Tuesday, February 24, 2015

Working to Better Understand the Effects of Self-Starvation A new questionnaire is designed to identify compulsive food restriction in anorexia nervosa.

Reprinted from Eating Disorders Review
January/February Volume 26, Number 1
©2015 IAEDP
Although self-starvation is the central behavior in anorexia nervosa (AN), that behavior itself has not always generated attention from researchers. The compulsive behavior seen in patients with AN has been compared to compulsive drug-seeking behavior, and the serious consequences and lack of control have clear similarities. Two psychiatrists at Oxford University recently developed a new measure of compulsive self-starvation in AN (Eat Behav. 2015; 17:10). They theorized that using the concept of ‘dependence’ on starvation might be useful to understand compulsive behaviors in eating disorders such as AN.

The Self-Starvation Scale

The scale was developed as an offshoot of the Yale Food Addiction Scale, or YFAS (Appetite. 2009; 52:430), and was redesigned to capture the compulsive element of self-starvation among people with eating disorders and specifically those with AN. The questionnaire includes 17 items as “I have restricted my food intake so much that I have concentrated on this instead of doing other activities.”
Three groups of patients were studied with the Self-Starvation Scale: those with acute AN, (n=41), weight-restored individuals with a history of AN (n=18), and individuals who were fully recovered from AN (n=19). Using the Internet, Drs. Lauren R. Godier and Rebecca J. Park recruited 126 healthy individuals with no prior history of eating disorders and who were between 18 and 65 years of age. A second group included 78 participants between 18 and 65 years of age. 
In addition to the Self-Starvation Scale, the researchers used a set of online questionnaires, including the Eating Disorders Examination Questionnaire (EDE-Q) and the Clinical Impairment Assessment, the Patient Health Questionnaire-9 and the Generalised Anxiety Disorder Assessment-7. The mirror image concept of serotonin addiction-‘food addiction’ was measured with the YFAS. Impulsivity, excessive exercise, and substance abuse symptoms were also measured. Demographic data, including gender, age, weight, and occupation, were gathered, and self-reported height and weight, were gathered to calculate body mass index.

‘Starvation dependence’ may have clinical importance

The scale showed good reliability and criterion-related results. With the scale, the authors were able to predict eating disorders symptoms, use of compulsive exercise, and depression and anxiety in all study participants. The scale was particularly good for predicting these symptoms when the results were compared with the restraint subscale of the EDE-Q. That subscale centers on the rigid control of eating behavior and thus the results are likely to reflect the compulsive and rule-driven nature of self-starvation measured by the SS. The predictive ability was especially good in the group of patients with acute AN. The authors feel this suggests that the test may be particularly useful in clinical samples, in which extreme self-starvation behaviors are likely to be more evident and of potential importance in maintaining the illness. Thus, the concept of ‘starvation dependence’ may be of clinical and theoretical importance among people with eating disorders.

An Interactive Model Studies Binge Eating Among Men A trio of risky traits helped identify increased binge eating among college men.

Reprinted from Eating Disorders Review
January/February Volume 26, Number 1
©2015 IAEDP
The lifetime prevalence of binge eating disorder, or BED, is quite similar among men and women, and subclinical binge eating might be more common in men than in women (Biol Psychiatry. 2007; 61:348). In spite of this, just as with other eating disorders, most BED measures have been designed to study women. 
Dr. Allison M. Minnich and colleagues applied Joiner’s triple interaction model for binge eating in a study of 302 male undergraduates from a large Midwestern University (Eat Behav. 2014; 15:625). The triple interaction model, which examines the effects of the combination of high perfectionism, high body dissatisfaction, and low self-esteem, was developed after Joiner et al. (1997) reported that perfectionism was a risk factor for bulimic symptoms for women who perceived themselves as overweight. 
Dr. Minnich and her team sought to show that the triple interaction model could also be used to predict binge eating among college men. The researchers recruited participants from undergraduate psychology classes at a Midwestern university. The men completed self-report questionnaires on a secure online system at two different time points at least 8 weeks apart (Time 1 and Time 2) and received course credit for their participation. The mean age of the men who completed the assignment was 19.2 years (age range: 18-24 years). 
The study showed that at Time 1, perfectionism, body dissatisfaction, and self-esteem did, in fact, predict binge eating. Interestingly, the three variables were not associated with depression or anxiety. Thus, the model appeared to be specific for disordered eating. This interesting work suggests that similar variables may lead to binge eating in both women and men.

