Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Feb 1.
Published in final edited form as: Int J Eat Disord. 2022 Nov 29;56(2):428–438. doi: 10.1002/eat.23858

Binge-Eating Disorder With and Without Lifetime Anorexia Nervosa: A Comparison of Sociodemographic and Clinical Features

Pratiksha S Pawar 1, Laura M Thornton 2, Rachael E Flatt 2,3, Christina M Sanzari 4, Emily A Carrino 3, Jenna P Tregarthen 5, Stuart Argue 5, Cynthia M Bulik 2,6,7, Hunna J Watson 2,8,9
PMCID: PMC9904171  NIHMSID: NIHMS1851425  PMID: 36448187

Abstract

Objective:

To compare individuals who have experienced binge-eating disorder (BED) and anorexia nervosa (AN) (BED AN+) to those who have experienced BED and not AN (BED AN−).

Method:

Participants (N = 898) met criteria for lifetime BED and reported current binge eating. Approximately 14% had a lifetime diagnosis of AN. Analyses compared BED AN+ and BED AN− on sociodemographic variables and clinical history.

Results:

The presence of lifetime AN was associated with more severe eating disorder symptoms including earlier onset, more frequent, more chronic, and more types of eating disorder behaviors over the lifetime, and a higher lifetime prevalence of bulimia nervosa (BN). Participants with lifetime AN reported being more likely to have received treatments for BED or BN, had significantly lower minimum, current, and maximum BMIs, more severe general anxiety, and were significantly more likely to be younger and female. In the full sample, the lifetime prevalence of unhealthy weight control behaviors was high and treatment utilization was low, despite an average 15-year history since symptom onset. Gastrointestinal disorders and comorbid anxiety, depression, and attention-deficit/hyperactivity disorder symptoms were prevalent.

Discussion:

Individuals fared poorly on a wide array of domains, yet those with lifetime AN fared considerably more poorly. All patients with BED should be screened for mental health and gastrointestinal comorbidities and offered referral and treatment options.

Keywords: anorexia nervosa, binge-eating disorder, Binge Eating Genetics Initiative (BEGIN), comorbidity, DGBI, gastrointestinal

1. Introduction

Diagnostic crossover is commonly observed in eating disorders (EDs) (Milos et al., 2005; Schaumberg et al., 2019). Between 43% and 62% of individuals with anorexia nervosa restricting type (AN-R) develop binge-eating symptoms (Eddy et al., 2002) and between 21% and 54% of individuals with AN transition to bulimia nervosa (BN) (Castellini et al., 2011; Monteleone et al., 2011). Limited evidence suggests that a smaller percentage, between 0% and 10%, of individuals with AN transition to binge-eating disorder (BED) (Mathisen et al., 2018; Utzinger et al., 2015). Comorbidity, both within discrete periods of time or over the lifetime—such as from AN to BED or vice versa—is important to research because comorbidity can bear negatively on course and prognosis, involve broader and more complex therapeutic approaches, increase severity of illness and psychosocial and medical burden, and reveal developmental paths toward and etiologies of a condition.

Several studies of BN report that a history of AN predicts poorer outcome. Longitudinal studies have shown lower recovery rates, poorer treatment response, and higher odds of subsequent AN episodes in individuals with BN with a history of AN (Eddy et al., 2007; Vaz-Leal et al., 2011). Further, greater hospitalization, depression, substance abuse, purging, and longer duration of illness have been found in those with BN with a history of AN (Bardone-Cone et al., 2008; Eddy et al., 2007; Sullivan et al., 1996; Vaz-Leal et al., 2011). A history of AN also predicts lower BMI in BN (Bardone-Cone et al., 2008; Goldschmidt et al., 2013; Sullivan et al., 1996; Vaz-Leal et al., 2011). In a ten-year follow-up study of mortality in BN, lower minimum BMI and previous suicide attempt were the two significant predictors of death (Huas et al., 2013).

Much less is known about the prognostic significance of lifetime AN in individuals with BED. One study found that patients with BED with a history of AN or BN (AN/BN) had elevated mood disorder prevalence and ED symptoms before and after BED treatment than those without such a history (Utzinger et al., 2015). That study considered AN/BN as a composite variable, and so did not isolate the unique effect of history of AN. In another study, history of multiple EDs (AN, BN, or BED) was associated with a higher prevalence of gastrointestinal disorders than a history of a single ED (Wiklund et al., 2021). The broad literature on comorbidity in psychopathology suggests that comorbidity increases burden of illness, outcomes, severity of the component illnesses, and treatment utilization (Angold et al., 1999).

The clinical phenomenology of comorbid lifetime BED and AN contrasted with BED only is expected to be influenced by the epidemiology and naturalistic course of AN. Whereas BED has a roughly equal gender distribution, AN disproportionately affects females (Udo & Grilo, 2018), and decades-long follow-up studies find a lower weight status in AN than controls even after recovery from AN (Bulik et al., 2000; Fichter et al., 2017). Hence, those who have experienced BED and AN may be more likely to be female and to report lower minimum, current, and maximum BMIs than those who have experienced BED only.

This study compared adults BED and AN in their lifetime (BED AN+) to those with BED in their lifetime only (BED AN−). We compared sociodemographics, history of ED behaviors, treatment history, weight history, and current ED and comorbid symptoms. Based on the literature, we hypothesize that compared with BED AN−, BED AN+ will have a female preponderance, worse ED history (i.e., prevalence, chronicity, frequency, and earlier age at onset of behavioral symptoms), greater treatment utilization for BED, lower minimum, current, and maximum BMIs, and greater current ED symptoms, anxiety, depression, attention-deficit/hyperactivity disorder (ADHD), and gastrointestinal comorbidity. These hypotheses are based on the aforementioned comorbidity and epidemiology research (Angold et al., 1999; Bardone-Cone et al., 2008; Bulik et al., 2000; Eddy et al., 2007; Fichter et al., 2017; Sullivan et al., 1996; Utzinger et al., 2015; Vaz-Leal et al., 2011; Wiklund et al., 2021). We selected anxiety, depression, ADHD, and gastrointestinal comorbidity based on studies in the literature that indicated a higher prevalence of these conditions in people with EDs and because of the emergent focus on the gut microbiome, genomic, and neurobiological mechanisms underlying binge eating and EDs, which is the focus of the parent study from which these data were drawn (Bulik et al., 2020; Bulik et al., 2022). The findings are expected to provide clinicians and researchers with insights into BED that may inform clinical care.

