Skip to main content

Some NLM-NCBI services and products are experiencing heavy traffic, which may affect performance and availability. We apologize for the inconvenience and appreciate your patience. For assistance, please contact our Help Desk at info@ncbi.nlm.nih.gov.

NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Nov 1.
Published in final edited form as: Int J Eat Disord. 2011 Aug 8;45(7):850–855. doi: 10.1002/eat.20948

Diagnostic Concordance of the Interview and Questionnaire Versions of the Eating Disorder Examination

Kelly C Berg 1,2, E Colleen Stiles-Shields 2, Sonja A Swanson 3, Carol B Peterson 1, Jocelyn Lebow 4, Daniel Le Grange 2
PMCID: PMC3213320  NIHMSID: NIHMS308043  PMID: 21826696

Abstract

Objective

The diagnostic concordance of the interview (EDE) and questionnaire (EDE-Q) versions of the Eating Disorder Examination was examined.

Method

Two-hundred seventeen patients seeking eating disorder (ED) treatment completed the EDE and EDE-Q before beginning treatment. Diagnostic algorithms were generated for the DSM-IV-TR and proposed DSM-5 criteria using data first from the EDE and then from the EDE-Q; thus, each participant received four diagnoses.

Results

The sensitivity of the EDE-Q for individual diagnoses ranged from 27.8% to 84.3% (DSM-IV-TR) and from 36.8% to 80.8% (DSM-5). The specificity of the EDE-Q for individual diagnoses ranged from 71.1% to 98.5% (DSM-IV-TR) and from 77.3% to 98.0% (DSM-5). The overall diagnostic concordance was moderate (κ=0.57-0.60).

Discussion

The proposed DSM-5 criteria improved the diagnostic concordance of the two instruments and reduced the prevalence of Eating Disorder Not Otherwise Specified (EDNOS). However, concordance improvement was modest and both instruments still diagnosed most respondents with EDNOS.

Keywords: Eating Disorder Examination, Eating Disorder Examination-Questionnaire, eating disorders, diagnostic concordance, DSM-5, prevalence


Developed to measure both cognitive and behavioral symptoms of eating disorders (ED), the interview (EDE)1 and questionnaire (EDE-Q)2 versions of the Eating Disorder Examination are among the most widely used assessments of ED pathology. The psychometric properties of both the EDE and EDE-Q have been extensively studied and are detailed in a recent review3. Additionally, over a dozen studies have examined the convergence of EDE and EDE-Q scores on both the subscales and behavioral frequency items, finding good convergence for the subscale scores and the items that measure the frequency of compensatory behaviors (i.e., self-induced vomiting). However, scores on the items used to measure binge eating (BE) frequency have demonstrated weaker convergence4. These results have important implications for several reasons: a specific frequency of BE is required for a diagnosis of bulimia nervosa (BN) in DSM-IV-TR5, the proposed DSM-5 criteria6 require a specific frequency of BE for both BN and binge eating disorder (BED) diagnoses, and both instruments can be used to generate ED diagnoses.

There is surprisingly little research on the concordance of ED diagnoses derived by the EDE and EDE-Q. Three studies have poor to fair agreement between the EDE and EDE-Q for the classification of individuals as “binge eaters”7-9, with kappas ranging from .26 to .47. In a sample of 60 women seeking treatment for AN10, the overall agreement between the EDE and EDE-Q was moderate for the overall diagnosis of AN (κ=.56), and good for the subtyping of those with AN (κ=.79). Lastly, in a sample of women seeking treatment for substance abuse11, the diagnostic concordance of the EDE and EDE-Q improved when full-threshold and subthreshold ED diagnoses were combined.