Media Matters to Males, Too In this study, different media affected drive for thinness and muscularity in boys.

Reprinted from Eating Disorders Review
January/February Volume 26, Number 1
©2015 IAEDP
Those who work in the field of eating disorders often stress the impact of media images upon females, but a recent study suggests that media affect males, too. In a study of 182 adolescent boys (average age: 15.2 years), Amy Slater and Marika Tiggemann of the University of the West of England, Bristol, and Flinders University, Sydney, Australia, found that different types of media had differing and significant effects upon the boys (Eat Behav. 2014; 15:679).
According to the authors, one limitation of existing research into the effects of media on teenage boys is that different types of media, including television and magazines, have usually been considered equivalent. However, different types of media may have very different effects on male readers. For example, an earlier researcher found that reading fashion magazines was related to decreased drive for muscularity, while reading health/fitness magazines had the opposite effect increased drive for muscularity (J Soc Clin Psychol. 2005; 24:261). In an earlier study, one of the authors (Dr. Tiggemann) found that when adolescent boys watched TV soap operas, they had increased drive for thinness and decreased muscularity; when they watched music videos, they had an increased drive for muscularity.
The boys in the current study were predominantly white ninth and tenth graders recruited from two coeducational secondary schools in Adelaide, South Australia. The boys had an average body mass index of 21.04 kg/m2. The boys were asked about specific television programs and magazines. Disordered eating attitudes were measured with two scales, the Drive for Thinness subscale of the Eating Disorders Inventory and the Drive for Muscularity scale (Eat Behav 2003; 4:107). The Drive for Muscularity Scale is a 15-item self-report measure designed to assess dysfunctional attitudes and behaviors related to muscularity. 

Reading different magazines and TV programs produced different results

Drs. Slater and Tiggemann’s study was one of the first to evaluate the effects of reality programs and watching these programs. Watching such programs was correlated with drive for thinness among the boys. The major finding was that reading men’s magazines emerged as a unique predictor of both drive for thinness and muscularity.
The authors note that their study had some limitations, such as the fact that the boys were drawn from only two schools, where the student population was relatively homogeneous. The ways in which drive for thinness and muscularity were measured were also relatively simple. In addition, it is plausible that boys who are already concerned about their bodies are more likely to seek out particular types of media (rather than media use leading to body dissatisfaction)

A History of Overweight and Obesity and Development of Restrictive Eating A sizeable number of teens may take drastic weight control measures.

Reprinted from Eating Disorders Review
January/February Volume 26, Number 1
©2015 IAEDP
Prior obesity or overweight is common among treatment-seeking adolescents with restrictive eating disorders. A history of overweight/obesity likely delays case finding, and thus may worsen prognosis for these young patients, according to a team from the Mayo Clinic and the University of Miami Miller School of Medicine (J Adolesc Health. 2015; 56:19).
According to Jocelyn Lebow, PhD, and her colleagues, healthcare professionals are more likely to identify and refer clearly malnourished adolescents than teens who are obese or overweight, and may not recognize that obese adolescents’ attempts at weight loss carry a risk for developing an eating disorder. Obesity itself is a significant risk for the development of eating disorders, and stigma and weight-based teasing may add to the likelihood of disordered eating. One population-based sample of teenaged girls showed that approximately 25% of the obese girls were currently using extreme weight-control behaviors, such as fasting or smoking (Am J Prev Med.2007; 33:359).
The researchers conducted a chart review of consecutive patients 9 to 22 years of age presenting over 6 ½ years for an eating disorder intake evaluation. A body mass index (BMI, kg/m2) history was collected, and teens completed the Eating Disorder Examination Questionnaire (EDE-Q). The results were used to assess dietary restraint, weight, shape, and eating concerns.
Approximately 36% of teens presenting for treatment for a restrictive eating disorder had a weight history above the 85th percentile for age and gender. In this subgroup, 17% had a history of being overweight and a further 19% had a history of obesity. Another finding was that adolescents with a history of overweight or obesity experienced a significantly greater drop in BMI, and the time from the onset of their eating disorder to evaluation took approximately 10 months. While there was no significant difference between the two groups on the EDE-Q subscales for Restraint or Weight Concerns or Eating Concerns, there was a trend toward a difference between the two groups on the Shape Concerns subscale, with the overweight history group having higher scores.
According to the authors, clinicians should remain vigilant for disordered eating in young patients presenting at all BMIs. Early detection of eating disorders will be helped by eliminating the “faulty assumption” that obesity and restrictive eating disorders cannot co-occur.