2. Methods

2.1. Participants and Procedure

The study included 898 participants. Participants were from the Binge Eating Genetics Initiative-United States (BEGIN-US) study (Bulik et al., 2020). The goal of BEGIN-US was to recruit 1000 individuals with BN or BED to investigate the behavioral, genomic, and gut microbiota factors that underlie etiology, risk, and course of illness. BEGIN participants were recruited online and through a smartphone app, Recovery Record, which is an ED self-monitoring tool that assists with meal, ED behavior, and emotion tracking. BEGIN-US participants met the following inclusion and exclusion criteria: Inclusion criteria. 1) lifetime Diagnostic and Statistical Manual for Mental Disorders 5th edition criteria (DSM-5) (American Psychiatric Association, 2013) diagnosis of BN or BED with self-reported current binge eating; 2) resident of US; 3) age 18-45 years; 4) reads, speaks English; 5) existing iPhone user with iPhone 5 or later; 6) willing/able to wear Apple Watch for entire study period; 7) willing/able to use Recovery Record for the entire study period; 8) provides informed consent to have activity and self-reported Recovery Record data harvested; and 9) ambulatory (able to walk). Exclusion criteria. 1) currently pregnant or breastfeeding; 2) bariatric surgery due to the impact on eating patterns, including the following: (Roux-en-Y gastric bypass, laparoscopic adjustable gastric banding, sleeve gastrectomy, duodenal switch with biliopancreatic diversion, other gastrointestinal or weight loss surgery); 3) current use of hormone therapy; 4) inpatient treatment or hospitalization for EDs in the 2-weeks prior to study enrollment; 5) suicidal ideation at enrollment; or 6) antibiotic or probiotic use in the past 30 days (related to fecal sampling). For the present study, BEGIN-US participants with a lifetime history of BED were included (n = 904), and of these, six were excluded from this analysis because lifetime history of AN could not be determined. BEGIN participants completed all screening measures, consent, and measures within the Recovery Record app.

BEGIN-US was approved by the Biomedical Institutional Review Board (IRB) at the University of North Carolina at Chapel Hill, and all participants provided informed consent. BEGIN is registered at ClinicalTrials.gov (NCT04162574).

2.2. Measures

2.2.1. Sociodemographic variables

Age, race, ethnicity, and biological sex were measured. Age, race, and ethnicity were self-reported in a survey and race and ethnicity were self-reported according to NIH categories. Sex was ascertained by DNA genotyping and/or survey self-report, as discussed in the Supplement.

2.2.2. ED diagnoses and history of ED behaviors

The Eating Disorders 100,000 Questionnaire version 2 (ED100K.v2) was used to yield lifetime diagnoses of AN, BN, and BED based on DSM-5 criteria (Thornton et al., 2018). The ED100K, developed as part of the Anorexia Nervosa Genetics Initiative (ANGI) to enable rapid, cost-effective recruitment of 100,000 ED cases for large-scale global genetic studies, aims to capture community-recruited individuals with a personal history of EDs. The measure is self-report and web-based, and was developed based on the interviewer-administered Structured Clinical Interview for DSM-IV (SCID) assessment of lifetime ED diagnoses (Eating Disorders Module H). Diagnostic questions from Module H were adapted to a self-report format (Thornton et al., 2018). A key challenge in differential diagnoses among the EDs is the challenge in specifying the time frame of symptoms to ensure that symptoms co-occur at the requisite frequency and duration and are distinct from exclusionary criteria, such as presence of AN for diagnoses of BN and BED and presence of recurrent inappropriate compensatory behaviors or BN for a diagnosis of BED. The ED100K meets these requirements by careful wording and computerized diagnostic algorithms that provide diagnoses according to DSM criteria. The ED100K.v2 has been updated to DSM-5 diagnostic criteria.

In addition to establishing history of ED diagnoses, the ED100K contains questions on the history of ED behaviors, including objective binge eating; self-induced vomiting; laxative, diuretic, diet pill use; excessive exercise; and fasting. With these questions, we formed variables on: i) prevalence; ii) age at onset; iii) chronicity, defined as the duration of an ED behavior for longer than one year; and iv) frequency, defined as whether an ED behavior occurred at least weekly for vomiting, laxatives, diuretics, diet pills, and fasting, and for excessive exercise, every or nearly every day. The ED behaviors were combined to form a composite of “any ED behavior”. Binge eating was not included in the composites for “prevalence” and “frequency” because, by definition, all participants had engaged in binge eating at least once per week. Questions on chronicity began with the question stem, “for how long a period of time did you…” for each ED behavior, followed by the response options: “less than 1 month”, “1 to 2 months”, “3 to 5 months”, “6 to 12 months”, “more than 1 year”, and “don’t know”. Binary yes/no variables indicating whether duration was longer than a year were created from the responses. A duration of longer than a year was chosen a priori for analyses to reflect chronicity of symptoms. The variable distributions were checked for appropriateness before analysis to ensure this coding scheme did not produce small cell sizes in analyses, and thus prevent meaningful comparability. The question stem for the frequency variables was “how often did you usually use…” for each ED behavior, followed by the options “less than once a week”, “at least once a week”, “at least twice a week”, “every day/nearly every day”, and “don’t know”. Binary yes/no variables were created to indicate whether each behavior occurred at least once per week; for exercising excessively, once a week was considered unsuitable for conceptual reasons. “Every day/nearly every day” was chosen since this aligns with thoughts about this construct in previous research where “at least 5 times a week” has been used (Mond & Gorrell, 2021). The distributions of the frequency variables were checked before analysis to make sure the coding scheme did not produce small cell sizes. “Don't know” responses were recoded to missing.

Initial validation evidence supports the validity of the ED100K (Thornton et al., 2018). Random subsets of individuals drawn from samples in two countries completed ED100K and trained interviewers administered the SCID for comparison. The ED100K proved valid at assessing binge eating and also AN diagnosis, according to associations with SCID AN criterion B and C (positive predictive value ranged from .96-1.00 and negative predictive value was 1.00) and the high correlation between interview and self-report responses for lowest illness-related BMI (rs ranged between .91-.92). The validity of binge eating measured by the ED100K was also supported (Thornton et al., 2018).

