would not have chosen. Expectations about patients’ behavior during program hours

provide an example. The treatment setting could be identified as a symptom-free zone with the

expectation that patients who are having difficulty turn to staff or other group members for

support. Alternatively, the program could be based on the model that symptoms are an acceptable

coping response at times and their occurrence may be processed therapeutically. There is no

empirical basis for preferring one of these philosophical stances over the other; rather, the need for

a clearly articulated model is paramount.

The model should provide guiding principles for staff, a rationale for their decision making, and the

structure to provide a consistent, predictable environment for patients. Model adherence and

consistent care may be fostered by encouraging staff to function as a team and reserving adequate

team meeting time to allow collaborative decision making.

GOALS

Although there are different ways to operationalize treatment based on a biopsychosocial model of

eating disorders, the goals would generally include 1) medical stabilization as required; 2)

normalization of eating behavior through a balanced meal plan; 3) cessation of bingeing, purging

(i.e., vomiting, laxatives, overexercise), and other unhealthy behaviors used to control weight; 4)

weight gain for patients who are below a healthy weight range (i.e., BMI < 20); 5) therapeutic

exploration of underlying issues and skill development related to affect regulation and

interpersonal relationships; and 6) initiation of a process of comprehensive social and vocationalPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

8 of 15

10/05/2009 17:33

rehabilitation, which is generally not completed during the intensive treatment. Ideally, patients

are able to stop bingeing and purging and/or reach their target weight during intensive treatment

and then have several more weeks to consolidate their new patterns and practice normalized

eating. Follow-up treatment is critical to support the behavioral changes and to continue work on

body image and psychological and vocational rehabilitation. Intensive treatment should be viewed

as one step on a longer journey toward recovery.

Medical Stabilization

The physical safety of the patient is an important priority for any intensive treatment. All patients

require a physical examination. We suggest a complete blood count; liver function tests; a serum

creatinine level; the measurement of electrolytes, calcium, magnesium, and phosphate; and an

electrocardiogram for all patients beginning intensive treatment, given the high frequency of

abnormalities secondary to both malnutrition and purging behavior. Measurement of creatinine

clearance is warranted if serum creatinine is elevated. It is important to correct the serum

creatinine for the weight of the patient, as low weight can produce a falsely normal value.

Assessment of bone mineral density is indicated in all patients with AN who are admitted to

intensive treatment and in individuals with BN who have a history of low weight or amenorrhea for

6 months or longer. The routine performance of radiological examinations, such as computed

tomography or magnetic resonance imaging scans of the brain, is not normally indicated for such

patients.

The role of laboratory tests in detecting covert symptoms deserves some attention. The usual

consequences of purging behavior are acid–base and electrolyte abnormalities (Schulte and Mehler

1999). Vomiting can result in metabolic alkalosis (elevated serum bicarbonate). Laxative use can

result in a metabolic alkalosis or mild acidosis. Diuretics can result in a metabolic alkalosis. All

forms of purging can lead to hypokalemia and, less frequently, hypomagnesemia and hypocalcemia.

However, there is significant variability in the timing and extent to which these physical parameters

change in response to purging behavior. For some patients, serum salivary amylase can be a

sensitive indicator of vomiting, as amylase may be released when the parotid glands are irritated by

gastric acid. The exact mechanism of hyperamylasemia in eating disorder patients is unclear.

However, amylase is also released from the pancreas, and mild pancreatitis is a common

complication of refeeding individuals who are very underweight.

Because there are complete chapters and books devoted to the laboratory abnormalities and

medical complications of eating disorders, an exhaustive review of them will not be possible here.

Physicians who treat eating disorders should be familiar with these potential problems. We will,

however, mention electrocardiogram abnormalities, given the potential morbidity associated with

cardiac problems. Low potassium can cause T-wave flattening, U waves, and ST depression. A

prolonged QT interval can occur with low weight or low serum magnesium and calcium. A variety of

arrhythmias are more likely in the face of a prolonged QT and electrolyte abnormalities. Thus, the

anorexic patient who is purging is at highest risk. Also, a serious medical concern is ipecac, which

even with a relatively small amount of use can lead to cardiomyopathy.