The High Economic Burden of Care for Patients and Families Economic hardship was reported by 97% of patients and families.

Reprinted from Eating Disorders Review
January/February Volume 26, Number 1
©2015 IAEDP
The economic burden of treatment is one of the many hurdles patients with eating disorders and their families must overcome. A new report from Australia offers new information about the economic burden faced by patients with eating disorders and their families. In Australia, the cost of treating an episode of anorexia nervosa (AN) is second only to the cost for cardiac bypass surgery, and the overall cost of eating disorders, including treatment, lost production and loss to society was $69 billion in 2012. 
While the direct costs for most health care services are either fully or partially underwritten, limited availability of specialist services for treating and managing eating disorders forces many patients and their families to seek private treatment. Families also have economic and social consequences related to changes in education, employment, and to family members who provide necessary but unpaid informal care for patients. Lack of adherence to treatment is also affected by lack of drug coverage, for example, for antidepressants and antipsychotics for comorbid psychological conditions.

The first-ever study of economic burden in Australia

Dr. Lauren Gatt and a team at the George Institute for Global Health, University of Sydney, Australia, used an opt-in invitation and a study questionnaire posted to all persons who had been treated at two hospitals during January 2009 to March 2011, and sent possible participants a link to an online version of the study documents (BMC Psychiatry 2014; 14:338). Patients currently in treatment were also invited to participate and either self-administered a study questionnaire or completed it with the help a researcher. The primary outcome was the household economic burden of eating disorders based on out-of-pocket expenditures on medical and health-related expenses, household economic burden and the related cost of non-adherence to treatment. Hardship was measured using a series of questions about financial stress, such as failure to pay living and medical expenses, and spending more than income by depleting savings or borrowing to finance the expenses.
Ninety persons completed the questionnaires. The majority (54.4%) had self-reported diagnoses of AN (purging or restrictive subtypes), 16.% had bulimia nervosa (BN), and 28.9% were diagnosed with either binge eating disorder (BED) or eating disorder not otherwise specified (EDNOS). Although gender was not a basis for exclusion, 99% of respondents were female.

The economic burden was highest for those with BN

Economic hardship during the prior 12 months affected 97% of study participants, and all groups of patients were affected. The greatest financial stress involved paying for medical appointments (44.5%), followed by dental appointments (44.3%), utility bills (29.5%), and food costs (25%). Most of the families used their savings (61.4%) or sought help from family and friends (45.3%) to pay for treatment. Not adhering to treatment because of cost was most common among those with AN or BN.
Patients with a diagnosis of BN spent twice the amount of out-of-pocket money during the prior 12 months than did those with a diagnosis of AN ($3175 vs $1525, respectively). Hospitalization costs were the greatest for all patients, followed by psychologist and counseling appointments. Overall, families spent 20.7% of their income on treatment for the eating disorder.
The authors point out that unlike the situation with other chronic medical conditions, the current model of care for patients with eating disorders in Australia is outside the primary health care setting and includes a mixture of specialist eating disorder programs, most of which are offered in private settings. Private health care insurance policies will pay for inpatient care but coverage outsider that setting is variable for services in outpatient or day care programs. Thus patients who seek outpatient care tend to have a greater financial burden.

An effort to improve the situation

Dr. Gatt and her colleagues note that the results of their study show that current models of funding for treating eating disorders often limit treatment options beyond admission for severe acute episodes. The study findings may also underrepresent those with poorer outcomes who may be less likely to participate in research. However, there may be good news on the horizon. The Australian National Framework for Eating Disorders has called for development of a sustainable and integrated service model and as part of this a review of Medicare coverage and private health care insurance, to better support patients with eating disorders.

An Alternative Treatment for Anorexia Nervosa: Systemic Family Therapy A treatment option that may help patients with obsessive-compulsive symptoms.