2.2.3. Treatment and weight

BEGIN participants completed self-report questions on treatment and weight history. Participants were asked, “Have you ever received any of the following treatments for binge-eating disorder or bulimia nervosa? Check all that apply”, with a list which included inpatient treatment, residential treatment, emergency room treatment, cognitive-behavioral treatment (individual or group) (CBT), interpersonal psychotherapy (individual or group) (IPT), other type of psychotherapy, and never received any outpatient treatment for BED or BN. A question asked, “Have you ever received any of the following medications for binge eating? Check all that apply”. Participants were presented with a list of prescribed psychiatric medications used in the treatment of BED and BN, weight-loss agents, the option to select “other medication” and provide information in a free-text comment (see Supplement), and the option to select never having ever taken any medication for binge eating. The list included chemical and brand names of drugs. Examples include selective serotonin reuptake inhibitors (i.e., Prozac, Zoloft), lisdexamfetamine (Vyvanse), topiramate (Topamax), bupropion (Wellbutrin), duloxetine (Cymbalta), and weight loss agents such as phentermine (Adipex), orlistat (Alli), phentermine/topiramate (Qsymia), naltrexone/bupropion (Contrave), and lorcaserin (Belviq). After data collection, responses were reviewed for each participant and binary yes/no variables were created for “psychiatric medication to treat BED or BN” and “prescription weight-loss medication”. Participants reported their current weight, and their lowest and highest (nonpregnancy) weight at their adult height, from which current, minimum, and maximum BMI (kg/m2) were calculated.

2.2.4. Current mental health

Current ED symptoms were assessed with the Eating Disorder Examination-Questionnaire (EDE-Q) v6.0 (Fairburn & Beglin, 2008). Several screening instruments measured current depression, anxiety, and ADHD symptoms: the 9-item Patient Health Questionnaire-9 (PHQ-9) (Kroenke et al., 2001), which is based on the nine criteria for depression in the DSM-IV (American Psychiatric Association, 2000); the 7-item Generalized Anxiety Disorder-7 (GAD-7) (Spitzer et al., 2006), which was developed and validated according to DSM-IV GAD criteria, and the 6-item World Health Organization Adult ADHD Self-Report Scale Screener (ASRS-6) (Kessler et al., 2005), which is consistent with DSM-IV ADHD criteria and was developed in conjunction with revisions of the World Health Organization Composite International Diagnostic Interview. These validated measures have excellent psychometric properties (Johnson et al., 2019; Kessler et al., 2005; Kroenke et al., 2001; Spitzer et al., 2006). The theoretical scores range from 0 to 27 (PHQ-9), 0 to 21 (GAD-7), and 0 to 24 (ASRS-6). These measures had acceptable internal consistency in this study: EDE-Q Restraint (α = .80), Eating Concern (α = .75), Shape Concern (α = .82), Weight Concern (α = .68), and Global (α = .89), PHQ-9 (α = .84), GAD-7 (α = .89), and ASRS-6 (α = .85). The cut-offs for positive screens for major depression (MDD) and generalized anxiety disorder (GAD) are 10 on the PHQ-9 and GAD-7, and for ADHD, four or more checkmarks in the shaded boxes of the ASRS-6 (Kessler et al., 2005; Kroenke et al., 2001; Spitzer et al., 2006).

2.2.5. Disorders of gut-brain interaction (DGBIs)

Current diagnoses of DGBIs (formerly called functional gastrointestinal disorders; FGIDs) were assessed with the Rome III Adult Questionnaire (RIIIAQ). RIIIAQ is consistent with Rome III diagnostic criteria and was developed and validated by the Rome Foundation (Drossman, 2006, 2006). Participants completed the functional bowel disorders section, which assesses irritable bowel syndrome (IBS), functional bloating, functional constipation, functional diarrhea, and unspecified functional bowel disorder diagnoses. A variable summing these diagnoses represented the number of current DGBIs.

2.3. Statistical analysis

Descriptive statistics summarized the full sample and group characteristics. Comparisons between the BED AN+ and BED AN− groups on i) sociodemographic variables, ii) history of BN and ED behaviors, iii) treatment and weight, and iv) current mental health and gastrointestinal comorbidity (DGBIs) involved analysis of variance (ANOVA), chi square test of independence, and Fisher’s exact test. The false discovery rate (FDR) procedure was applied to reduce the probability of Type I errors due to multiple testing (Benjamini & Hochberg, 1995). Effect sizes quantified the magnitude of the group differences (e.g., Cohen’s d, phi (φ)) (Cohen, 1988). Group differences were interpreted if FDR P < .05. Uncorrected P values are also presented for completeness.

Several participants did not complete these psychometric measures: EDE-Q (11%), PHQ-9 (7%), GAD-7 (7%), ASRS-6 (7%), and RIIIAQ (7%). Seven further variables had a non-negligible amount of missing data (i.e., >5%): six had 5-9% missingness and one had 14% missingness. There was no specific pattern to these missing values. Multiple imputation handled missing data and the results were pooled according to Rubin (1987) and Schafer (1997). Sensitivity analyses were conducted to rule out lifetime BN as a confound, to compare methods for handling missing data, and to address an acute diagnosis of BED on core and comorbid symptoms at the time of the study. The key findings were unchanged, as reported in the Supplement.

3. Results

3.1. Participants

The full sample had 898 participants, of whom 14% had a lifetime AN diagnosis (BED AN+: n = 126) and 86% had no lifetime AN diagnosis (BED AN−, n = 772). In the BED AN+ group, 79% reported binge eating or vomiting, laxative use, or diuretic use for weight control while ill with AN, suggesting a lifetime prevalence of 11% for AN binge-purge subtype (AN-B/P) and 3% of AN-R in the full BED sample.

3.2. Sociodemographic characteristics

The mean age was 29.67 years (SD = 7.18) and sex was mostly female (87%), otherwise male (13%) or missing (0.3%). Most participants identified as White (86%) and non-Hispanic (89%). Non-White participants reported their race as African American, Asian, Native American or mixed race (breakdowns of these race categories are suppressed due to small numbers to comply with regulatory frameworks on protection of personal health information). BED AN+ participants were significantly more likely to be younger (27.87 ± 7.14 v 29.97 ± 7.15 yrs, F(1,896) = 9.31, FDR P = .01, Cohen’s d = −0.15) and female (95% v 86%, χ2(1) = 9.74, FDR P = .01, φ = .10) than BED AN− participants. There were no group differences on race (Fisher’s exact test FDR P = .49) or ethnicity (χ2(1) = 4.19, FDR P = .06, φ = .07).

