Rate of Weight Restoration Tied to Outcome in AN

The rate at which weight is restored during inpatient treatment for anorexia nervosa (AN) helps predict treatment outcome, according to Dr. Brian C. Lund and colleagues at the Laureate Institute for Brain Research, Tulsa, OK, and the University of Oklahoma School of Medicine, Oklahoma City (Int J Eat Disord 2009; 42:301). In this study, the rate of weight gain was a significant predictor of clinical outcome after discharge.

Although there is no single standard threshold for weight restoration in terms of body mass index (BMI), most agree that clinical goals include medical stabilization, which may involve every organ system. And, the parameter during weight restoration that is most consistent with post-discharge outcome has been BMI at discharge. This ranges from 15.5 to 19 kg/m2,   according to the authors, suggesting that more extensive weight restoration is always better. Another parameter, rate of weight gain, is not often studied but was a significant predictor of outcome in at least one study (Int J Eat Disord 2004; 36:22).

A study of consecutively admitted patients

Dr. Lund and co-workers recruited female teens and adults from consecutive new patient admissions to the Laureate Eating Disorders Program for an ongoing longitudinal study to examine weight restoration parameters as predictors of clinical deterioration after discharge. These patients selected for the study were required to have a weight restoration goal of at least 8 lb (3.6 kg) over their admission weight, a lifetime DSM-IV-TR diagnosis of AN with or without amenorrhea, and complete discharge and 1-year follow-up information. The authors also used the Clinical Global Impression-Severity Score for the eating disorder during the 1-year follow-up, as well as assessment with the Eating Disorder Inventory-3 and the Structured Clinical Interview for DSM-IV Axis II Disorders.

A total of 79 females with AN were available for analysis, and the mean patient age was 21.6 years (range: 13 to 51 years). The mean duration of eating disorder illness was 4.8 years, but more than half had their eating disorder less than 3 years. Almost half (48%) had no prior hospitalizations for eating disorder treatment; 22% had been hospitalized once; and 30% had two or more prior hospitalizations for eating disorders treatment.

Clinical deterioration during the time from discharge to the 1-year follow-up was seen in 32 (41%) of the women. None of the continuous weight restoration parameters was a significant predictor of clinical deterioration. In contrast, the rate of weight gain was significantly associated with outcome results. Participants’ who gained more than the threshold of 0.8 kg per week were significantly less likely to experience clinical deterioration (21%) than were those who gained weight below the threshold (53%).

Association between discharge BMI and outcome

One of the surprising findings was that there was no significant association between discharge BMI and outcome—this is usually the most consistently reported weight-related predictor of outcome. The authors hypothesize that this pattern might be explained by the weight restoration these patients achieved. The mean discharge BMI was 20.8 kg/m2 and only 13 participants (16.5%) were discharged with BMIs of less than 20 kg/m2. The authors note that the highest previous discharge BMI threshold shown to predict outcome was 19 kg/m2 (Am J Psychiatry 1999; 156:1697).

Dr. Lund and colleagues point to two important findings that have implications for clinical practice.  First, maintaining a weight gain of at least 0.8 kg (1.8 lb) per week during inpatient eating disorder treatment was associated with improved post-treatment course. Second, the lack of association between clinical outcome and BMI at discharge suggests that achieving a discharge BMI of 20 kg/m2 may be an important clinical threshold for optimizing outcome.