Reprinted from Eating Disorders Review
January/February Volume 26, Number 1
©2015 IAEDP
Family-based therapy (FBT) has become a mainstay in the treatment of anorexia nervosa (AN) in adolescents, and it has substantial research support. On the other hand, FBT clearly does not work for all adolescents, so alternative treatments are certainly needed. Stewart Agras and colleagues recently reported the results of their study comparing FBT to an alternative type of treatment, systemic family therapy (JAMA Psychiatry. 2014; 71:1279). Systemic family therapy views the problems of an individual in relation to the different contexts in which this individual lives: i.e, as a partner in a couple relationship, as a family member, or as a person with particular cultural and/or religious allegiances, while also taking into account socioeconomic circumstances and political processes. 
The study involved 164 male and female adolescents. All participated in 16 one-hour sessions, which were provided over the course of 9 months. Typical eating disorder outcomes were examined, as well as measures of anxiety, obsessionality, and mood.

The results with FBT and systemic family therapy did not differ, either at the end of treatment or at the end of the follow-up period. At the end of treatment, remission rates were 33.1% and 25.3% with FBT and systemic family therapy, respectively. By the end of follow-up, remission rates were 40.7% with FBT and 39.0% with systemic family therapy. An economic analysis of the trial showed that costs per remitted patient were lower with FBT, perhaps due in part to the faster weight gain that was observed in the FBT group, which may have led to shorter hospital stays. Also of note, response to treatment was influenced by obsessive-compulsive symptoms, such that higher levels of obsessive-compulsive symptoms (as measured by the Children's Yale-Brown Obsessive Compulsive Scale were linked to better response to systemic family therapy. 
The results are somewhat surprising. The researchers had hypothesized that FBT would be more effective than systemic family therapy, which seemed a reasonable hypothesis based on prior research. On the other hand, the results are very useful, as they provide an alternative treatment approach for adolescents with AN, and may indicate a particular group (those with high obsessive-compulsive symptoms) who might be particularly good candidates for systemic family therapy.

Refeeding Young Patients with Anorexia Nervosa

By Andrea Garber, PhD, RD, and Daniel Le Grange, PhD
Eating Disorders Program, Departments of Pediatrics and Psychiatry
University of California, San Francisco 
Reprinted from Eating Disorders Review
January/February Volume 26, Number 1
©2015 IAEDP
Adolescents with anorexia nervosa (AN) can rapidly become malnourished from caloric restriction, over-exercising, purging, and other weight-control behaviors. Restoring weight and nutritional status in the hospital or in outpatient settings is a crucial first step toward recovery. 
Adequate weight gain during the early stages of refeeding predicts better long-term recovery.1-4 For medically unstable adolescents, the weight gained during hospitalization or upon hospital discharge (typically 2 to 3 weeks after admission) predicts weight recovery at 1 year.1-3 For those stable enough to begin outpatient psychotherapy, greater weight gain during the first 3 to 4 weeks (1.7 to 1.9 lb/wk) predicts full remission (both weight and cognitive recovery) at 1 year.4 These findings indicating that early weight gain sets the stage for long-term recovery underscore the pressing need to develop approaches to refeeding that safely optimize weight gain in adolescents with AN. 
For decades, the recommendations for refeeding in AN have been quite conservative. For instance, in the U.S., the normal recommendations are to begin refeeding at around 1200 kcal per day, and then to advance slowly by adding about 200 kcal every other day.5-7 In Europe, the UK, and Australia, 8, 9 refeeding regimens starting with as little as 200 to 600 kcal per day have been recommended. These collective Lower Calorie Refeeding (LCR) approaches were intended to ensure patient safety by minimizing risk for the refeeding syndrome. This syndrome is characterized by rapid fluid and electrolyte shifts that can result in severe medical sequelae, including cardiac arrhythmias, cardiac arrest, muscle weakness, hemolytic anemia, delirium, seizures, coma, and even death.10 The refeeding syndrome was first documented in World War II prisoners of war and other starved populations, and is now understood to occur in response to influx of nutrients at the cellular level. 
LCR became the standard of care for refeeding based on clinical consensus that nutrition should be reintroduced as slowly as possible to prevent the refeeding syndrome. Although several studies have now identified severely malnourished patients (<70% of expected weight) as being at the highest risk, 11 LCR has been applied broadly to safeguard all patients with AN. This approach has been generally safe: only a handful of cases of the refeeding syndrome have been reported in patients with AN.8,12-14 However, the results of recent studies indicate that LCR has contributed to the poor weight gain and longer hospital stays that have become expected as a normal part of the course of recovery in AN. For example, we reported that lower starting calorie regimens were associated with initial weight loss, slower weight gain, and longer hospital stays in adolescents with AN followed prospectively over the course of their hospital stay.15 Such findings have contributed to recognition of the so-called “underfeeding syndrome”14 and sparked an interest in Higher Calorie Refeeding (HCR). 