3.3. History of BN and ED behaviors

BED AN+ and BED AN− were compared on history of BN and ED behaviors as shown in Table 1 and Figure 1. Supplementary Table 1 contains the sample descriptives and detailed analysis results for the findings in Figure 1. A lifetime diagnosis of BN was common in BED AN+ and BED AN−, and significantly more common in BED AN+. Most participants reported engaging in ED behaviors besides binge eating at some point in their life (92%), particularly fasting and excessive exercise (Table S1). BED AN+ were significantly more likely to report excessive exercise, vomiting, laxative use, diuretic use and younger onset of ED behaviors compared to BED AN− (φs: .08–.16). Many reported chronic (i.e., lasting longer than 1 year) and frequent (occurred at least once per week at some time(s) in their lives, and for excessive exercise, every or nearly every day) ED behaviors, although a significantly higher proportion in BED AN+ (Figure 1B-C). Effect sizes on the chronicity and frequency variables ranged from φ = .04–.24 and φ = .03–.23, respectively. There was no significant difference in binge eating age at onset between BED AN+ and BED AN−. Participants were asked whether they had ever used vomiting, laxatives, diuretics, diet pills, excessive exercise, or fasting to compensate for binge eating. The majority of participants had used compensatory behaviors (87%). The BED AN+ group was significantly more likely than the BED AN− group to have compensated for binge eating with vomiting, laxative use, diuretic use, excessive exercise, and any ED behavior (φs: .09–.18) (Figure 1D).

Table 1.

Comparisons Between the Lifetime Binge-Eating Disorder with Lifetime Anorexia Nervosa (BED AN+; n = 126) and Lifetime Binge-Eating Disorder Only (BED AN−; n = 772) Groups

Variable Full sample
(n = 898)		 BED AN+
(n = 126)		 BED AN−
(n = 772)		 Test Test statistics P FDR P Effect size
Lifetime BN diagnosis 743 (83%) 116 (92%) 627 (82%) Chi square χ2(1)=7.87 .01** .01* φ = .09
Age at onset of binge eating 17.76 (6.93) 18.09 (5.84) 17.70 (7.12) ANOVA F(1,896)=0.16 .69 .73 d = 0.06
Age at onset of any ED behavior 15.75 (5.92) 14.48 (3.87) 15.96 (6.19) ANOVA F(1,896)=5.55 .02* .04* d = 0.25
History of treatment (“Have you ever received any of the following treatments for BED or BN?”)
 Residential 68 (8%) 27 (21%) 41 (5%) Chi square χ2(1)=40.21 <.001*** <.001*** φ = .21
 Inpatient 67 (7%) 27 (21%) 40 (5%) Chi square χ2(1)=41.41 <.001*** <.001*** φ = .21
 Emergency 28 (3%) 15 (12%) 13 (2%) Chi square χ2(1)=37.46 <.001*** <.001*** φ = .20
 CBT 346 (39%) 69 (55%) 277 (36%) Chi square χ2(1)=16.30 <.001*** <.001*** φ = .13
 IPT 241 (27%) 46 (37%) 195 (25%) Chi square χ2(1)=6.98 .01** .02* φ = .09
 Other therapy 33 (4%) 9 (7%) 24 (3%) Chi square χ2(1)=4.98 .03* .05 φ = .07
 Psychiatric medication 393 (44%) 70 (56%) 74 (10%) Chi square χ2(1)=8.28 .004** .01* φ = .10
 Any of the above 546 (62%) 88 (72%) 458 (60%) Chi square χ2(1)=5.50 .02* .04* φ = .08
Current BMI (kg/m2) 33.58 (9.93) 26.31 (7.53) 34.77 (9.78) ANOVA F(1,896)=85.87 <.001*** <.001*** d = 0.89
Lowest adult BMI (kg/m2) 24.02 (6.57) 16.99 (1.95) 25.17 (6.34) ANOVA F(1,896)=206.75 <.001*** <.001*** d = 1.38
Highest adult BMI (kg/m2) 36.53 (10.26) 29.22 (7.93) 37.72 (10.11) ANOVA F(1,896)=77.46 <.001*** <.001*** d = 0.86
Prescription weight-loss medication 92 (10%) 18 (14%) 74 (10%) Chi square χ2 (1)=2.60 .11 .15 φ = .05
EDE-Q binge eating 13.24 (9.55) 12.17 (9.55) 13.41 (9.54) ANOVA F(1,896)=1.67 .20 .25 d = 0.13
EDE-Q restraint 2.78 (1.55) 3.05 (1.58) 2.74 (1.54) ANOVA F(1,896)=3.75 .05 .08 d = 0.20
EDE-Q eating concern 3.68 (1.26) 3.79 (1.31) 3.66 (1.25) ANOVA F(1,896)=0.51 .48 .53 d = 0.10
EDE-Q shape concern 4.71 (1.12) 4.72 (1.27) 4.71 (1.09) ANOVA F(1,896)=0.08 .78 .80 d = 0.01
EDE-Q weight concern 4.53 (1.08) 4.43 (1.28) 4.54 (1.05) ANOVA F(1,896)=1.33 .25 .31 d = 0.10
EDE-Q global 3.93 (1.00) 4.00 (1.13) 3.91 (0.98) ANOVA F(1,896)=0.32 .57 .61 d = 0.09
PHQ-9 score 11.82 (5.29) 12.64 (5.67) 11.68 (5.22) ANOVA F(1,896)=2.66 .10 .14 d = 0.18
PHQ-9 MDD screen 548 (65%) 86 (72%) 462 (64%) Chi square χ2(1)=2.25 .13 .17 φ = .05
GAD-7 score 9.52 (5.34) 11.04 (5.37) 9.27 (5.29) ANOVA F(1,896)=9.25 .002** .01* d = 0.33
GAD-7 GAD screen 396 (47%) 69 (57%) 327 (46%) Chi square F(1,896)=4.46 .03* .05 φ = .10
ASRS-6 score 12.19 (5.76) 13.09 (6.16) 12.03 (5.68) ANOVA F(1,896)=3.49 .06 .09 d = 0.18
ASRS-6 ADHD screen 427 (51%) 65 (54%) 362 (50%) Chi square χ2(1)=0.51 .48 .53 φ = .02
Rome III IBS 257 (31%) 43 (36%) 214 (30%) Chi square χ2(1)=1.95 .16 .21 φ = .05
Rome III functional bloating 233 (28%) 41 (24%) 192 (27%) Chi square χ2(1)=2.84 .09 .13 φ = .06
Rome III functional constipation 50 (6%) 9 (7%) 41 (6%) Chi square χ2(1)=0.39 .53 .57 φ = .02
Rome III functional diarrhea 15 (2%) a a Fisher's exact n.a. .77 .80 a
Rome III unspecified functional bowel disorder 614 (73%) 89 (74%) 525 (73%) Chi square χ2(1)=0.03 .86 .87 φ = .01
Number of DGBIs 1.39 (0.81) 1.53 (0.77) 1.37 (0.81) ANOVA F(1,896)=3.76 .05 .08 d = 0.20
Open in a new tab

Note. ANOVA = analysis of variance. ASRS-6 = Adult ADHD Self-Report Scale Screener. BMI = body mass index. BN = bulimia nervosa. CBT = cognitive-behavioral treatment. DGBI = disorder of gut-brain interaction. ED = eating disorder. EDE-Q = Eating Disorder Examination-Questionnaire. FDR = false discovery rate. GAD-7 = Generalized Anxiety Disorder-7. IBS = irritable bowel syndrome. IPT = interpersonal psychotherapy. PHQ-9 = Patient Health Questionnaire-9.