Of the five studies described above, only one included a clinical ED sample10. Although ED symptoms were reported in the other samples, the majority of participants in those studies did not meet criteria for an ED, which could potentially inflate the diagnostic agreement between the EDE and EDE-Q12. Additionally, the one ED sample10 was composed entirely of patients who met criteria for full-threshold Anorexia Nervosa (AN), which limits the generalizability of the findings. Thus, the objective of the current study was to examine the diagnostic concordance of the EDE and EDE-Q in a sample of ED patients with varied clinical presentations. Additionally, given that there is no published research on the diagnostic concordance of the EDE and EDE-Q when using the proposed DSM-5 ED criteria, a secondary objective of the study was to compare the diagnostic concordance of the EDE and EDE-Q when using the DSM-IV-TR criteria versus the proposed DSM-5 criteria.

Method

Participants

Data were collected from 498 patients assessed at the Eating Disorders Program at The University at Chicago Medical Center (UCMC) between 1999 and 2009. Participants were either self-, parent-, or physician-referred to the program, which offers outpatient ED treatment. Given that the purpose of this study was to examine the diagnostic concordance of the EDE and EDE-Q, patients were excluded from the current analyses if they had missing data that precluded the generation of either an EDE or an EDE-Q diagnosis. Due to missing data, 281 patients (56.4%) were excluded from the current analyses.

The 217 participants that were included in the study ranged in age from 9 to 61 years (M=19.6±9.6). Participants' percent of ideal body weight (%IBW) ranged from 60.5% to 278.9% (Q1=92.0%, Q2=104.1%, Q3=117.3%). The majority of participants were female (90.3%) and most identified as Caucasian (73.2%), followed by African American (13.1%), and Latino (10.3%), with other groups constituting 5.1% of the sample. Patients who were and were not included in the analyses did not differ with regard to age, gender, ethnicity, %IBW, menstrual status, or global scores on either the EDE or EDE-Q (all ps>.30). There were no significant differences between the two groups' scores on any of the EDE or EDE-Q items that were included in the diagnostic algorithms (all ps>.10), except OBE frequency (ps<.05). Patients who were not included in the analyses reported significantly fewer OBEs on both instruments.

Measures

Eating Disorder Examination

The EDE1 is a clinician-administered interview that assesses the cognitive and behavioral symptoms of EDs. Items that assess the cognitive symptoms of EDs are based on a 28-day time frame and are scored on a 7-point Likert scale ranging from 0 to 6, with higher scores indicating more severe pathology. Behavior frequency items are based on the past 3 months and scores on these items represent the specific frequency of each behavior. Research supports the reliability of EDE scores and demonstrates that EDE scores can differentiate between cases and non-cases of EDs3.

Eating Disorder Examination-Questionnaire

The EDE-Q2 is a self-report questionnaire that was developed to address logistical limitations of the EDE (e.g., time required to administer). The EDE-Q measures the same constructs using nearly identical language and rating scales as the EDE. The primary differences between the EDE and EDE-Q are that the EDE allows the interviewer to ask additional questions for clarification purposes and all EDE-Q questions are based on a 28-day timeframe. Scores on the EDE-Q have demonstrated reliability and the ability to distinguish between cases and non-cases of EDs3.

Physical Assessment

Height and weight were measured using calibrated instruments. Percent of ideal body weight (%IBW) was defined as current weight divided by 50th centile weight. The 50th centile weight for adolescent participants (age≤19) was calculated using the Center for Disease Control's Child and Adolescent BMI Percentile Calculator which takes into account age, gender, and height13. For adult participants (age≥20), the 50th centile weight was defined as the median weight, for gender and height, published in the 1959 Metropolitan Life Insurance Tables10,14. Menstrual status was based on self-report.

Procedure

All assessments were completed in the context of an intake appointment, prior to the start of treatment, and on the same day. The order in which the EDE and EDE-Q were administered was not controlled and the entire assessment procedure lasted approximately three hours. Written consent for adult patients or parental/guardian consent and patient assent for child/adolescent patients were obtained. Refusal to consent did not alter patients' assessment or treatment in the clinic. The use of these data was approved by the UCMC Institutional Review Board.