Normalized Eating

Patients should be provided with an individualized balanced meal plan. The expected topography of

meals forms an important part of the treatment model. This should include the degree of variety

required, expectations about including phobic foods (e.g., how many, how soon), role of food

supplements, and methods for establishing calorie levels and changes in calorie levels. Most

programs expect patients to tackle increasingly difficult eating tasks as they progress through

treatment.

Meals within an intensive program are an important component of treatment. There has been a

gradual shift over the past 20 years from having patients in intensive treatment eat alone to eating

as a group or in more public settings. When patients eat together as a group, they feel supported by

engaging in a challenging task together. With guidelines and time limits clearly established inPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

9 of 15

10/05/2009 17:33

advance, staff are able to supervise, support, and coach patients who are having difficulty. Staff can

also encourage light social conversation around the table as part of normalizing eating and as a

means of distraction.

When patients are not in a hospital, either because they are enrolled in a day hospital program or

because they are on a pass from an inpatient program, work toward the goal of normalizing eating

must continue. Considerable group time should be devoted to planning meals outside of the

program, developing strategies to facilitate symptom control and adherence to the meal plan, and

reporting back after evenings and weekends. Patients can be encouraged to share strategies and

offer feedback to one another.

Even patients who are living in the hospital will need to work together outside of scheduled

programming hours to remain free of symptoms and to structure free time. It is generally much

more difficult than it might appear to make a hospital ward a guaranteed “symptom-free zone.”

Patients admitted to the hospital may be allowed some freedom within the institution and will

benefit from planning and reviewing excursions.

Symptom Control

The therapeutic environment is designed to inhibit symptoms. Whereas traditionally this has meant

physical containment, increasingly the importance of psychological containment via support from

staff and co-patients is recognized. This is as true for patients who are living in a hospital as it is for

those who are living at home. Cognitive-behavioral strategies are recommended to control

symptoms outside of program hours. As with normalized eating, considerable group time may be

devoted to working toward symptom control.

Weight Gain

Expectations about rate of weight gain and a protocol for increasing calories form part of the

operational model for treatment. Calories beyond the normal maintenance level may be provided as

food supplements to increase patients’ familiarity with their maintenance meal plan. Activity levels

should be kept to a minimum to promote weight gain. There is a paucity of research on what

constitutes an optimal rate of weight gain for underweight patients during intensive treatment. The

usual considerations relate to the length of treatment versus the likely longer-term outcome after

weight restoration. Weight gain targets of about 1 kg per week are common for inpatient and day

hospital treatment settings. A recent study found that setting lower weekly weight gain

requirements during inpatient treatment was associated with better weight maintenance after

discharge (T. Herzog et al. 2004).

There is some controversy about what represents an appropriate weight target for patients with

AN, and there are no empirical data to support a specific choice. Most clinicians suggest that a

useful guideline is a weight at which normal menstruation occurs, although the resumption of

menses is often delayed in AN patients. Suggested target weights are in the range of 85%–95% of

chart average, or a BMI of 19–21.

Underlying Therapeutic Issues

Underlying issues and stressors that maintain the eating disorder may be identified and processed

in group, family, and individual therapy. The schedule for therapy groups should acknowledge the

central importance of body image concerns as well as relationship issues related to autonomy,

intimacy and caring, past abuse, ambivalence about recovery, and functions served by the eating

disorder.

COGNITIVE-BEHAVIORAL FRAMEWORK

The necessary emphasis on behavioral change makes a cognitive-behavioral framework ideal for

intensive treatment. The following cognitive-behavioral components are easily built into the

structure of an intensive program.