*

P < .05.

**

P < .01.

***

P < .001. Values of d = 0.2 and φ = .1 are considered to be small effects, d = 0.5 and φ = .3 medium effects, and d = 0.8 and φ = .5 large effects.

a

Small cell sizes are suppressed to comply with regulations on health data privacy.

Figure 1. Comparison of the history and course of eating disorder behaviors in the lifetime binge-eating disorder with lifetime anorexia nervosa (BED AN+; n = 126) and lifetime binge-eating disorder only (BED AN−; n = 772) groups.

Figure 1.

Open in a new tab

(A) Comparison of unhealthy weight control behaviors in the full sample. (B) Comparison of the chronicity of eating disorder behaviors in the full sample. (C) Comparison of the frequency of eating disorder behaviors in the full sample. (D) Comparison of the prevalence of binge-eating compensatory behaviors in the full sample. (E) Comparison of the chronicity of eating disorder behaviors in the subsample analyses: subsample analyses only included individuals who endorsed ever using that behavior. (F) Comparison of the frequency of eating disorder behaviors in the subsample analyses: subsample analyses only included individuals who endorsed ever using that behavior. Eating disorder behaviors were measured with the ED100K. In each figure, “any” is a composite of vomiting, laxatives, diuretics, diet pills, excessive exercise, and fasting, and for B and E also includes binge eating. For exercise, the frequency plotted in C and F is “every day/nearly every day”; for all other behaviors it is at least once a week. Summary statistics, detailed analysis results, and subsample Ns can be found in Supplementary Table 1. * p < .05. ** p < .01. *** p < .001.

In further analyses, only individuals with a lifetime history of each specific ED behavior were analyzed. AN was still associated with significantly more frequent and chronic ED behaviors (Figure 1E-F). Effect sizes for the significant findings ranged from φ = .14–.23 for the frequency and φ = .10–.20 for the chronicity comparisons. These analyses addressed whether BED-AN+ demonstrated greater chronicity and frequency simply due to these behaviors being more common in the BED-AN+ sample, and therefore greater opportunity to observe more chronic and frequent behaviors (i.e., a confound). The results established that BED-AN+ participants had greater chronicity and frequency of these behaviors in their lifetime compared with BED AN−, irrespectively, thus supporting the hypothesis that comorbidity is associated with greater burden.

3.4. History of ED treatment, weight, and prescription weight-loss treatment

As shown in Table 1, 62% of the full sample reported being treated for BED or BN. Supporting our hypothesis, those in the BED AN+ group reported significantly greater treatment utilization for their binge-eating problem compared to those in the BED AN− group, more specifically, significantly higher utilization of psychiatric medication, CBT, IPT, residential, inpatient treatment, and emergency room services.

The majority in the full sample had a BMI ≥ 25 (79%) and 20% had a BMI of 18.5 to 24.9. Supporting our hypothesis, BED AN+ had a significantly lower current BMI, minimum adult BMI, and maximum adult BMI compared to BED AN−. There were no significant group differences in the percentage who had received prescription weight-loss treatment.

3.5. Current mental health

Descriptive statistics and group comparisons on ED, depression, anxiety, and ADHD symptoms at the time of the study can be found in Table 1. The full sample EDE-Q Global M ± SD score was 3.9 ± 1.0 and the EDE-Q frequency of binge eating over the past 28 days was 13.2 ± 9.6. These scores are comparable to clinical samples, including a treatment-seeking sample of 86 people with BED (EDE-Q Global, 3.5 ± 0.8, EDE-Q frequency of binge eating 13.7 ± 11.7) (Aardoom et al., 2012; Reas et al., 2006). GAD-7 was significantly higher in the BED AN+ group compared to the BED AN− group, in support of our hypothesis of a higher burden of illness in participants with lifetime AN. There were no statistically significant differences between the groups on the other measures. A sensitivity analysis incorporating current BED diagnostic status as a covariate produced the same findings (Supplementary section 3.3). A supplementary analysis comparing current symptoms among those with current BED did not reveal any statistically significant differences and effect sizes were small to nil (Supplementary section 4, Supplementary Table 4).

3.6. DGBIs

In the full sample, DGBI screening-based diagnoses were common, particularly IBS, functional bloating, and unspecified functional bowel disorder (Table 3). Cohen’s d indicated a higher mean number of DGBIs in the BED AN+ group compared to the BED AN− group.

4. Discussion

We compared adults with lifetime BED with and without lifetime AN to characterize and compare the clinical phenomenology. Due to the recruitment criteria, all participants endorsed current binge eating. Both groups reported substantial clinical burden over the average 15 years between reported symptom onset and study recruitment; however, a diagnosis of AN was broadly associated with more pathology.

The BED AN+ group had a younger onset of ED behaviors, broader range of ED symptoms, and were more likely to have frequent and chronic ED behaviors and a lifetime BN diagnosis, and worse general anxiety compared to the BED AN− group. These findings align with research demonstrating poorer outcomes for people with BN with a history of AN (Eddy et al., 2007; Vaz-Leal et al., 2011) and evidence that comorbidity, broadly-speaking, increases burden of illness (Angold et al., 1999). Unlike anxiety, other symptoms including eating pathology, depression, ADHD, and gastrointestinal problems were not significantly more severe in the comorbid group. This suggests that the absence of lifetime AN comorbidity does not diminish impairment or association in these areas, which were highly impacted across the whole sample. Since those findings were not in line with expectations, future studies with other samples will be helpful for clarifying the association between lifetime AN comorbidity and illness burden. Group differences hypothesized from the epidemiological and clinical profile of AN were supported (Bulik et al., 2000; Fichter et al., 2017; Udo & Grilo, 2018). Those in the BED AN+ group were more likely to be female with lower current, minimum, and maximum BMIs, and to have a younger onset of ED behaviors and higher prevalence of weight control and compensatory behaviors in their clinical history.

