Treatment Outcome in People with Subthreshold Compared with Full-Syndrome Binge Eating Disorder


University of Tuebingen, Department of Psychosomatic Medicine and Psychotherapy, Osianderstrasse 5, 72076 Tuebingen, Germany. E-mail:


Objective: There is a controversial discussion in the literature as to whether individuals with subthreshold binge eating disorder (subBED) differ clinically significantly from individuals with full-syndrome binge eating disorder (BED). This study was designed to compare eating-related and general psychopathology at baseline and in response to a multimodal treatment program in obese people with subBED compared with BED.

Research Methods and Procedures: A total of 96 obese participants (BMI ≥ 30 kg/m2) were assessed for eating-related and general psychopathology at baseline. Thirty-nine participants meeting criteria for BED and 19 participants meeting criteria for subBED attended a 15-session outpatient group therapy including cognitive behavioral therapy extended by interpersonal therapy, nutritional counseling, and a supervised walking exercise. Participants with eating disorders were reassessed at the end of treatment and at 3-month follow-up. The obese control group without an eating disorder (n = 38) was assessed once. This was not a randomized controlled trial.

Results: Intent-to-treat analyses revealed no differences between subBED and full-syndrome BED participants with regard to eating-related and general psychopathology at baseline and with regard to treatment outcome. All participants experienced substantial improvements, and the results remained stable during follow-up (except for dietary restraint). At follow-up, participants with subBED and BED remained different from non-eating disorder controls in eating-related but not general psychopathology.

Discussion: The findings indicate that our multimodal treatment program is equally effective in obese subBED and BED participants, suggesting that a differentiation currently seems not to be of clinical significance.


Since the introduction of the research criteria for the binge eating (BE)1 disorder (BED) diagnosis in the DSM-IV as a specific example of an eating disorder (ED) not otherwise specified, several randomized controlled trials have shown the efficacy of psychotherapeutic interventions in reducing BE frequency and psychiatric comorbidity (1).

A substantial proportion of obese individuals seeking treatment fulfill some but not all of the diagnostic criteria for BED (2). There is almost no evidence of how to guide the management of people who report typical BE episodes but do not fulfill all of the BED research criteria.

Given baseline similarities of psychopathology and stress responsiveness that have been reported in previous studies (3, 4), it is of great interest to evaluate whether obese people with subthreshold BED (subBED) merit psychotherapeutic treatment similar to that of patients who fulfill all of the BED criteria.

Research Methods and Procedures

Diagnostic Assessment

Clinical interviews were conducted by one of the therapists (A.Q.) to assess lifetime diagnoses of an ED [Structured Interview on Anorexic and Bulimic Disorders-Expert Assessment (SIAB-EX)] (5) and of psychiatric comorbidity (Structured Clinical Interview for DSM-IV Axis I Disorders) (6). The interviews were administered to all (n = 96) obese participants (BMI ≥ 30 kg/m2) at baseline, and the SIAB-EX was readministered in ED individuals (n = 58) at the end of treatment and at 3-month follow-up. Thirty-nine obese participants met full DSM-IV criteria for BED, and 19 met criteria for subBED, which was defined as fulfilling the DSM-IV BED criteria except for either the frequency criterion of objective BE episodes (less than twice a week but at least once per 28 days) or the large amount of food criterion (subjective BE, defined as eating <800 kcal/binge with a frequency of at least twice a week) (Table 1). Thirty-eight obese control participants without a lifetime diagnosis of an ED were recruited by advertisement and carefully matched (age, gender, BMI) to the ED group. They were assessed only once, without participating in the treatment intervention. All participants gave written informed consent, and the study was approved by the local Ethics Committee of the Medical University of Heidelberg.

Table 1.  Baseline sociodemographic characteristics in controls and participants with BED and subBED
 Obese controls (N = 38)SubBED (N = 19)BED (N = 39)p*
  • *

    Comparison between obese controls and participants with an ED (subBED and BED). Comparison between participants with subBED and BED at baseline showed no significant differences.