Exposure and Response PreventionPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

10 of 15

10/05/2009 17:33

Intensive treatment is set up for repeated experiences of exposure and response prevention, with

meals arriving at regular intervals and a treatment environment designed to discourage symptoms.

Patients may be exposed to regular feedback about their weight (from the scale), and during

therapy groups they may be exposed to interpersonal conflict, the experience of affect expression,

and other triggers for symptomatic behavior.

Stimulus Control and Self-Control Strategies

Patients can be taught to “smart schedule.” This involves making detailed plans for time outside

the program, whether in the hospital or at home. Plans may include how and where to have meals

or snacks, strategies for avoiding symptoms, and plans to cope with anticipated stressful events or

risky situations. Plans may be reviewed each day, and patients should be encouraged to evaluate

their own and each other’s plans in terms of the expected probability that the plan will lead to

symptom control and adherence to the meal plan. Each morning, the program can begin with

patients sharing their experiences over the evening or weekend. Symptom occurrences should be

viewed as learning opportunities that provide valuable information for subsequent plans.

Graded Task Assignment

From the first day, patients should expect to devote themselves to normalized eating and symptom

control. Nevertheless, some patients will eat normally only during program hours; this is the lowest

level of task difficulty and requires the highest level of support. Over time, patients should be

encouraged to order more phobic foods in the program and on meal outings (i.e., increased task

difficulty) and to “take home” what they have accomplished in the program (i.e., reduce the level of

support). In the patient’s first attempts to normalize eating outside the hospital, having an easy

(less phobic) snack with a friend and staying away from home until bedtime to avoid bingeing may

be viewed as smart scheduling. Later in treatment, the same patient may be able to have a

moderately risky snack at home alone and avoid acting on any subsequent urges. The final goal is

for patients to independently consume any food, even in challenging situations, and abstain from

symptoms. The trick to smart scheduling is to help the patient correctly identify her level of coping

so that she can make plans that match her needs and abilities. Patients often become adept at

helping one another with this.

Psychoeducation

It is helpful to provide formal psychoeducation on topics such as setpoint weight, metabolism,

nutrition, medical complications, autonomy and boundary setting, thinking errors, self-control

strategies, relapse prevention, and assertiveness training. At other times throughout the day, staff

may spontaneously offer psychoeducational information in response to patient experiences or

questions. Group members may also share information based on what they have learned, what they

have experienced, and what has been helpful for them. Peer psychoeducation has a powerful

influence, as it is based on shared challenges and experiences.

Self-Monitoring

Patients may be asked to do intensive self-monitoring. This might take the form of a weekly diary

recording all foods eaten, symptoms and symptom urges, strategies utilized, and feelings. The

self-monitoring diaries are intended to increase patients’ awareness of their eating behavior,

symptom triggers, and coping strategies. It is helpful to have the diaries handed in and read by

staff. A parallel process in the group might involve sharing this information each morning and at a

weekly goal-setting group. Although many patients experience being accountable to the group as a

deterrent to symptomatic behavior outside of program hours, co-group members are usually

sympathetic to difficulties and enthusiastic about accomplishments.

GROUP THERAPY FEATURES

A weekly schedule, a defined focus for group time, and an expectation that patients be punctual,

present, and ready to work toward program goals provide structure and safety for the group. Staff

must take responsibility for adhering to the schedule and for addressing any lack of adherence withPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

11 of 15

10/05/2009 17:33

program expectations. These guidelines set a baseline for group homogeneity, as they require

patients to be committed to the program and to clearly defined goals. In treatment programs with

an open group format, new patients join a preexisting group with members at various stages in

their stay. A group culture has already been established, and new members quickly feel the

pressure to join the group in working toward recovery. The empathic realization that other group

members may find it more difficult if one breaks group norms encourages a feeling of responsibility

and commitment to the group. For some patients, this operates at the level of being reliable in

attendance or working with the group during program hours. For others, it may include challenging

food, body image, or interpersonal risks. Making a commitment to the group about one’s behavior is

usually taken very seriously by group members (e.g., I will have breakfast tomorrow). Thus, the

interpersonal aspects of the group therapy can significantly augment the cognitive-behavioral

structure of the program in promoting change.

