Epidemiological data offer conflicting views of the natural course of binge-eating disorder (BED), with large retrospective studies suggesting a protracted course and small prospective studies suggesting a briefer duration. We thus examined changes in BED diagnostic status in a prospective, community-based study that was larger and more representative with respect to sex, age of onset, and body mass index (BMI) than prior multi-year prospective studies.

Probands and relatives with current DSM-IV BED (n = 156) from a family study of BED (‘baseline’) were selected for follow-up at 2.5 and 5 years. Probands were required to have BMI > 25 (women) or >27 (men). Diagnostic interviews and questionnaires were administered at all timepoints.

Of participants with follow-up data (n = 137), 78.1% were female, and 11.7% and 88.3% reported identifying as Black and White, respectively. At baseline, their mean age was 47.2 years, and mean BMI was 36.1. At 2.5 (and 5) years, 61.3% (45.7%), 23.4% (32.6%), and 15.3% (21.7%) of assessed participants exhibited full, sub-threshold, and no BED, respectively. No participants displayed anorexia or bulimia nervosa at follow-up timepoints. Median time to remission (i.e. no BED) exceeded 60 months, and median time to relapse (i.e. sub-threshold or full BED) after remission was 30 months. Two classes of machine learning methods did not consistently outperform random guessing at predicting time to remission from baseline demographic and clinical variables.

Among community-based adults with higher BMI, BED improves with time, but full remission often takes many years, and relapse is common.

Loss of control eating is more likely to occur in the evening and is uniquely associated with distress. No studies have examined the effect of treatment on within-day timing of loss of control eating severity. We examined whether time of day differentially predicted loss of control eating severity at baseline (i.e. pretreatment), end-of-treatment, and 6-month follow-up for individuals with binge-eating disorder (BED), hypothesizing that loss of control eating severity would increase throughout the day pretreatment and that this pattern would be less pronounced following treatment. We explored differential treatment effects of cognitive-behavioral guided self-help (CBTgsh) and Integrative Cognitive-Affective Therapy (ICAT).

Individuals with BED (N = 112) were randomized to receive CBTgsh or ICAT and completed a 1-week ecological momentary assessment protocol at baseline, end-of-treatment, and 6-month follow-up to assess loss of control eating severity. We used multilevel models to assess within-day slope trajectories of loss of control eating severity across assessment periods and treatment type.

Within-day increases in loss of control eating severity were reduced at end-of-treatment and 6-month follow-up relative to baseline. Evening acceleration of loss of control eating severity was greater at 6-month follow-up relative to end-of-treatment. Within-day increases in loss of control severity did not differ between treatments at end-of-treatment; however, evening loss of control severity intensified for individuals who received CBTgsh relative to those who received ICAT at 6-month follow-up.

Findings suggest that treatment reduces evening-shifted loss of control eating severity, and that this effect may be more durable following ICAT relative to CBTgsh.

While negative affect reliably predicts binge eating, it is unknown how this association may decrease or ‘de-couple’ during treatment for binge eating disorder (BED), whether such change is greater in treatments targeting emotion regulation, or how such change predicts outcome. This study utilized multi-wave ecological momentary assessment (EMA) to assess changes in the momentary association between negative affect and subsequent binge-eating symptoms during Integrative Cognitive Affective Therapy (ICAT-BED) and Cognitive Behavior Therapy Guided Self-Help (CBTgsh). It was predicted that there would be stronger de-coupling effects in ICAT-BED compared to CBTgsh given the focus on emotion regulation skills in ICAT-BED and that greater de-coupling would predict outcomes.

Adults with BED were randomized to ICAT-BED or CBTgsh and completed 1-week EMA protocols and the Eating Disorder Examination (EDE) at pre-treatment, end-of-treatment, and 6-month follow-up (final N = 78). De-coupling was operationalized as a change in momentary associations between negative affect and binge-eating symptoms from pre-treatment to end-of-treatment.

There was a significant de-coupling effect at follow-up but not end-of-treatment, and de-coupling did not differ between ICAT-BED and CBTgsh. Less de-coupling was associated with higher end-of-treatment EDE global scores at end-of-treatment and higher binge frequency at follow-up.

Both ICAT-BED and CBTgsh were associated with de-coupling of momentary negative affect and binge-eating symptoms, which in turn relate to cognitive and behavioral treatment outcomes. Future research is warranted to identify differential mechanisms of change across ICAT-BED and CBTgsh. Results also highlight the importance of developing momentary interventions to more effectively de-couple negative affect and binge eating.

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.

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.

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.

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.