Statistical Analyses

Diagnostic algorithms

Algorithms were created to classify individuals into one of five diagnostic groups: AN, BN, BED, Eating Disorder Not Otherwise Specified (EDNOS), or no ED diagnosis. Separate diagnostic algorithms were developed for the DSM-IV-TR criteria and the proposed DSM-5 criteria, first using scores from the EDE and then using scores from the EDE-Q. Thus, four sets of diagnostic algorithms were created and four diagnoses were generated for each participant (i.e., DSM-IV diagnosis based on EDE scores). The algorithms were derived from those suggested by Fairburn and Cooper (1994) and are described in detail in Table 1.

Table 1. Diagnostic Algorithms for DSM-IV-TR and Proposed DSM-5 Eating Disorders.
Criteria DSM-IV-TR DSM-5
Anorexia Nervosa
  A. Low Weight %IBW<85% %IBW<85%
  B. Fear of Weight Gain Fear of Weight Gain ≥ 4 Fear of Weight Gain ≥ 4 or Compensatory Behaviors ≥ Once per Week
  C. Body Image Disturbance Importance of Weight ≥ 4 or Importance of Shape ≥ 4 Importance of Weight ≥ 4 or Importance of Shape ≥ 4
  D. Amenorrhea Amenorrhea or Male or Prepubertal or Birth Control N/A

Bulimia Nervosa
  A. Binge Eating OBEs Endorsed OBEs Endorsed
  B. Compensatory Behavior Vomiting, Laxative Use, or Diuretic Use Endorsed Vomiting, Laxative Use, or Diuretic Use Endorsed
  C. Frequency/Durationa Total OBEs ≥ Twice per Week and Total Compensatory Behaviors ≥ Twice per Week Total OBEs ≥ Once per Week and Total Compensatory Behaviors ≥Once per Week
  D. Overevaluation of Shape and Weight Importance of Weight ≥ 4 or Importance of Shape ≥ 4 Importance of Weight ≥ 4 or Importance of Shape ≥ 4
  E. Not AN %IBW ≥ 85% %IBW ≥ 85%

Binge Eating Disorder
  A. Binge Eating OBEs Endorsed OBEs Endorsed
  B. Associated Featuresb -- --
  C. Distressb -- --
  D. Frequency/Durationa Total OBEs ≥ Twice per Week Total OBEs ≥ Once per Week
  E. No Regular Compensatory Behaviors, Not BN, Not ANa Total Compensatory Behaviors ≤ Once per Month and %IBW ≥ 85% Total Compensatory Behaviors ≤ Once per Month and %IBW ≥ 85%

Eating Disorder Not Otherwise Specified
  1. Clinically significant eating disorder that does not meet criteria for AN, BN, or BED DSM-IV-TR criteria were not met for AN, BN, BED, or No Eating Disorder Diagnosis DSM-5 criteria were not met for AN, BN, BED, or No Eating Disorder Diagnosis

No Eating Disorder Diagnosis
  1. No Clinically Significant ED Cognitions Fear of Weight Gain ≤ 2 and Importance of Weight ≤ 2 and Importance of Shape ≤ 2 Fear of Weight Gain ≤ 2 and Importance of Weight ≤ 2 and Importance of Shape ≤ 2
  2. No Clinically Significant ED Behaviors Total OBEs = 0 and Total SBEs = 0 and Total Compensatory Behaviors = 0 Total OBEs = 0 and Total SBEs = 0 and Total Compensatory Behaviors = 0

Note : Proposed changes to the criteria are shown in boldface. DSM-IV-TR=Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision; DSM-5=Diagnostic and Statistical Manual of Mental Disorders, 5th edition; IBW=ideal body weight; OBE=objective bulimic episode; AN=anorexia nervosa; BN=bulimia nervosa; BED=binge eating disorder; ED=eating disorder; SBE=subjective bulimic episode.

a

Duration for the Eating Disorder Examination was based on the past 3 months whereas the duration for the Eating Disorder Examination-Questionnaire was based on the past month.

b

These criteria were not assessed.