Several findings have clinical implications. Many participants screened positive for GAD, MDD, and gastrointestinal disorders. Screening for major mental health comorbidities in all ED patients at clinical assessment is therefore recommended. As many patients commonly report that negative moods drive binges, mental health comorbidities are an essential component of case conceptualization (Dingemans et al., 2017). The triad of EDs, anxiety, and DGBI is becoming increasingly recognized as a potential mechanism of risk and maintenance of EDs (Bulik et al., 2021; Peters et al., 2021; Wiklund et al., 2021; Zucker & Bulik, 2020). Our findings encourage referral for management of comorbid gastrointestinal conditions in order to improve health outcomes for ED patients. The prevalence of IBS of 31% and unspecified functional bowel disorder of 73% in our sample are relatively higher than the 9% (95% CI: 7-10%) and 10% (95% CI: 9-11%) prevalence, respectively, among adults in the general population in the United States (Palsson et al., 2020). It remains to be seen whether the higher prevalence of DGBIs is specific to EDs or common to psychological disorders.

Both groups reported near universal prevalence of ED behaviors besides binge eating (i.e., vomiting, laxative use, diuretic use, excessive exercise, and fasting) in their lifetime clinical history. These behaviors are sometimes referred to as unhealthy weight control behaviors (UWCB). Given that the hallmark of BED is a prolonged period of binge eating in the absence of the regular use of UWCB, this finding is notable. There is a large theoretical and empirical literature implicating restraint as a developmental precursor to binge eating, but how emergence of UWCB relates to onset of BED is unclear. A recent prospective study indicated that recurrent UWCB predicted future onset of clinical EDs, including BED marginally (OR = 1.17; 95% 0.99-1.35) (Stice et al., 2021). In over a third of individuals seeking treatment for BED, UWCB had predated binge eating (Stice et al., 2021). Similarly, in our study, UWCB emerged an average two years prior to binge eating and during adolescence. Identifying and intervening with adolescents who engage in UWCB may help to prevent the onset of future EDs, including BED. A clinical implication is that if there is a history of such behaviors in BED patients, clinicians may want to incorporate these features into case formulation and address them in relapse prevention. Our clinical experience with patients with BED is that within their clinical history some extensively ruminate on these behaviors and then abandon them after early attempts, some regularly engage in these behaviors outside BED episodes, and some engage in them occasionally.

The 14% prevalence of lifetime AN in our BED population is higher than the 1%-10% previously reported in treatment-seeking BED samples (Mathisen et al., 2018; Utzinger et al., 2015). This finding could be due to differences in diagnostic assessment, for instance, Utzinger et al. used a clinical diagnostic interview for diagnosing EDs. Mathisen et al. used a similar approach to us of screening for DSM-5 criteria using a self-report questionnaire. Longitudinal studies have found very low crossover between AN and BED in either direction (0%-2%) while actively following research participants over an average 6 to 8 years (Castellini et al., 2011; Schaumberg et al., 2019; Stice et al., 2013). The lower rate in these studies is due to the more limited time frame and reliance on assessment of current diagnosis at each wave, whereas the aforementioned studies measured lifetime history. Longitudinal epidemiological samples using well-validated diagnostic methods are needed to fully understand the prevalence of comorbid lifetime BED and AN, sequence of onset, and risk factors for diagnostic crossover.

In this study, individuals with lifetime AN endorsed greater treatment utilization. The questions regarding treatment utilization captured treatment received for either BN or BED (BN/BED). The difference remained apparent in sensitivity analyses that addressed BN as a confound, suggesting that something about lifetime AN history might influence treatment utilization. The perceived severity of AN, or the experience of AN treatment may lead patients, family, and health care providers to be more likely to engage in or recommend treatment among these BED patients. Treatment for AN-B/P and treatment for BN may also have been conflated by participants because clinically, many patients have difficulty distinguishing between these diagnoses, and thus treatment for AN-B/P may have been reported as treatment for BN. Recognizing all types of EDs is crucial for primary care providers and screening for EDs facilitates access to care (Citrome, 2017). Patients may not disclose their BED symptoms because of embarrassment or may lack knowledge that they have a mental health condition that is treatable (Citrome, 2017). Participants in this study, who were all community-recruited, reported an average age at onset of binge eating 12 years prior to the study and 2 in every 5 people had never been treated for BN/BED.

Future research in other samples and comparisons of more diverse factors would be helpful. Although we compared a large range of variables, they were focused in nature. AN and BED lifetime comorbidity is an intriguing clinical presentation. Some with AN are never able to override restrictive eating behaviors, and become chronically unwell and at risk of starvation-related death. The comorbidity pattern of AN and BED involves extreme nutrition transitions, and points to “switches” for turning up and down self-starvation and control over eating. Such mechanisms might reside in novel biological, genomic, or behavioral factors that inform treatment development. Further investigation into the factors associated with AN and BED comorbidity might also yield insight into the developmental course and etiology of particular presentations of BED.

There were several study limitations. There was a reliance on self-report instruments and a clinician-administered diagnostic interview was not used. Evidence that self-report instruments can produce reliable and valid diagnoses has largely been restricted to current ED diagnosis, and further validation of the ED100K to capture lifetime diagnoses is warranted. We had no data to determine if BED occurred before or after AN, although AN is reported to predate BED in most instances (Castellini et al., 2011; Stice et al., 2013; Welch et al., 2016). As recruitment involved community-based methods, the findings can be generalized to individuals in the community who have had BED and are still engaging in binge eating. However, it is unclear whether the findings generalize to BED clinic samples. The use of convenience sampling means that population representativeness cannot be assured. We were unable to investigate the role of AN subtype because of small sample cell sizes. White women comprised the majority of participants and only adults aged 18-45 years were recruited, which limits generalizability. Some of the measures in this study were retrospective, which raises the possibility that memory or reporting bias may have affected the results.

In summary, our findings suggest that lifetime AN may be associated with poorer outcomes and a greater burden of illness in people with BED. Mental health and gastrointestinal comorbidity was very common in our full sample, emphasizing the importance of a comprehensive approach to BED care that addresses screening, referral, and management of mental health and somatic comorbidities.

Supplementary Material

supinfo

Funding information:

Foundation of Hope, Raleigh North Carolina (Bulik, PI); National Eating Disorders Association (Bulik and Tregarthen, PIs); Brain and Behavior Research Foundation (BBRF: NARSAD Distinguished Investigator Grant; Bulik, PI); National Institute of Mental Health (NIMH: R01MH119084, Bulik/Butner, MPIs; R01MH124871, Sullivan/Bulik MPI), and National Science Foundation Graduate Research Fellowship Program (grant no. DGE-1650116, Flatt; Grant No. 1645421, Sanzari). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NIH nor the National Science Foundation. No funding bodies were involved in the design of the study and collection, analysis, interpretation of data, or writing the manuscript.

