28 February 2020 in Psychological Medicine
Background: Psychiatric disorders, including eating disorders (EDs), have clinical outcomes that range widely in severity and chronicity. The ability to predict such outcomes is extremely limited. Machine-learning (ML) approaches that model complexity may optimize the prediction of multifaceted psychiatric behaviors. However, the investigations of many psychiatric concerns have not capitalized on ML to improve prognosis. This study conducted the first comparison of an ML approach (elastic net regularized logistic regression) to traditional regression to longitudinally predict ED outcomes.Methods: Females with heterogeneous ED diagnoses completed demographic and psychiatric assessments at baseline (n= 415) and Year 1 (n= 320) and 2 (n= 277) follow-ups. Elastic net and traditional logistic regression models comprising the same baseline variables were compared in ability to longitudinally predict ED diagnosis, binge eating, compensatory behavior, and underweight BMI at Years 1 and 2.Results: Elastic net models had higher accuracy for all outcomes at Years 1 and 2 [average Area Under the Receiving Operating Characteristics Curve (AUC) = 0.78] compared to logistic regression (average AUC = 0.67). Model performance did not deteriorate when the most important predictor was removed or an alternative ML algorithm (random forests) was applied. Baseline ED (e.g. diagnosis), psychiatric (e.g. hospitalization), and demographic (e.g. ethnicity) characteristics emerged as important predictors in exploratory predictor importance analyses.Conclusions: ML algorithms can enhance the prediction of ED symptoms for 2 years and may identify important risk markers. The superior accuracy of ML for predicting complex outcomes suggests that these approaches may ultimately aid in advancing precision medicine for serious psychiatric disorders.
27 May 2020 in Clinical Psychology Review
1 October 2016 in Journal of Adolescent Health
DNDianne Neumark-Stzainer
1 July 2017 in Journal of Abnormal Psychology
Despite robust support for the role of affect in the maintenance of binge eating and purging, the relationship between affect and restrictive eating remains poorly understood. To investigate the relationship between restrictive eating and affect, ecological momentary assessment data from 118 women with anorexia nervosa (AN) were used to examine trajectories of higher-order dimensions of negative affect (NA) and positive affect (PA), as well as lower-order dimensions of NA (Fear, Guilt) and PA (Joviality, Self-Assurance) relative to restrictive eating. Affect trajectories were modeled before and after restrictive eating episodes and AN subtype was examined as a moderator of these trajectories. Across the sample, Guilt significantly increased before and decreased after restrictive eating episodes. Global NA, Global PA, Fear, Joviality, and Self-Assurance did not vary relative to restrictive eating episodes across the sample. However, significant subtype by trajectory interactions were detected for PA indices. Among individuals with AN restricting subtype, Global PA, Joviality, and Self-Assurance decreased prior to and Self-Assurance increased following restrictive eating episodes. In contrast, Global PA and Self-Assurance increased prior to, but did not change following, restrictive eating episodes among individuals with AN binge eating/purging subtype. Results suggest that dietary restriction may function to mitigate guilt across AN subtypes and to enhance self-assurance among individuals with AN restricting subtype. (PsycINFO Database Record
15 February 2017 in International Journal of Eating Disorders
DNDianne Neumark‐Sztainer
+1
2 January 2018 in Eating Disorders
The relationship between emotion regulation difficulties and restrictive eating has not been established in non-clinical samples. In this study, undergraduates (n = 98) provided information regarding general and specific emotion regulation difficulties on the Difficulties in Emotion Regulation Scale (DERS) and whether they had engaged in recent restrictive eating. Generalized linear models were used to determine if individuals endorsing versus denying recent restrictive eating differed on emotion regulation problems. Results indicated that individuals endorsing restrictive eating had elevated DERS Total (p < .001), Goals (p = .001), Impulse (p < .001), and Strategies (p < .001) scores. Results remained primarily unchanged after controlling for the related construct of dietary restraint. Findings indicate that endorsement of restrictive eating among non-clinical individuals is uniquely associated with emotion regulation deficits, especially those reflecting emotional under-control. Interventions targeting emotion regulation may enhance prevention and treatment of restrictive eating across severity.