It is important to note that DSM-IV-TR considers BED to be an example of EDNOS and includes BED in the appendices as a diagnosis warranting further study5. Given that the proposed DSM-5 criteria for BED are different than the research criteria for BED published in the DSM-IV-TR appendices8, BED was diagnosed separately from EDNOS in the current study to examine whether changes to the BED criteria impact the diagnostic concordance of the EDE and EDE-Q for the diagnosis of BED. The algorithm for DSM-IV-TR BED was based on the research criteria for BED included in the DSM-IV-TR appendices5.

Although the diagnostic algorithms used in this study closely resemble the DSM-IV-TR and DSM-5 criteria, they are not exact replications. First, neither the EDE nor the EDE-Q includes items that assess whether there is a disturbance in the way ones shape or weight is experienced or a lack of recognition of the seriousness of low body weight, both of which are criteria included in the DSM-IV-TR and the proposed DSM-5 criteria for AN. Second, the EDE-Q does not include items that measure the associated features of BED or distress regarding BE. Third, the DSM-5 criteria allow one's weight history, body build, and physiological symptoms to be considered when determining whether one's weight meets the A criteria for AN. These factors are not assessed by the EDE or EDE-Q and could not be included in the diagnostic algorithm for DSM-5 AN. To be consistent with previous research10, a cutoff of 85% IBW was used for both the DSM-IV-TR and DSM-5 criteria for AN.

In addition to not using certain criteria in the algorithms, the item content of the EDE and EDE-Q required changes to aspects of the criteria for AN, BN, and BED algorithms. First, as stated previously, the time periods assessed by the EDE and EDE-Q differ; thus, the duration criteria for BE and compensatory behaviors was based on 3 months when using the EDE, but only 28 days when using the EDE-Q. Second, only purging compensatory behaviors (i.e., self-induced vomiting, laxative and diuretic misuse) were included in the diagnostic algorithms. Nonpurging compensatory behaviors were not included because 1) fasting, as defined by DSM-IV-TR5, is not assessed by either the EDE or EDE-Q, 2) although excessive exercise is assessed by both instruments, the extent to which exercise is compensatory is not assessed by either instrument, and 3) there is a dearth of research on the reliability and validity of the items that assess nonpurging compensatory behaviors3. Finally, dietary restriction was not included in the DSM-5 algorithm for AN because of concerns that measures of dietary restraint, including the EDE and EDE-Q, may not be valid measures of caloric intake15,16.

Diagnostic Concordance

The sensitivitya, specificityb, positive predictive value (PPV)c, and negative predictive value (NPV)d were calculated separately for the DSM-IV-TR and DSM-5 criteria using the EDE diagnoses as the “gold standard”. The overall diagnostic concordance was also calculated separately for DSM-IV-TR and DSM-5 diagnoses using Cohen's kappa.

Results

Prevalence

For both DSM-IV-TR and DSM-5 diagnoses, the EDE and EDE-Q identified approximately the same proportion of the sample as having each ED diagnosis (see Table 2). The largest discrepancy between the EDE and EDE-Q prevalence estimates was for the diagnosis of BED when using DSM-IV-TR criteria (8.3% vs. 5.1% respectively). More individuals were diagnosed with a full threshold ED when using DSM-5 criteria, regardless of which instrument was used to make the diagnosis. Subsequently, the prevalence of EDNOS decreased when using DSM-5 criteria. Including BED as part of EDNOS resulted in a 14.7% and 13.4% reduction in EDNOS when using the EDE and EDE-Q, respectively. However, EDNOS remained the most common DSM-5 ED diagnosis (58.3%-67.2%) regardless of which instrument was used.