Open information equips group members to support and challenge one another in problem areas.

However, it is helpful to keep some details (e.g., actual weights, lowest weights, target weights)

unstated in the group to minimize comparisons among group members. Similarly, comprehensive

information about mental status, medical issues, unusual psychiatric symptoms, or self-harm may

be collected privately. This allows the patient to control how much information she shares with the

group regarding symptoms that may set her apart from others, and it protects group members from

potentially upsetting or provocative details.

ENHANCEMENT OF MOTIVATION IN THE GROUP

Some patients respond immediately to intensive treatment (Olmsted et al. 1996), while others

experience obstacles to engaging with the treatment. Early in treatment, barriers may relate to fear

of change or weight gain or a lack of belief or hope that the treatment will work or that the patient

can do what the treatment requires. Patients may be encouraged to view change as an experiment

and not a permanent commitment. The patient’s right to choose how to spend her life may be

acknowledged directly, along with the importance of conducting a fair experiment over a

reasonable length of time to collect good information relevant to her decision. Data from other

group members’ experiments may bolster the patient’s faith or courage.

It is common for patients to experience difficulties in making progress in intensive treatment. This

is often apparent by the third or fourth week of day hospital treatment and by the sixth or seventh

week of inpatient treatment. Such difficulties might include the inability to adequately reduce

bingeing and purging, complete meals, or reduce exercise. These difficulties should be discussed

directly with the patient in the spirit of understanding what is happening and how she would like to

proceed. Options include the patient withdrawing from the program with the knowledge that she is

welcome to return for a future admission; the patient being discharged and possibly referred to a

more intensive treatment if available (e.g., discharged from day hospital and referred to inpatient

treatment); or the patient deciding that she can commit herself to making specific behavioral

changes at this time. In our experience, well-applied cognitive-behavioral strategies are usually

associated with improvement. Patients who are not able to use the strategies may be overwhelmed

by other stressors in their lives or may not be ready to give up their symptoms. The group can

support a member who is “not doing the program” for a limited time, provided that her struggle is

explicit and her intentions are sincere. However, it is important to maintain the focus on behavioral

change as promised to all group members before admission.

As in individual therapy, it is essential that therapists accept the patient’s beliefs as being genuine

and the patient’s symptoms and interpersonal behavior as serving a function or need in the

patient’s life. Open discussion of the advantages and disadvantages of having specific symptoms,

having an eating disorder, and being in recovery help to create an atmosphere of respect and

understanding for decisions the patients have made in the past and will make in the future.

STAFF ISSUES

Working in an intensive treatment program can be intense and demanding (Kaplan and Garfinkel

1999). Staff are in close regular contact with sick patients, who may be (understandably) underPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

12 of 15

10/05/2009 17:33

stress and irritable and who may direct some of their frustration toward staff. Some patients will

not recover, and some will die from their illness, often after many years of connection with staff

during repeated efforts at treatment. Many will need to address issues that can be emotionally

draining for staff. In order to maintain the therapeutic capacity of the treatment environment, staff

may benefit from frequent opportunities to de-stress, debrief, and refocus. Staff self-care should be

clearly separated from clinical work, so that when staff members are with the patients, they are

able to be fully present and therapeutic. For example, staff members can get some distance from

patient issues by doing paperwork or taking a break, rather than by reading a magazine while

sitting with patients at a meal.