Footnotes

Conflict of interest: Cynthia M. Bulik reports: Shire (grant recipient, Scientific Advisory Board member); Lundbeckfonden (grant recipient); Pearson (author, royalty recipient); Equip Health Inc. (clinical advisory board); and Recovery Record (research collaborator). Jenna P. Tregarthen and Stuart Argue are co-founders and shareholders of Recovery Record. All other authors have no conflicts of interest to disclose.

Data availability statement:

The Binge Eating Genetics Initiative (BEGIN) dataset will become publicly available in NIMHs National Data Archive (collection #3361).

References

  1. Aardoom JJ, Dingemans AE, Slof Op't Landt MC, & Van Furth EF (2012). Norms and discriminative validity of the Eating Disorder Examination Questionnaire (EDE-Q). Eating Behaviors, 13(4), 305–309. 10.1016/j.eatbeh.2012.09.002 [DOI] [PubMed] [Google Scholar]
  2. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th, text revised ed.). American Psychiatric Association. [Google Scholar]
  3. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). American Psychiatric Association. [Google Scholar]
  4. Angold A, Costello EJ, & Erkanli A (1999). Comorbidity. Journal of Child Psychology and Psychiatry and Allied Disciplines, 40(1), 57–87. [PubMed] [Google Scholar]
  5. Bardone-Cone AM, Maldonado CR, Crosby RD, Mitchell JE, Wonderlich SA, Joiner TE Jr., . . . le Grange D (2008). Revisiting differences in individuals with bulimia nervosa with and without a history of anorexia nervosa: Eating pathology, personality, and maltreatment. International Journal of Eating Disorders, 41(8), 697–704. 10.1002/eat.20567 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Benjamini Y, & Hochberg Y (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society, Series B, 57(1), 289–300. [Google Scholar]
  7. Bulik CM, Butner JE, Tregarthen J, Thornton LM, Flatt RE, Smith T, . . . Deboeck PR (2020). The Binge Eating Genetics Initiative (BEGIN): Study protocol. BMC Psychiatry, 20(1), 307. 10.1186/s12888-020-02698-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bulik CM, Carroll IM, & Mehler P (2021). Reframing anorexia nervosa as a metabo-psychiatric disorder. Trends in Endocrinology and Metabolism, 32(10), 752–761. 10.1016/j.tem.2021.07.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bulik CM, Coleman JRI, Hardaway JA, Breithaupt L, Watson HJ, Bryant CD, & Breen G (2022). Genetics and neurobiology of eating disorders. Nature Neuroscience, 25(5), 543–554. 10.1038/s41593-022-01071-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bulik CM, Sullivan PF, Fear JL, & Pickering A (2000). Outcome of anorexia nervosa: Eating attitudes, personality, and parental bonding. International Journal of Eating Disorders, 28(2), 139–147. [DOI] [PubMed] [Google Scholar]
  11. Castellini G, Lo Sauro C, Mannucci E, Ravaldi C, Rotella CM, Faravelli C, & Ricca V (2011). Diagnostic crossover and outcome predictors in eating disorders according to DSM-IV and DSM-V proposed criteria: A 6-year follow-up study. Psychosomatic Medicine, 73(3), 270–279. 10.1097/PSY.0b013e31820a1838 [DOI] [PubMed] [Google Scholar]
  12. Citrome L (2017). Binge-eating disorder and comorbid conditions: Differential diagnosis and implications for treatment. Journal of Clinical Psychiatry, 78 Suppl 1, 9–13. 10.4088/JCP.sh16003su1c.02 [DOI] [PubMed] [Google Scholar]
  13. Cohen J (1988). Statistical power analysis for the behavioral sciences. Lawrence Erlbaum Associates. [Google Scholar]
  14. Dingemans A, Danner U, & Parks M (2017). Emotion regulation in binge eating disorder: A review. Nutrients, 9(11), 1274. 10.3390/nu9111274 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Drossman DA (2006). The functional gastrointestinal disorders and the Rome III process. Gastroenterology, 130(5), 1377–1390. [DOI] [PubMed] [Google Scholar]
  16. Drossman DA (2006). Rome III: The new criteria. China Journal of Digestive Diseases, 7(4), 181–185. 10.1111/j.1443-9573.2006.00265.x [DOI] [PubMed] [Google Scholar]
  17. Eddy KT, Dorer DJ, Franko DL, Tahilani K, Thompson-Brenner H, & Herzog DB (2007). Should bulimia nervosa be subtyped by history of anorexia nervosa? A longitudinal validation. International Journal of Eating Disorders, 40 Suppl, S67–71. 10.1002/eat.20422 [DOI] [PubMed] [Google Scholar]
  18. Eddy KT, Keel PK, Dorer DJ, Delinsky SS, Franko DL, & Herzog DB (2002). Longitudinal comparison of anorexia nervosa subtypes. International Journal of Eating Disorders, 31(2), 191–201. 10.1002/eat.10016 [DOI] [PubMed] [Google Scholar]
  19. Fairburn CG, & Beglin SJ (2008). Eating Disorder Examination Questionnaire (EDE-Q 6.0). In Fairburn CG (Ed.), Cognitive behavior therapy and eating disorders (pp. 309–314). Guilford Press. [Google Scholar]
  20. Fichter MM, Quadflieg N, Crosby RD, & Koch S (2017). Long-term outcome of anorexia nervosa: Results from a large clinical longitudinal study. International Journal of Eating Disorders, 50(9), 1018–1030. 10.1002/eat.22736 [DOI] [PubMed] [Google Scholar]
  21. Goldschmidt AB, Peterson CB, Wonderlich SA, Crosby RD, Engel SG, Mitchell JE, . . . Berg KC (2013). Trait-level and momentary correlates of bulimia nervosa with a history of anorexia nervosa. International Journal of Eating Disorders, 46(2), 140–146. 10.1002/eat.22054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Huas C, Godart N, Caille A, Pham-Scottez A, Foulon C, Divac SM, . . . Rouillon F (2013). Mortality and its predictors in severe bulimia nervosa patients. European Eating Disorders Review, 21(1), 15–19. 10.1002/erv.2178 [DOI] [PubMed] [Google Scholar]
  23. Johnson SU, Ulvenes PG, Øktedalen T, & Hoffart A (2019). Psychometric Properties of the General Anxiety Disorder 7-Item (GAD-7) Scale in a Heterogeneous Psychiatric Sample. Frontiers in Psychology, 10, 1713. 10.3389/fpsyg.2019.01713 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kessler RC, Adler L, Ames M, Demler O, Faraone S, Hiripi E, . . . Walters EE (2005). The World Health Organization Adult ADHD Self-Report Scale (ASRS): A short screening scale for use in the general population. Psychological Medicine, 35(2), 245–256. 10.1017/s0033291704002892 [DOI] [PubMed] [Google Scholar]