Table 2. Prevalence Rates of DSM-IV-TR and DSM-5 Eating Disorders Generated by the EDE and EDE-Q.

Diagnosis DSM-IV-TR DSM-5


EDE EDE-Q EDE EDE-Q




N % N % N % N %
AN 11 5.1% 11 5.1% 15 6.9% 16 7.4%
BN 39 18.0% 45 20.7% 48 22.1% 52 24.0%
BED 18 8.3% 11 5.1% 19 8.8% 17 7.8%
EDNOS 134 61.8% 137 63.1% 120 55.3% 119 54.8%
No ED 15 6.9% 13 6.0% 15 6.9% 13 6.0%

Total 217 100.0% 217 100.0% 217 100.0% 217 100.0%

Note : DSM-IV-TR=Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision; DSM-5=Diagnostic and Statistical Manual of Mental Disorders, 5th edition; EDE=Eating Disorder Examination; EDE-Q=Eating Disorder Examination-Questionnaire; AN=anorexia nervosa; BN=bulimia nervosa, BED=binge eating disorder; EDNOS=eating disorder not otherwise specified; No ED=no eating disorder diagnosis

Diagnostic Concordance of the EDE and EDE-Q

With the exception of BED, the EDE-Q demonstrated acceptable sensitivity (60.0%-84.3%) and PPV (64.4%-82.5%) for both DSM-IV-TR and DSM-5 diagnoses (see Table 3). The sensitivity (27.8%-36.8%) and PPV (41.2%-45.5%) of the EDE-Q for the detection of BED were poor for both sets of criteria. The EDE-Q demonstrated good to excellent specificity (71.1% to 98.5%) and NPV (73.8% to 98.5%) for all DSM-IV-TR and DSM-5 diagnoses. Overall, the sensitivity, specificity, PPV, and NPV of the EDE-Q to detect EDE diagnoses were equivalent or improved when using DSM-5 versus DSM-IV-TR criteria. The overall diagnostic concordance was moderate for both for DSM-IV-TR and DSM-5 diagnoses (κs =.57 and .60, respectively).

Table 3. Sensitivity, Specificity, Positive Predictive Value, and Negative Predictive Value of the EDE-Q to Detect Diagnoses Generated by the EDE.

Diagnosis DSM-IV-TR DSM-5


Sn Sp PPV NPV Sn Sp PPV NPV
AN 72.7% 98.5% 72.7% 98.5% 80.0% 98.0% 75.0% 98.5%
BN 74.4% 91.0% 64.4% 94.2% 79.2% 91.7% 73.1% 93.9%
BED 27.8% 97.0% 45.5% 93.7% 36.8% 94.9% 41.2% 94.0%
EDNOS 84.3% 71.1% 82.5% 73.8% 80.8% 77.3% 81.5% 76.5%
No ED 60.0% 98.0% 69.2% 97.1% 60.0% 98.0% 69.2% 97.1%

Note : EDE=Eating Disorder Examination; EDE-Q=Eating Disorder Examination-Questionnaire; DSM-IV-TR=Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision; DSM-5=Diagnostic and Statistical Manual of Mental Disorders, 5th edition; Sn = sensitivity; Sp = specificity; PPV = positive predictive value; NPV = negative predictive value; AN = anorexia nervosa, BN = bulimia nervosa; BED = binge eating disorder, EDNOS = eating disorder not otherwise specified; No ED = no eating disorder diagnosis.

Discussion

The purpose of this study was to examine the diagnostic concordance of the EDE and EDE-Q in a clinical sample of ED patients with varied symptom presentations. Regardless of which criteria set was used, the EDE and EDE-Q identified approximately the same proportion of individuals in each diagnostic group. However, the two instruments identified slightly different groups of individuals as having each diagnosis. Thus, although the EDE and EDE-Q arrived at nearly identical prevalence estimates for each diagnostic group, the overall diagnostic concordance between the two instruments was only moderate. The diagnostic concordance was particularly poor for the identification of BED.