HCR: As Yet, No Standard Definition

Due to insufficient evidence, there is currently no standard definition of HCR or recommendations for implementing it. HCR typically starts with between 1400 to 2400 kcal per day, and may advance anywhere from 0 to 200 kcal per day.16-18These variations are likely to have an impact on the rate of weight gain and may explain inconsistent findings across studies.16, 18 
There are also major differences in methods of delivery. Some programs use meals to deliver higher amounts of calories, 16,17 while others use enteral (tube) feeding alone, 19-21 and still others use a combination of both.22 A handful of studies using meal-based HCR in moderately malnourished adolescents have shown that it is feasible and produces better outcomes in hospitalized patients. Whitelaw et al. (2010) demonstrated satisfactory weight gain and no cases of the refeeding syndrome in 29 adolescents hospitalized with AN, by beginning refeeding with about 1900 kcal per day and increasing this rapidly by approximately 100 kcal per day for the first 5 days.9 In 2013, in a study of 56 adolescents, we reported nearly double the rate of weight gain and almost 6 days shorter hospitalization for the adolescents who were started on an average of 1775 kcal per day, compared to those started on about 1165 kcal per day.17 A concurrent large retrospective study of 310 adolescents by Golden et al. reported 3 days shorter hospital stay in adolescents when refeeding was started at an average of 1550 vs. 1165 kcal per day.17 These findings are quickly being translated into clinical practice despite gaps in the evidence. 

Finding Safe Levels

Safety is an urgent question as clinical practice moves forward with HCR. The justifiable concern that higher levels of nutrients introduced more rapidly would increase risk for the refeeding syndrome has not been substantiated by recent studies of HCR. However, several limitations preclude establishing HCR as a safe approach at this time. First, no study to date has had (and perhaps no future study will have) a sufficiently large sample to examine the full spectrum of clinical features of the relatively rare occurrence of the refeeding syndrome. Studies thus far have focused on low serum phosphorus, the hallmark of the refeeding syndrome, shown to occur in 14% of adolescents with AN across multiple refeeding studies.23 However, the wide variety of electrolyte replacement approaches make it impossible to determine whether this hypophosphatemia occurs earlier or is greater in magnitude in patients on HCR. Another limitation inherent to the available retrospective and observational studies is systematic bias. Physicians may be more likely to prescribe lower calories and/or electrolyte replacement disproportionately to severely malnourished patients, knowing that they are at highest risk.11 These limitations underscore the need for randomized controlled trials (RCTs) to systematically compare and examine the safety of refeeding protocols. 
In addition to the questions of medical safety, it is not known whether there are unintended psychological consequences associated with HCR. Whether increased anxiety related to HCR could negatively impact cognitive recovery is a particular concern, based on several studies demonstrating that mealtime anxiety increases in proportion to caloric load in AN patients.24-27 Increasing caloric loads of the same volume administered to AN patients via tube feeding (such that patients were unaware of the calories ingested) produced increased psychological and physiologic stress, including increases in energy expenditure and plasma cortisol and fear of becoming fat with higher calories. The possibility that increased meal-based anxiety on HCR accumulates to impede long-term cognitive recovery is not supported by the available data showing that greater weight gain in early psychotherapy predicts superior weight and cognitive recovery.28 However, an essential next step is to examine whether the caloric level and/or rate of refeeding (not just amount of weight gain) impacts the course of cognitive recovery. 
The promising outcomes in hospitalized patients make it tempting to assume HCR would produce better long-term weight recovery; however, this has not been examined. In fact, several lines of evidence suggest HCR may not produce better weight gain over the long-term. In the only randomized control trial of refeeding in AN to date, Rigaud et al. compared two groups—one that received higher calories via tube feeding to a lower-calorie meal-based group, and found no difference in weight at 12 months.24
In our 2013 study, discharge weights were equal between HCR and LCR groups due to the shorter hospital stay among patients treated with HCR. Based on this finding, we would anticipate no difference in long-term weight recovery between groups, given that discharge weight is a strong predictor of weight recovery at one year.1Earlier discharge was initially worrisome to clinicians who were accustomed to protracted hospitalizations for these patients. However, a recent large randomized control trial provides strong evidence that there is no benefit to extending hospitalization in adolescents with AN. Madden et al. (2014) compared short-term medical stabilization to longer-term weight recovery in hospitalized patients, both followed by family-based therapy, and found no difference in 12-month weight recovery although patients spent about 20 additional days in the hospital.29Collectively, these findings suggest that the potential benefit of HCR may not be improved weight recovery but less time in the hospital. 