Staff constitute a working group, brought together to accomplish a shared task. Although most

treatment teams have an identified team leader, teams vary in how hierarchical they are. Less

hierarchical teams are more empowering for staff members, which may translate into an

environment in which patients feel safe, contained, and able to work throughout the day, regardless

of which staff members are present at the moment. Informal or assumed roles and relationships

are equally important. These may reflect team members’ perceived levels or areas of competence,

strength or courage, and personality styles, as well as the effect team members have on one

another, to list a few possibilities. It is helpful for the team to have a forum for regular discussion

of how they are functioning together. This may provide an opportunity for staff to articulate how

they would like to work together and to develop strategies for change when needed. The quality of

relating among team members can have a significant impact on the therapeutic environment for

patients; it can also dramatically increase job-related stress for staff—or, conversely, provide them

with immense support and encouragement.

The operational model for treatment may have an equalizing effect for staff, provided that each

team member has a full understanding of the model. The model provides the “authority” for staff to

make certain decisions without consultation. In more complicated situations, a team that functions

nonhierarchically will need to meet as a group for decision making. At times this may feel like a

bureaucratic delay to team members who would like to proceed on their own initiative. Over time,

the treatment model may need to be modified to incorporate new ideas; having a process to

consider changes to the model encourages staff to think about potential improvements and

facilitates short-term adherence to the model in staff members who disagree with some features.

Whenever possible, the team should not include individuals who have fundamental disagreements

with the treatment model. It is especially problematic to have staff members who cannot personally

endorse the type of long-term normalized eating being recommended for patients.

SUMMARY

Intensive therapies are generally effective for patients who are willing to accept this type of

treatment. However, significant refusal, dropout, and relapse rates indicate that intensive

treatment is not a panacea. We have described suitable patients, criteria for admission to a

hospital, issues related to discharge from a hospital, treatment of comorbidity, and structure and

goals of intensive treatment, as well as some cognitive-behavioral aspects, group therapy features,

and motivational enhancement strategies. It is important to view intensive treatment as one step in

the comprehensive treatment of eating disorders. Effective interventions focused on increasing

motivation for recovery and willingness to participate in intensive treatment—along with

maintenance therapies to maintain changes and continue the process of psychological, social, and

vocational rehabilitation—are needed to complete the continuum of care.

REFERENCES

American Psychiatric Association: Practice guideline for the treatment of patients with eating disorders

(revision). Am J Psychiatry 157 (suppl 1):1–39, 2000

Beumont P, Carney T: Can psychiatric terminology be translated into legal regulation? The anorexia nervosa

example. Aust N Z J Psychiatry 38:819–829, 2004 [PubMed]

Bowers WA, Anderson AE: Inpatient treatment of anorexia nervosa: review and recommendations. Harv RevPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

13 of 15

10/05/2009 17:33

Psychiatry 2:193–203, 1994 [PubMed]

Carter JC, Blackmore E, Sutandar-Pinnock K, et al: Relapse in anorexia nervosa: a survival analysis. Psychol

Med 34:671–679, 2004 [PubMed]

Colton P, Woodside DB, Kaplan AS: Laxative withdrawal in eating disorders: treatment protocol and 3- to

20-month follow-up. Int J Eat Disord 25:311–317, 1999 [PubMed]

Crisp AH, Norton K, Gowers S, et al: A controlled study of the effect of therapies aimed at adolescent and

family psychopathology in anorexia nervosa. Br J Psychiatry 159:325–333, 1991 [PubMed]

Dancyger I, Fornari V, Schneider M, et al: Adolescents and eating disorders: an examination of a day

treatment program. Eating and Weight Disorders 8:242–248, 2003 [PubMed]

Deter HC, Herzog W: Anorexia nervosa in a long-term perspective: results of the Heidelberg-Mannheim study.

Psychosom Med 56:20–27, 1994 [PubMed]

Eckert ED, Halmi KA, Marchi P, et al: Ten-year follow-up of anorexia nervosa: clinical course and outcome.