Given that there was only moderate diagnostic concordance between the EDE and EDE-Q, it is notable that neither instrument appeared to diagnose more individuals with full threshold EDs relative to the other. This observation is in contrast to findings that individuals consistently score higher on the subscales of the EDE-Q regardless of their diagnostic status4. One possible explanation for this discrepancy may be the fact that EDE and EDE-Q subscales are scored as continuous variables whereas scores on the EDE and EDE-Q must be dichotomized when used in diagnostic algorithms. Regardless, these data suggest that despite the tendency for individuals to endorse more severe pathology on the EDE-Q, they are not more likely to receive a full-threshold diagnosis when assessed by the EDE-Q rather than the EDE.

Implications for Assessment

These data have important implications for researchers interested in using the EDE or EDE-Q diagnostically. The results suggest that researchers interested in estimating the prevalence of EDs may arrive at similar results regardless of whether the EDE or EDE-Q is used. However, the results also suggest that researchers who use the EDE to identify cases of EDs may identify slightly different samples of participants than researchers who use the EDE-Q for the same purpose. Given the similarities between the EDE and EDE-Q with regard to item content and scoring, these findings may overestimate the diagnostic concordance of a structured interview and self-report questionnaire that are not directly related.

Implications for DSM-5

Although the primary objective of this study was to examine the diagnostic concordance of the EDE and EDE-Q, these data provided an opportunity to evaluate whether using the proposed DSM-5 criteria reduces the prevalence of EDNOS. These data illustrate that more individuals met criteria for a full threshold ED when using the proposed DSM-5 criteria, resulting in a reduction of the relative prevalence of EDNOS. Similar results have been reported in other clinical samples18. However, the majority of participants in this sample were still diagnosed with EDNOS when using the proposed DSM-5 criteria.

It is possible that EDNOS remained the most prevalent ED diagnosis in this sample because the algorithms used in the current study did not fully replicate the proposed DSM-5 criteria. Several posthoc analyses were conducted to examine this possibility. First, because nonpurging compensatory behaviors were not included in the diagnostic algorithms for BN, patients with the nonpurging subtype of BN may have been misdiagnosed. Posthoc analyses indicated that 0.0% (EDE-Q) to 2.5% (EDE) of patients diagnosed with DSM-5 EDNOS met all the criteria for BN except purging frequency. Second, neither nonpurging compensatory behaviors nor dietary restriction were included in the diagnostic algorithms for DSM-5 AN. However, posthoc analyses indicate that only 2.5% (EDE-Q) to 5.0% (EDE) of patients diagnosed with DSM-5 EDNOS met all the criteria for AN, except fear of weight gain. Together, these data suggest that the inclusion of nonpurging compensatory behaviors and dietary restriction would not have had a substantial impact on the overall prevalence of EDNOS.

Of the individuals diagnosed with DSM-5 EDNOS, 70.8% (EDE) and 72.3% (EDE-Q) met criteria for one of the specific feeding and eating conditions not elsewhere classified proposed for DSM-58. The most common of these diagnoses was Subthreshold BED (50.6%-52.3%), followed by Atypical AN (32.9%-37.2%), Subthreshold BN (4.7%-10.6%), and Purging Disorder (5.8%-5.9%). Night Eating Syndrome (NES) was not represented in these analyses because symptoms of NES are not evaluated by either the EDE or EDE-Q. The remaining participants diagnosed with DSM-5 EDNOS represented a mixture of symptom configurations.

Limitations and Strengths

There are limitations to consider when interpreting these findings. Perhaps most importantly, missing data precluded the use of data from all patients presenting to the clinic for treatment, which may limit the generalizability of the findings. However, with the exception of OBE frequency, there were no significant differences between the patients who were and were not included in the analyses. Additionally, posthoc analyses using samples with different levels of missing data did not change the results substantially (e.g., including only those with complete information on both instruments, N=180), suggesting that the subset of participants used in these analyses may have been representative of the larger sample with regard to diagnostic concordance between the EDE and EDE-Q.