Weighing the Possibility of Relapse with Shorter Hospital Stays

Any benefit of shorter hospitalization associated with HCR could be easily outweighed by an increase in the incidence of relapse. In addition to the historically long initial stay in the hospital, relapse contributes to the recognition of AN as among the top “common and costly” primary mental health diagnoses in pediatrics.30 Forty-three percent of patients require medical re-hospitalization within 12 months of first admission.31 Yet Rigaud et al.’s trial found delayed relapse in the higher-calorie group.20 Although this finding cannot be attributed to differing lengths of stay (the severely malnourished adult participants were hospitalized for equally extended periods of 2 months), there is, nevertheless, no indication that patients refed on HCR would require earlier or more frequent re-hospitalization to supplement the initially shorter stay. 

Considering the Real Cost of AN

If the same weight recovery can indeed be achieved with less time spent in the hospital, this would be a potentially major benefit of HCR, with important implications for the treatment and prognosis of adolescents with AN. Madden et al. (2014) estimated the difference in treatment times between medical stabilization and weight recovery groups amounted to $80,000 in the U.S. healthcare system.29 These sizeable savings do not even begin to cover what is perhaps the larger “cost” of AN — the cost of missed school and work, lost relationships, and decreased quality of life. Future studies should strive to examine these direct and indirect costs to determine the true effectiveness of HCR. However, cost-effectiveness and other potential benefits of HCR cannot be fully realized until the risks are better understood. Future studies are needed to establish the safety and elucidate the possible unintended consequences of HCR so any risks can be weighed against the promise of rapid weight recovery. Only then can clinicians, families, and patients embrace HCR as the new standard of care. 

About the Authors

Dr. Garber is Associate Professor of Pediatrics, Division of Adolescent & Young Adult Medicine, at the Benioff Children's Hospital, University of California, San Francisco.
Dr. Le Grange is Benioff UCSF Professor in Children's Health, Departments of Psychiatry and Pediatrics, University of California, San Francisco. 


