Weight gained during treatment predicts 6-month body mass index in a large sample of patients with anorexia nervosa using ensemble machine learning.

OBJECTIVE: This study used machine learning methods to analyze data on treatment outcomes from individuals with anorexia nervosa admitted to a specialized eating disorders treatment program. METHODS: Of 368 individuals with anorexia nervosa (209 adolescents and 159 adults), 160 individuals had data available for a 6-month follow-up analysis. Participants were treated in a 6-day-per-week partial-hospital program. Participants were assessed for eating disorder-specific and non-specific psychopathology. The analyses used established machine learning procedures combined in an ensemble model from support vector machine learning, random forest prediction, and the elastic net regularized regression with an exploration (training; 75%) and confirmation (test; 25%) split of the data. RESULTS: The models predicting body mass index (BMI) at 6-month follow-up explained a 28.6% variance in the training set (n = 120). The model had good performance in predicting 6-month BMI in the test dataset (n = 40), with predicted BMI significantly correlating with actual BMI (r = .51, p = 0.01). The change in BMI from admission to discharge was the most important predictor, strongly correlating with reported BMI at 6-month follow-up (r = .55). Behavioral variables were much less predictive of BMI outcome. Results were similar for z-transformed BMI in the adolescent-only group. Length of stay was most predictive of weight gain in treatment (r = .56) but did not predict longer-term BMI. CONCLUSIONS: This study, using an agnostic ensemble machine learning approach in the largest to-date sample of individuals with anorexia nervosa, suggests that achieving weight gain goals in treatment predicts longer-term weight-related outcomes. Other potential predictors, personality, mood, or eating disorder-specific symptoms were relatively much less predictive. PUBLIC SIGNIFICANCE: The results from this study indicate that the amount of weight gained during treatment predicts BMI 6 months after discharge from a high level of care. This suggests that patients require sufficient time in a higher level of care treatment to meet their specific weight goals and be able to maintain normal weight.

Adverse childhood experiences, low self-esteem, and salient stimulus response in eating disorders.

BACKGROUND: Adverse childhood experiences (ACEs) are elevated in individuals with eating disorders (EDs), but how the neurobiology of EDs and ACEs interact is unclear. METHODS: Women 18-45 years old with anorexia nervosa (AN, n = 38), bulimia nervosa (BN, n = 32), or healthy controls (n = 60) were assessed for ACEs and ED behaviours and performed a taste-conditioning task during brain imaging. Mediation analyses tested relationships between ACE score, self-esteem, and ED behaviours. RESULTS: ACE scores were elevated in EDs and correlated positively with body mass index (p = 0.001), drive for thinness (p = 0.001), and body dissatisfaction (p = 0.032); low self-esteem mediated the relationship between ACEs and body dissatisfaction, drive for thinness, and bulimia severity. ACE scores correlated negatively (FDR-corrected) with unexpected, salient stimulus receipt in AN (substantia nigra) and BN (anterior cingulate, frontal and insular cortex, ventral striatum, and substantia nigra). When ACE scores were included in the model, unexpected stimulus receipt brain response was elevated in EDs in the anterior cingulate and ventral striatum. CONCLUSIONS: ACEs attenuate unexpected salient stimulus receipt response, which may be a biological marker for altered valence or hedonic tone perception in EDs. Low self-esteem mediates the relationships between ACEs and ED behaviours. Adverse childhood experiences should be assessed in biological studies, and their effects targeted in treatment.

Why Don't You Just Eat? Neuroscience and the Enigma of Eating Disorders.

Eating disorders are severe psychiatric illnesses that are associated with high mortality. Research has identified environmental, psychological, and biological risk factors that could contribute to the psychopathology of eating disorders. Nevertheless, the patterns of self-starvation, binge eating, and purging behaviors are difficult to reconcile with the typical mechanisms that regulate appetite, hunger, and satiety. Here, the authors present a neuroscience and human brain imaging-based model to help explain the detrimental and often persistent behavioral patterns seen in individuals with eating disorders and why it is so difficult to overcome them. This model incorporates individual motivations to change eating, fear conditioning, biological adaptations of the brain and body, and the development of a vicious cycle that drives the individual to perpetuate those behaviors. This knowledge helps to explain these illnesses to patients and their families, and to develop more effective treatments, including biological interventions.

Perceptual Dysfunction in Eating Disorders.

Eating disorders (EDs) are characterized by abnormal responses to food and weight-related stimuli and are associated with significant distress, impairment, and poor outcomes. Because many of the cardinal symptoms of EDs involve disturbances in perception of one's body or abnormal affective or cognitive reactions to food intake and how that affects one's size, there has been longstanding interest in characterizing alterations in sensory perception among differing ED diagnostic groups. Within the current review, we aimed to critically assess the existing research on exteroceptive and interoceptive perception and how sensory perception may influence ED behavior. Overall, existing research is most consistent regarding alterations in taste, visual, tactile, and gastric-specific interoceptive processing in EDs, with emerging work indicating elevated respiratory and cardiovascular sensitivity. However, this work is far from conclusive, with most studies unable to speak to the precise etiology of observed perceptual differences in these domains and disentangle these effects from affective and cognitive processes observed within EDs. Further, existing knowledge regarding perceptual disturbances in EDs is limited by heterogeneity in methodology, lack of multimodal assessment protocols, and inconsistent attention to different ED diagnoses. We propose several new avenues for improving neurobiology-informed research on sensory processing to generate actionable knowledge that can inform the development of innovative interventions for these serious disorders.

Examining anxious temperament in anorexia nervosa: Behavioural inhibition and intolerance of uncertainty and their contribution to trait anxiety in adolescents with anorexia nervosa.

BACKGROUND: Anorexia nervosa (AN) is a serious and complex psychiatric disorder yet treatment results are suboptimal. Insight into the etiology of this illness is much needed. Research highlights the implication of anxiety-related traits in the development and maintenance of AN. This study investigates firstly, behavioural inhibition and intolerance for uncertainty (IU) in adolescents with and without AN, and secondly relations between these traits. METHODS: In a cross-sectional study, 165 adolescent girls (AN = 94, HC = 71) completed questionnaires measuring behavioural inhibition, IU and trait anxiety. ANOVAs tested differences between AN and HC groups, and mediation models with IU as a mediator between behavioural inhibition and trait anxiety were run. RESULTS: AN adolescents reported significantly higher levels of behavioural inhibition, IU and trait anxiety compared to their peers. In both AN and HC, a direct and a total effect of behavioural inhibition on trait anxiety was found. However, only in the AN group IU partially mediated the relation between behavioural inhibition and trait anxiety. LIMITATIONS: Data is cross-sectional and longitudinal studies are required. A mean illness duration of nearly 2 years may mean early effects of malnourishment and habituation and future studies should include patients with shorter illness duration. CONCLUSIONS: Results highlight that behavioural inhibition and IU may contribute to anxiety in AN whilst their peers may have developed better executive and social-emotional skills to manage uncertainty. Adolescents with AN may benefit from interventions targeting behavioural inhibition and IU.