ABSTRACT
Background: The body mass index is a key predictor of treatment outcome in patients with anorexia nervosa. In adolescents, higher premorbid BMI is a strong predictor of a favorable treatment outcome. It is unclear whether this relationship holds true for adults with anorexia nervosa. Here, we examine adult patients with AN and investigate the lowest and highest lifetime BMI and weight suppression as predisposing factors for treatment outcome. Methods: We included 107 patients aged 17-56 with anorexia nervosa and tracked their BMI from admission to inpatient treatment, through discharge, to follow-up at 1-6 years. Illness history, including lowest and highest lifetime BMI were assessed prior to admission. We used multiple linear regression models with minimal or maximal lifetime BMI or weight suppression at admission as independent variables to predict BMI at admission, discharge and follow-up, while controlling for patients' age, sex, and duration of illness. Results: Low minimal BMI had a negative influence on the weight at admission, which in turn resulted in a lower BMI at discharge. Higher maximal BMI had a substantial positive influence on BMI at discharge and follow-up. Weight suppression was highly correlated with maximal BMI and showed similar effects to maximal BMI. Conclusion: Our findings strongly support a relationship between low minimal lifetime BMI and lower BMI at admission, and between higher maximal lifetime BMI or weight suppression and a positive treatment outcome, even years after discharge. Overall, maximal BMI emerged as the most important factor in predicting the weight course in adults with AN.
ABSTRACT
Impaired decision-making under conditions of uncertainty seems to contribute to the expression and maintenance of anorexia nervosa (AN), but it is not clear whether this impairment is a disease state that would remit with treatment, or a persisting trait in patients with AN. To examine this question, a longitudinal study was conducted in 12 female inpatients with AN (age M = 22.2, SE = 1.36), before (Time-1) and after reaching a body mass index of >17.5 kg/m2 (Time-2). Intolerance of uncertainty (IU) was assessed via a decision-making task, the wheel of fortune (WOF). Weight gain at Time-2 was accompanied with significant changes in uncertainty-related performance compared to Time-1 [(Time × Uncertainty), p < 0.05]. At Time-1, reaction times (RTs) varied in function of uncertainty, while at Time-2, uncertainty did not modulate RTs. These findings support a change in decision-making under uncertainty with successful weight-rehabilitation in AN. While IU was present in underweight patients, it became non-significant after weight restoration.
ABSTRACT
BACKGROUND: The present study determined the psychiatric comorbidity of Axes I and II in a sample of subjects with eating disorders (EDs). The objective was to investigate associations between comorbidity and current and past treatment. METHODS: In a sample of 248 women (77 anorexia nervosa, 137 bulimia nervosa, 34 eating disorders not otherwise specified), psychiatric comorbidity of Axes I and II was determined with the Structured Clinical Interview of DSM-IV. Current and past treatment since ED onset were also assessed. RESULTS: High levels of psychiatric comorbidity were found in the total sample (71% Axis I and 68% Axis II). Only 17% of cases had no psychiatric comorbidity. Anxiety (52%) and affective disorders (50%) were the most common Axis I diagnoses. Personality disorders of Clusters C (52%) and B (23%) were most common for Axis II. Twenty-one percent of participants who were not in treatment at the time had a history of inpatient treatment, and an additional 59% had a history of outpatient treatment. Thirty-eight percent of participants currently in outpatient treatment had a history of inpatient treatment. Participants with multiple comorbidity (Axes I and II) had the highest proportion of cases who had been treated by health professionals. Higher levels of comorbidity were associated with experiences in more intense treatment settings (ranging from no treatment to inpatient treatment). CONCLUSIONS: ED subjects with greater comorbidity require more treatment encounters and more intense treatment settings. The association between comorbidity and treatment experiences may represent a bias in the assessment of comorbidity when samples with heterogeneous treatment history are recruited.
