Inpatient Refeeding: Adding Science to Art

Methods and goals usually
vary from center to center.

Reprinted from Eating Disorders Review
November/December 2011 Volume 22, Number 6
©2011 Gürze Books

There is as yet no single optimal approach to refeeding underweight eating disorder patients, and refeeding such patients can be described as more art than science. Refeeding methods and goals differ between treatment centers and countries and, according to a group of Australian clinicians, there is little if any scientific evidence to use as an overall guideline. For example, although the American Psychiatric Association recommends a weight gain of 0.1 to 1.4 kg per week for underweight eating disorders patients, there have been few actual descriptions of weight gain during inpatient treatment.

With all this in mind, Dr. Susan Hart and her colleagues at the University of Sydney and the University of Queensland, Australia, examined weight records of female patients admitted to a specialty eating disorders clinic for treatment between January 2000 and December 2006. Patients were categorized as severely underweight (body mass index, or BMI, <14.0 kg/m2 ); moderately underweight (BMI 14.01-17.49 kg/m2); and slightly underweight (BMI 17.5 to 18.99 kg/m2). Those in the moderately underweight category had diagnoses of restricting or purging anorexia nervosa, and those in the slightly underweight category included patients with bulimia nervosa and eating disorder not otherwise specified (EDNOS). The patients included in the analysis were all females because there were too few male patients available to participate (Eur Eat Disorders Rev 2011; 19:390).


All women were placed on the refeeding program of the unit with weight gain goal of 1.0 kg pr week to a minimum BMI of 19.0. All were medically stable prior to admission to the study. Their refeeding program consisted of 3 meals and 3 snacks with no nasogastric or parenteral nutrition. The refeeding program was designed to be an age-appropriate “normal” diet similar to what the patients’ peers might eat, as determined by the dietitian and team. Oral liquid supplements were sometimes included for underweight patients, to meet their energy requirements if they had trouble achieving their weight gain target or could not tolerate the volume of food needed for weight gain. The refeeding regimen for those eating food only was a mean intake of 1980 kilocalories, with 20.5% from protein, 47% from carbohydrate, and 32.5% from fat. For 96 patients, from 1 to 4 liquid meals were added to their meal plans (liquid meals contained 19% of nutrition from protein, 46.5% from carbohydrate, and 34.5% from fat). Each patient had an individual meal plan that was reviewed weekly by an experienced dietitian, and registered nurses supervised all meals (one nurse for 6 to 8 patients). Once progress was seen, the women were given ever-greater autonomy until they were eating without supervision, eating outside of the hospital, and eating at home with their families by the time of discharge.


The final study group included 247 female patients out of 414 admissions admitted for refeeding. For a majority of patients (73%), this was their first admission for inpatient treatment for an eating disorder. The severely underweight group gained significantly more weight between the first admission and discharge from their last admission and significantly more weight; however, it took longer for them to gain their weight. Nearly a third of patients with BMIs less than 17.49 voluntarily included the liquid supplements to help meet weight gain targets, compared with only a single patient in the slightly underweight group.

The authors also reported that there were 4 deaths among patients who had six or more admissions or occasions of care. The patients were not receiving any eating disorder treatment when they died3 died from complications of malnutrition, and a fourth who was in the normal weight range when she was discharged from the clinic. The fourth patient died from unknown causes.

Based on the results of their study, the authors suggest there is a need for more research on identifying patients who best respond to inpatient weight restoration programs. As the authors reported, some patients lose weight or only gain a small amount while hospitalized, as was the case with 88 women in their series. The authors would like to see much more research on identifying the differences between patients who do well in terms of nutritional rehabilitation and weight change and those who do poorly.

No Comments Yet

Comments are closed