  1. Lock J, Litt I. What predicts maintenance of weight for adolescents medically hospitalized for anorexia nervosa? Eat Disord. 2003;11:1.
  2. Lund BC, Hernandez ER, Yates WR, et al. Rate of inpatient weight restoration predicts outcome in anorexia nervosa. Int J Eat Disord. 2009; 42:301.
  3. Baran SA, Weltzin TE, Kaye WH. Low discharge weight and outcome in anorexia nervosa. Am J Psychiatry. 1995;152:1070.
  4. Le Grange D, Accurso EC, Lock J, et al. Early weight gain predicts outcome in two treatments for adolescent anorexia nervosa. Int J Eat Disord. 2014; 47:124.
  5. Practice guideline for the treatment of patients with eating disorders (revision). American Psychiatric Association Work Group on Eating Disorders. Am J Psychiatry. 2000;157 (1 Suppl):1.
  6. Treatment of patients with eating disorders,3rd ed. American Psychiatric Association. Am J Psychiatry. 2006;163(7 Suppl):4.
  7. Position of the American Dietetic Association: Nutrition intervention in the treatment of anorexia nervosa, bulimia nervosa, and other eating disorders. J Am Diet Assoc. 2006;106:2073.
  8. Hall DE, Kahan B, Snitzer J. Delirium associated with hypophosphatemia in a patient with anorexia nervosa. J Adolesc Health. 1994;15:176.
  9. Whitelaw M, Gilbertson H, Lam PY, et al. Does aggressive refeeding in hospitalized adolescents with anorexia nervosa result in increased hypophosphatemia? J Adolesc Health. 2010; 46:577.
  10. Kraft MD, Btaiche IF, Sacks GS. Review of the refeeding syndrome. Nutr Clin Pract. 2005; 20:625. 
  11. Refeeding hypophosphatemia in hospitalized adolescents with anorexia nervosa: a position statement of the society for adolescent health and medicine. J Adolesc Health. 2014; 55:455.
  12. Fisher M, Simpser E, Schneider M. Hypophosphatemia secondary to oral refeeding in anorexia nervosa. Int J Eat Disord. 2000; 28:181.
  13. Kohn MR, Golden NH, Shenker IR. Cardiac arrest and delirium: presentations of the refeeding syndrome in severely malnourished adolescents with anorexia nervosa. J Adolesc Health. 1998; 22:239.
  14. Beumont PJ, Large M. Hypophosphataemia, delirium and cardiac arrhythmia in anorexia nervosa. Med J Aust. 199;155:519.
  15. MARSIPAN Working Group. Management of Really Sick Patients with Anorexia Nervosa. [cited 2012 June 25]; CR 162. Royal College Psychiatrists and Royal College of Physicians London, Oct 2010:[Available from: http://www.rcpsych.ac.uk/files/pdfversion/CR162.pdf]
  16. Garber AK, Michihata N, Hetnal K, et al. A prospective examination of weight gain in hospitalized adolescents with anorexia nervosa on a recommended refeeding protocol. J Adolesc Health. 2012; 50:24.
  17. Golden NH, Keane-Miller C, Sainani KL, et al. Higher caloric intake in hospitalized adolescents with anorexia nervosa is associated with reduced length of stay and no increased rate of refeeding syndrome. J Adolesc Health. 2013; 53(5):573-578.
  18. Garber AK, Mauldin K, Michihata N, et al. Higher calorie diets increase rate of weight gain and shorten hospital stay in hospitalized adolescents with anorexia nervosa. J Adolesc Health. 2013; 53(5):579-584.
  19. Gentile MG, Pastorelli P, Ciceri R, et al. Specialized refeeding treatment for anorexia nervosa patients suffering from extreme undernutrition. Clin Nutr. 2010;5:627.
  20. Rigaud D, Brondel L, Poupard AT, et al. A randomized trial on the efficacy of a 2-month tube feeding regimen in anorexia nervosa: A 1-year follow-up study. Clin Nutr. 2007; 26:421.
  21. Gaudiani JL, Sabel AL, Mascolo M, et al. Severe anorexia nervosa: outcomes from a medical stabilization unit. Int J Eat Disord. 2012; 45:85.
  22. Kohn MR, Madden S, Clarke SD. Refeeding in anorexia nervosa: increased safety and efficiency through understanding the pathophysiology of protein calorie malnutrition. Curr Opin Pediatr. 2011;23:390.
  23. O'Connor G, Nicholls D. Refeeding hypophosphatemia in adolescents with anorexia nervosa: a systematic review. Nutr Clin Prac. 2013;28:358.
  24. Rigaud D, Verges B, Colas-Linhart N, et al. Hormonal and psychological factors linked to the increased thermic effect of food in malnourished fasting anorexia nervosa. J Clin Endocrinol Metab. 2007; 92:1623.
  25. Steinglass JE, Sysko R, Mayer L, et al. Pre-meal anxiety and food intake in anorexia nervosa. Appetite. 2010;55:214.
  26. Van Wymelbeke V, Brondel L, Marcel Brun J, et al. Factors associated with the increase in resting energy expenditure during refeeding in malnourished anorexia nervosa patients. Am J Clin Nutr. 2004;80:1469.
  27. Birmingham CL, Hlynsky J, Whiteside L, et al. Caloric requirement for refeeding inpatients with anorexia nervosa: the contribution of anxiety exercise, and cigarette smoking. Eat Weight Disord. 2005;10:e6.
  28. Le Grange D, Accurso EC, Lock J, et al. Early weight gain predicts outcome in two treatments for adolescent anorexia nervosa. Int J Eat Disord. 2014;47:124.
  29. Madden S, Miskovic-Wheatley J, Wallis A, et al. A randomized controlled trial of in-patient treatment for anorexia nervosa in medically unstable adolescents. Psychol Med. 2014;14:1.
  30. Bardach NS, Coker TR, Zima BT, et al. Common and costly hospitalizations for pediatric mental health disorders. Pediatrics. 2014;133:602.
  31. Steinhausen HC, Grigoroiu-Serbanescu M, Boyadjieva S, et al. Course and predictors of rehospitalization in adolescent anorexia nervosa in a multisite study. Int J Eat Disord. 2008; 41:29.