Age [mean (SD)]46.8 (8.9)43.2 (10.1)45.3 (10.0)0.27
BMI [mean (SD)]35.8 (6.0)37.7 (5.4)37.6 (5.5)0.13
Women [no. (%)]29 (76.3)16 (84.2)30 (76.9)0.73
Employment status [no. (%)]   0.27
 Full time18 (47.4)4 (21.1)14 (35.9) 
 Part time8 (21.1)5 (26.3)12 (30.8) 
 Not working12 (31.6)10 (52.6)13 (33.3) 
Age at onset of obesity [mean (SD)]22.8 (12.9)20.1 (16.3)23.4 (13.6)0.70
Psychotherapy in the past [no. (%)]9 (23.7)9 (47.4)20 (51.3)0.01
Current psychiatric comorbidity [no. (%)]    
 Any axis I disorder7 (18.4)9 (47.4)20 (51.3)0.002
 Mood disorders overall4 (10.5)6 (31.6)16 (41.0)0.003
 Anxiety disorders overall5 (13.2)5 (26.3)4 (10.3)0.75

Self-Report Questionnaires

The Three-Factor Eating Questionnaire (TFEQ) (7) and the Patient Health Questionnaire (PHQ-D) (8) for common mental disorders were administered at all time-points. From the PHQ-D, the scores of the depression module and of the psychosocial stress severity scale are reported.

Multimodal Treatment Program

The multimodal treatment program was a 15-session group intervention program, which comprised cognitive behavioral therapy, elements of interpersonal psychotherapy, diet counseling, and exercise therapy. All sessions were run by a psychological psychotherapist trained in ED treatment (A.Q.) and a nutritionist (S.S.). Each two hour session was subdivided into 60 minutes of psychotherapy with integrated diet counseling and 60 minutes of supervised walking exercise. For the intervention, a detailed treatment manual was developed (9). Participants were informed that overcoming episodes of BE and establishing regular and healthy eating patterns were the primary aims, whereas weight loss was of secondary concern. One of the authors (S.Z.) provided session-by-session supervision, and adherence to the treatment manual was ensured by regular feedback on video-taped sessions.

Statistical Analysis

All analyses of the treatment results were intent-to-treat analyses using the last observation carried forward procedure. Between-group differences were analyzed with Student's t test or χ2 test if appropriate. Repeated measure ANOVAs were performed to determine treatment effects across time-points in participants with BED and subBED. p < 0.05 (two-tailed) was considered statistically significant.



Tables 1 and 2 summarize the sociodemographic characteristics and general and ED-related psychopathology of all participants. Before treatment, subBED participants were not significantly different from BED participants. ED participants (BED, subBED) differed significantly at baseline from obese controls with respect to ED-related (TFEQ) and general (PHQ-D) psychopathology, except for dietary restraint.

Table 2.  ED psychopathology (TFEQ), general psychopathology (PHQ-D), and weight in obese controls and individuals with subBED and full-syndrome BED at baseline and intent-to-treat outcomes at posttreatment and follow-up in eating-disordered participants
 BaselinePosttreatment3-Month follow-upMain time effectGroup × time interaction
 Obese controls (N = 38)SubBED (N = 19) [mean (SD)]BED (N = 39) [mean (SD)]SubBED (N = 19) [mean (SD)]BED (N = 39) [mean (SD)]SubBED (N = 19) [mean (SD)]BED (N = 39) [mean (SD)]FpFp
  • *

    Comparison between obese controls and participants with an ED (subBED and BED), ≤0.001. Comparison between participants with subBED and BED at baseline showed no differences.

 Dietary restraint7.7 (3.7)7.7 (3.5)6.5 (3.4)10.0 (3.7)9.8 (4.1)9.9 (3.5)8.8 (4.6)15.8<0.0010.520.59
 Disinhibition7.7 (3.9)*12.3 (2.4)12.6 (1.8)9.9 (3.5)10.8 (3.0)10.3 (3.7)10.3 (3.6)15.7<0.0010.470.63
 Hunger4.8 (3.1)*9.2 (2.9)9.7 (2.7)6.3 (3.3)7.1 (2.9)6.6 (3.1)7.0 (3.3)29.5<0.0010.170.85
 Depression5.1 (5.0)*8.1 (5.0)9.5 (4.0)7.4 (6.0)7.4 (4.6)6.2 (5.2)7.3 (4.3)7.50.0010.860.43
 Stress5.6 (3.7)*8.2 (3.6)8.1 (3.0)6.8 (3.6)6.2 (3.6)6.5 (3.4)6.4 (3.9)9.0<0.0010.280.76
 Weight (kg)101.2 (24.1)104.6 (21.1)109.5 (19.0)102.7 (21.9)106.3 (18.2)101.8 (23.9)105.9 (19.2)7.10.0010.290.80