Psychol Med 25:143–156, 1995 [PubMed]

Engel K, Wilfarth B: Therapy results and follow-up of an integrated inpatient treatment for severe cases of

anorexia nervosa. Psychother Psychosom 50:5–14, 1988 [PubMed]

Franzen U, Backmund H, Gerlinghoff M: Day treatment group programme for eating disorders: reasons for

dropout. European Eating Disorders Review 12:153–158, 2004

Freeman C: Day patient treatment for anorexia nervosa. British Review of Bulimia and Anorexia Nervosa

6:3–8, 1992

Gerlinghoff M, Backmund H, Franzen U: Evaluation of day treatment programme for eating disorders. European

Eating Disorders Review 6:96–106, 1998

Gleaves DH, Post G, Eberenz KP, et al: A report of 497 women hospitalized for treatment of bulimia nervosa.

Eating Disorders: The Journal of Treatment and Prevention 1:134–146, 1993

Gowers S, Norton K, Yeldham D, et al: The St. George’s prospective treatment study of anorexia nervosa: a

discussion of methodological problems. Int J Eat Disord 8:445–454, 1989

Gowers S, Norton K, Halek C, et al: Outcome of outpatient psychotherapy in a random allocation treatment

study of anorexia-nervosa. Int J Eat Disord 15:165–177, 1994 [PubMed]

Guarda AS, Heinberg LJ: Effective weight gain in a step-down partial hospitalization program for eating

disorders. Poster presented at the annual meeting of the Academy for Eating Disorders, San Diego, CA, May

1999

Heinberg LJ, Haug NA, Freeman YL, et al: Clinical course and short-term outcome of hospitalized adolescents

with eating disorders: the success of combining adolescents and adults on an eating disorders unit. Eating and

Weight Disorders 8:326–331, 2003 [PubMed]

Herzog DB, Dorer DJ, Keel PK, et al: Recovery and relapse in anorexia nervosa and bulimia nervosa: a 7.5-year

follow-up study. J Am Acad Child Adolesc Psychiatry 38:829–837, 1999 [PubMed]

Herzog T, Zeeck A, Hartmann A, et al: Lower targets for weekly weight gain lead to better results in inpatient

treatment of anorexia nervosa: a pilot study. European Eating Disorders Review 12:164–168, 2004

Howard WT, Evans KK, Quintero-Howard CV, et al: Predictors of success and failure of transition to day

hospital treatment for inpatients with anorexia nervosa. Am J Psychiatry 156:1697–1702, 1999 [Full Text]

[PubMed]

Isager T, Brinch M, Kreiner S, et al: Death and relapse in anorexia nervosa: survival analysis of 151 cases. J

Psychiatry Res 19:515–521, 1985 [PubMed]

Kahn C, Pike KM: In search of predictors of dropout from inpatient treatment for anorexia nervosa. Int J Eat

Disord 30:237–244, 2001 [PubMed]Print: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

14 of 15

10/05/2009 17:33

Kaplan AS, Garfinkel P: Difficulties in treating patients with eating disorders: a review of patient and clinician

variables. Can J Psychiatry 44:665–670, 1999 [PubMed]

Kaplan AS, Olmsted MP, Molleken L: Day treatment of eating disorders. Balliere’s Clinical Psychiatry

3(2):275–289, 1997

Lowe MR, Davis WN, Annunziato RA, et al: Inpatient treatment for eating disorders: outcome at discharge and

3-month follow-up. Eating Behaviors 4:385–397, 2003 [PubMed]

Martin FE: The treatment and outcome of anorexia nervosa in adolescents: a prospective study and five year

follow-up. J Psychiatry Res 19:509–514, 1985 [PubMed]

Mitchell JE, Pyle RL, Pomeroy C, et al: Cognitive-behavioural group psychotherapy of bulimia nervosa:

importance of logistical variables. Int J Eat Disord 14:277–287, 1993 [PubMed]

Morgan HG, Russell GFM: Value of family background and clinical features as predictors of long-term outcome

in anorexia nervosa: four-year follow-up study of 41 patients. Psychol Med 5:355–371, 1975 [PubMed]