A second limitation of the current study is that the diagnostic algorithms used in this study were not perfect replications of the DSM-IV-TR or DSM-5 criteria. Although posthoc analyses suggest that specific changes to the algorithms would not have changed the results substantially, the total impact of all deviations from the DSM-IV-TR and DSM-5 criteria could not be evaluated. Additionally, these results only illustrate the level of diagnostic agreement between the EDE and EDE-Q; they do not necessarily provide evidence of the validity of EDE- or EDE-Q-derived diagnoses. It is also notable that the sample included both adolescents and adults, which could obscure potential age-related discrepancies between the EDE and EDE-Q. However, stratifying the sample by age did not change the results. Lastly, these data were collected in the context of an outpatient clinic; thus, these results may not generalize to non-treatment-seeking samples.

Despite these limitations, this study is the first attempt to examine the diagnostic concordance of the EDE and EDE-Q in a sample of ED patients with varied symptom presentations. It is also the first time the diagnostic concordance of the EDE and EDE-Q has been examined when using the proposed DSM-5 criteria. Further research is needed to examine the validity of EDE- and EDE-Q-derived diagnoses and to examine whether inconsistencies between EDE- and EDE-Q-derived diagnoses result in differential findings across studies. Additionally, revisions made to the EDE and EDE-Q to reflect the DSM-5 criteria should seek to maximize concordance between the two measures.

Acknowledgments

The investigators would like to thank the research and clinical staff at the Eating Disorders Program of The University at Chicago Medical Center for their assistance with data collection.

Footnotes

Financial Disclosures/Conflicts of Interest : This work was supported in part by a grant from NIMH (T32 MH082761-01). Additionally, the authors report the following financial disclosures: Dr. Berg, none; Ms. Stiles-Shields, none; Ms. Swanson, none; Dr. Peterson, royalties from Guilford Press; Ms. Lebow, none; Dr. Le Grange, royalties from Guilford Press and honoraria from the Training Institute for Child and Adolescent Eating Disorders, LLC.

a

In the current study, sensitivity is the probability that, if a participant was diagnosed by the EDE as having a particular ED (e.g., AN), the EDE-Q also identified the participant as having that particular ED (e.g., AN).

b

In the current study, specificity is the probability that, if a person was not diagnosed by the EDE as having a particular ED, the EDE-Q also identified the participant as not having that particular ED.

c

In the current study, PPV is the probability that, if the EDE-Q identified a participant as having a particular ED, the person was, in fact, diagnosed by the EDE as having that particular ED.

d

In the current study, NPV is the probability that, if the EDE-Q identified a participant as not having a particular ED, the person was, in fact, not diagnosed by the EDE as having that particular ED.