  25. Kroenke K, Spitzer RL, & Williams JB (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16(9), 606–613. 10.1046/j.1525-1497.2001.016009606.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mathisen TF, Rosenvinge JH, Friborg O, Pettersen G, Stensrud T, Hansen BH, . . . Sundgot-Borgen J (2018). Body composition and physical fitness in women with bulimia nervosa or binge-eating disorder. International Journal of Eating Disorders, 51(4), 331–342. 10.1002/eat.22841 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Milos G, Spindler A, Schnyder U, & Fairburn CG (2005). Instability of eating disorder diagnoses: Prospective study. British Journal of Psychiatry, 187, 573–578. 10.1192/bjp.187.6.573 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Mond J, & Gorrell S (2021). "Excessive exercise" in eating disorders research: problems of definition and perspective. Eat Weight Disord, 26(4), 1017–1020. 10.1007/s40519-020-01075-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Monteleone P, Di Genio M, Monteleone AM, Di Filippo C, & Maj M (2011). Investigation of factors associated to crossover from anorexia nervosa restricting type (ANR) and anorexia nervosa binge-purging type (ANBP) to bulimia nervosa and comparison of bulimia nervosa patients with or without previous ANR or ANBP. Comprehensive Psychiatry, 52(1), 56–62. 10.1016/j.comppsych.2010.05.002 [DOI] [PubMed] [Google Scholar]
  30. Palsson OS, Whitehead W, Törnblom H, Sperber AD, & Simren M (2020). Prevalence of Rome IV functional bowel disorders among adults in the United States, Canada, and the United Kingdom. Gastroenterology, 158(5), 1262–1273.e1263. 10.1053/j.gastro.2019.12.021 [DOI] [PubMed] [Google Scholar]
  31. Peters JE, Basnayake C, Hebbard GS, Salzberg MR, & Kamm MA (2021). Prevalence of disordered eating in adults with gastrointestinal disorders: A systematic review. Neurogastroenterology and Motility, e14278. 10.1111/nmo.14278 [DOI] [PubMed] [Google Scholar]
  32. Reas DL, Grilo CM, & Masheb RM (2006). Reliability of the Eating Disorder Examination-Questionnaire in patients with binge eating disorder. Behaviour Research and Therapy, 44(1), 43–51. 10.1016/j.brat.2005.01.004 [DOI] [PubMed] [Google Scholar]
  33. Rubin DB (1987). Multiple imputation for nonresponse in surveys. John Wiley & Sons. [Google Scholar]
  34. Schafer JL (1997). Analysis of incomplete multivariate data. Chapman and Hall. [Google Scholar]
  35. Schaumberg K, Jangmo A, Thornton LM, Birgegård A, Almqvist C, Norring C, . . . Bulik CM (2019). Patterns of diagnostic transition in eating disorders: A longitudinal population study in Sweden. Psychological Medicine, 49(5), 819–827. 10.1017/s0033291718001472 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Spitzer RL, Kroenke K, Williams JB, & Löwe B (2006). A brief measure for assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166(10), 1092–1097. 10.1001/archinte.166.10.1092 [DOI] [PubMed] [Google Scholar]
  37. Stice E, Desjardins CD, Rohde P, & Shaw H (2021). Sequencing of symptom emergence in anorexia nervosa, bulimia nervosa, binge eating disorder, and purging disorder and relations of prodromal symptoms to future onset of these disorders. Journal of Abnormal Psychology, 130(4), 377–387. 10.1037/abn0000666 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Stice E, Marti CN, & Rohde P (2013). Prevalence, incidence, impairment, and course of the proposed DSM-5 eating disorder diagnoses in an 8-year prospective community study of young women. Journal of Abnormal Psychology, 122(2), 445–457. 10.1037/a0030679 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sullivan PF, Bulik CM, Carter FA, Gendall KA, & Joyce PR (1996). The significance of a prior history of anorexia in bulimia nervosa. International Journal of Eating Disorders, 20(3), 253–261. [DOI] [PubMed] [Google Scholar]
  40. Thornton LM, Munn-Chernoff MA, Baker JH, Juréus A, Parker R, Henders AK, . . . Bulik CM (2018). The Anorexia Nervosa Genetics Initiative (ANGI): Overview and methods. Contemporary Clinical Trials, 74, 61–69. 10.1016/j.cct.2018.09.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Udo T, & Grilo CM (2018). Prevalence and correlates of DSM-5–defined eating disorders in a Nntionally representative sample of U.S. adults. Biological Psychiatry, 84(5), 345–354. https://doi.org/ 10.1016/j.biopsych.2018.03.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Utzinger LM, Mitchell JE, Cao L, Crosby RD, Crow SJ, Wonderlich SA, & Peterson CB (2015). Clinical utility of subtyping binge eating disorder by history of anorexia or bulimia nervosa in a treatment sample. International Journal of Eating Disorders, 48(6), 785–789. 10.1002/eat.22422 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Vaz-Leal FJ, Rodríguez Santos L, García-Herráiz MA, Monge-Bautista M, & López-Vinuesa B (2011). Bulimia nervosa with history of anorexia nervosa: Could the clinical subtype of anorexia have implications for clinical status and treatment response? International Journal of Eating Disorders, 44(3), 212–219. 10.1002/eat.20805 [DOI] [PubMed] [Google Scholar]
  44. Welch E, Jangmo A, Thornton LM, Norring C, von Hausswolff-Juhlin Y, Herman BK, . . . Bulik CM (2016). Treatment-seeking patients with binge-eating disorder in the Swedish national registers: Clinical course and psychiatric comorbidity. BMC Psychiatry, 16, 163. 10.1186/s12888-016-0840-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Wiklund CA, Rania M, Kuja-Halkola R, Thornton LM, & Bulik CM (2021). Evaluating disorders of gut-brain interaction in eating disorders. International Journal of Eating Disorders, 54(6), 925–935. https://doi.org/ 10.1002/eat.23527 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Zucker NL, & Bulik CM (2020). On bells, saliva, and abdominal pain or discomfort: Early aversive visceral conditioning and vulnerability for anorexia nervosa. International Journal of Eating Disorders, 53(4), 508–512. 10.1002/eat.23255 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supinfo
NIHMS1851425-supplement-supinfo.docx (89.4KB, docx)

Data Availability Statement

The Binge Eating Genetics Initiative (BEGIN) dataset will become publicly available in NIMHs National Data Archive (collection #3361).

RESOURCES