Olmsted MP, Kaplan AS, Rockert W: Rate and prediction of relapse in bulimia nervosa. Am J Psychiatry

151:738–743, 1994 [Full Text] [PubMed]

Olmsted MP, Kaplan AS, Rockert W, et al: Rapid responders to intensive treatment of bulimia nervosa. Int J Eat

Disord 19:279–285, 1996 [PubMed]

Olmsted MP, Kaplan AS, Rockert W: Relative efficacy of a 4-day versus a 5-day day hospital program. Int J Eat

Disord 34:441–449, 2003 [PubMed]

Russell GF: Involuntary treatment in anorexia nervosa. Psychiatr Clin North Am 24:337–349, 2001 [PubMed]

Schulte M, Mehler P: Metabolic abnormalities in eating disorders, in Eating Disorders: A Guide to Medical Care

and Complications. Edited by Mehler P, Anderson A. Baltimore, MD, Johns Hopkins University Press, 1999, pp

76–85

Steinhausen HC: Evaluation of inpatient treatment of adolescent anorexic patients. J Psychiatr Res

19:371–375, 1985 [PubMed]

Strober M, Freeman R, Morrell W: The long-term course of severe anorexia nervosa in adolescents: survival

analysis of recovery, relapse and outcome predictors. Int J Eat Disord 22:339–360, 1997 [PubMed]

Surgenor LJ, Maguire S, Beumont PJV: Dropout from inpatient treatment for anorexia nervosa: can risk factors

be identified at point of admission? European Eating Disorders Review 12:94–100, 2004

Thornton CE, Beumont P, Touyz S: The Australian experience of day programs for patients with eating

disorders. Int J Eat Disord 32:1–10, 2002 [PubMed]

Thornton CE, Touyz SW, Beumont PJV, et al: The development of day program treatment of anorexia nervosa

using the stages of change approach, in Focus on Eating Disorder Research. Edited by Swain PI. New York,

Nova Publishing, 2004, pp 171–194

Touyz SW, Beumont PJV: Anorexia nervosa: a follow-up investigation. Med J Australia 141:219–222, 1984

[PubMed]

Touyz SW, Beumont PJV, Glaun D, et al: A comparison of lenient and strict operant conditioning programmes

in refeeding patients with anorexia nervosa. Br J Psychiatry 144:517–520, 1984 [PubMed]

Touyz SW, Beumont PJV, Dunn SM: Behavior therapy in the management of patients with anorexia nervosa.

Psychother Psychosom 48:151–156,1987 [PubMed]

Vandereycken W, Pierloot R: Long-term outcome research in anorexia nervosa: the problem of patient

selection and follow-up duration. Int J Eat Disord 2:237–242, 1983

Williamson DA, Prather RC, Bennett S, et al: An uncontrolled evaluation of inpatient and outpatient cognitive

behavior therapy for bulimia nervosa. Behav Modif 13:340–360, 1989 [PubMed]

Williamson DA, Thaw JM, Varnado-Sullivan PJ: Cost-effectiveness of a hospital-based cognitive behavioralPrint: Chapter 46. Intensive Treatments http://www.psychiatryonline.com/popup.aspx?aID=261793&print=yes…

15 of 15

10/05/2009 17:33

treatment program for eating disorders. Behavior Therapy 32:459–477, 2001

Woodside DB, Carter JC, Blackmore E: Predictors of premature termination of inpatient treatment for anorexia

nervosa. Am J Psychiatry 161:2277–2281, 2004 [Full Text] [PubMed]

Zeeck A, Herzog T, Hartmann A: Day clinic or inpatient care for severe bulimia nervosa? European Eating

Disorders Review 12:79–86, 2004

Zipfel S, Reas DL, Thornton CE, et al: Day hospitalization programs for eating disorders: a systematic review of

the literature. Int J Eat Disord 31:105–117, 2002 [PubMed]

Copyright © 2009 American Psychiatric Publishing, Inc. All Rights Reserved.