References

  • 1.Fairburn CG, Cooper Z. The eating disorder examination. In: Fairburn CG, Wilson GT, editors. Binge Eating: Nature, Assessment, and Treatment. 12th. New York: Guilford Press; 1993. pp. 317–360. [Google Scholar]
  • 2.Fairburn CG, Beglin SJ. Assessment of eating disorders: Interview or self-report questionnaire? Int J Eat Disord. 1994;16:363–370. [PubMed] [Google Scholar]
  • 3.Berg KC, Peterson CB, Frazier P, Crow SJ. Psychometric evaluation of the eating disorder examination and eating disorder examination-questionnaire: A systematic review of the literature. Int J Eat Disord. doi: 10.1002/eat.20931. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Berg KC, Peterson CB, Frazier P, Crow SJ. Convergence of scores on the interview and questionnaire versions of the eating disorder examination: A meta-analytic review. Psych Assess. doi: 10.1037/a0023246. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.American Psychiatric Association. Diagnostic and statistical manual of mental disorders, fourth edition, text revision (DSM-IV-TR) Washington, D.C.: APA; 1994. [Google Scholar]
  • 6.American Psychiatric Association. DSM-5 development. 2011. [Google Scholar]
  • 7.Mond JM, Hay PJ, Rodgers B, Owen C, Beumont PJV. Validity of the eating disorder examination questionnaire (EDE-Q) in screening for eating disorders in community samples. Behav Res Ther. 2004;42:551–567. doi: 10.1016/S0005-7967(03)00161-X. [DOI] [PubMed] [Google Scholar]
  • 8.de Zwaan M, Mitchell JE, Swan-Kremeier L, McGregor T, Howell ML, Roerig JL, et al. A comparison of different methods of assessing the features of eating disorders in post-gastric bypass patients: A pilot study. European Eating Disorders Review. 2004;12:380–386. [Google Scholar]
  • 9.Kalarchian MA, Wilson GT, Brolin RE, Bradley L. Assessment of eating disorders in bariatric surgery candidates: Self-report questionnaire versus interview. Int J Eat Disord. 2000;28:465–469. doi: 10.1002/1098-108x(200012)28:4<465::aid-eat17>3.0.co;2-2. [DOI] [PubMed] [Google Scholar]
  • 10.Wolk SL, Loeb KL, Walsh BT. Assessment of patients with anorexia nervosa: Interview or self-report. Int J Eat Disord. 2005;37:92–99. doi: 10.1002/eat.20076. [DOI] [PubMed] [Google Scholar]
  • 11.Black CMD, Wilson GT. Assessment of eating disorders: Interview versus questionnaire. Int J Eat Disord. 1996;20:43–50. doi: 10.1002/(SICI)1098-108X(199607)20:1<43::AID-EAT5>3.0.CO;2-4. [DOI] [PubMed] [Google Scholar]
  • 12.Sobell MB, Sobell LC, Klajner F, Pavan D, Basian E. The reliability of a timeline method for assessing normal drinker college students' recent drinking history: Utility for alcohol research. Addictive Behaviors. 1986;11:149–161. doi: 10.1016/0306-4603(86)90040-7. [DOI] [PubMed] [Google Scholar]
  • 13.Centers for Disease Control and Prevention. BMI percentile calculator for child and teen. 2010. [Google Scholar]
  • 14.Halmi KA, Agras WS, Crow SJ, Mitchell J, Wilson GT, Bryson SW, et al. Predictors of treatment acceptance and completion in anorexia nervosa. Arch Gen Psych. 2005;62:776–781. doi: 10.1001/archpsyc.62.7.776. [DOI] [PubMed] [Google Scholar]
  • 15.Stice E, Fisher M, Lowe MR. Are dietary restraint scales valid measures of acute dietary restriction? unobtrusive observational data suggest not. Psych Assess. 2004;16:51–59. doi: 10.1037/1040-3590.16.1.51. [DOI] [PubMed] [Google Scholar]
  • 16.Stice E, Cooper JA, Schoeller DA, Tappe K, Lowe MR. Are dietary restriction scales valid measures of moderate- to long-term dietary restriction? objective biological and behavioral data suggest not. Psych Assess. 2007;19:449–458. doi: 10.1037/1040-3590.19.4.449. [DOI] [PubMed] [Google Scholar]
  • 17.Gormally J, Black S, Daston S, Rardin D. The assessment of binge eating severity among obese persons. Addictive Behaviors. 1982;7:47–55. doi: 10.1016/0306-4603(82)90024-7. [DOI] [PubMed] [Google Scholar]
  • 18.Keel PK, Brown TA, Holm-Denoma J, Bodell LP. Comparison of DSM-IV versus proposed DSM-5 diagnostic criteria for eating disorders: Reduction of eating disorders not otherwise specified and validity. Int J Eat Disord. doi: 10.1002/eat.20892. in press. [DOI] [PubMed] [Google Scholar]

RESOURCES