Treatment Outcome

Ten participants dropped out of treatment (BED, n = 5, 12.8%; subBED, n = 5, 26.3%). There were no significant group differences between BED and subBED in the number of attended treatment sessions.

Eating Behavior and Weight

Less than one-half of the ED participants reported complete abstinence from BE at follow-up (subBED, 42.1%; BED, 48.6%). One-quarter of the BED individuals improved but still met criteria for subBED at follow-up (Table 3).

Table 3.  Intent-to-treat diagnostic outcome of obese participants with BED and subBED at posttreatment and at 3-month follow-up, based on a structured interview (5)
 BED baseline (N = 39)SubBED baseline (N = 19)
 BED N (%)SubBED N (%)No BE N (%)BED N (%)SubBED N (%)No BE N (%)
  • *

    Diagnostic criteria during the last 4 weeks.

Posttreatment*10 (25.6)12 (30.8)17 (43.6)1 (5.3)10 (52.6)8 (42.1)
Follow-up*10 (25.6)10 (25.6)19 (48.7)1 (5.3)10 (52.6)8 (42.1)

ED participants showed a significant treatment effect on weight, which remained stable during follow-up. There was no significant difference between BED (−3.3 kg; SD, 5.7 kg) and subBED (−1.9 kg; SD, 3.8 kg) participants (Table 2).

Eating-Related and General Psychopathology

ED participants showed significant main treatment effects on the TFEQ and the PHQ-D that were not significantly different between BED and subBED participants (Table 2). Treatment effects remained stable during the follow-up period, except for dietary restraint (TFEQ). At follow-up, ED participants persisted to be significantly different from obese controls with respect to disinhibition (TFEQ, t = 3.3, p = 0.002) and perceived hunger (TFEQ, t = 3.1, p = 0.002).


The main finding of the present study was that subBED compared with BED participants demonstrated an equivalent, substantial improvement in eating-related (TFEQ) and general (PHQ-D) psychopathology in response to treatment, which remained stable during short-term follow-up (except for dietary restraint). The different rate of drop-outs in the subBED and BED group did not influence the intent-to-treat results, as an additional analysis of completers confirmed. This combination of treatment components has not been tested before in patients with BED. However, because our treatment package did not include elements that are novel or unusual in the treatment of obese patients with BED, it is reasonable to assume that the two groups responded similarly.

Even though ED participants report significant improvement in eating behavior, they remained significantly different from obese controls with respect to eating-related (TFEQ) but not general (PHQ-D) psychopathology. Less than one-half of the participants no longer met criteria for subBED or BED diagnoses at follow-up. In the BED group, one-quarter of the participants showed partial remission, but they still fulfilled criteria for subBED. Transferring to less severe forms of BE in response to treatment (2) or spontaneously (10) appears to be common in BED and seems to be associated with a persistence of increased caloric intake.

In summary, the present findings suggest that currently, a differentiation between obese people with subBED and BED might not be of practical utility because they did not differ in general and eating-related psychopathology and showed an equivalent response to a multimodal treatment intervention.


This study was supported by the Medical Faculty of the University of Heidelberg (Grant F 203594). We thank M. Fichter for advice on the assessment of BED with the SIAB-EX.


  • 1

    Nonstandard abbreviations: BE, binge eating; BED, BE disorder; ED, eating disorder; subBED, subthreshold BED; SIAB-EX, Structured Interview on Anorexic and Bulimic Disorders-Expert Assessment; TFEQ, Three-Factor Eating Questionnaire; PHQ-D, Patient Health Questionnaire.

  • The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.