• Eating
  • Estimating the Prevalence of Binge...

Estimating the Prevalence of Binge Eating Disorder in a Community Sample From the United States: Comparing DSM-IV-TR and DSM-5 Criteria

Article Abstract

Objective: To estimate binge eating disorder (BED) prevalence according to DSM-5 and DSM-IV-TR criteria in US adults and to estimate the proportion of individuals meeting DSM-5 BED criteria who reported being formally diagnosed.

Methods: A representative sample of US adults who participated in the National Health and Wellness Survey were asked to respond to an Internet survey (conducted in October 2013). Assessments included 3-month, 12-month, and lifetime BED prevalence based on DSM-5 and DSM-IV-TR criteria and demographics, psychiatric comorbidities, and self-esteem (Rosenberg Self-Esteem Scale). Descriptive statistics are provided. Prevalence estimates were calculated using poststratification sampling weights.

Results: Of 22,397 respondents, 344 (women, n = 242; men, n = 102) self-reported symptoms consistent with DSM-5 BED symptom criteria. The 3-month, 12-month, and lifetime DSM-5 prevalence estimates (95% CIs) projected to the US population were 1.19% (1.04%-1.37%), 1.64% (1.45%-1.85%), and 2.03% (1.83%-2.26%), respectively. The 12-month and lifetime projected DSM-IV-TR prevalence estimates were 1.15% (1.00%-1.32%) and 1.52% (1.35%-1.70%), respectively. Of respondents meeting DSM-5 BED criteria in the past 12 months, 3.2% (11/344) reported receiving a formal diagnosis. Compared with non-BED respondents, respondents meeting DSM-5 BED criteria in the past 12 months were younger (mean ± SD age = 46.01 ± 14.32 vs 51.59 ± 15.80 years; P < .001), had a higher body mass index (mean ± SD = 33.71 ± 9.36 vs 27.96 ± 6.68 kg/m2; P < .001), and had lower self-esteem (mean ± SD score = 16.47 ± 6.99 vs 23.33 ± 6.06; P < .001).

Conclusions: DSM-5 BED criteria resulted in higher BED prevalence estimates than with DSM-IV-TR criteria. Most BED respondents did not report being formally diagnosed, indicating an unmet need in BED recognition and diagnosis.


This work may not be copied, distributed, displayed, published, reproduced, transmitted, modified, posted, sold, licensed, or used for commercial purposes. By downloading this file, you are agreeing to the publisher’s Terms & Conditions.

Estimating the Prevalence of Binge Eating Disorder in a Community Sample From the United States:

Comparing DSM-IV-TR and DSM-5 Criteria

Vertical divider

ABSTRACT

Objective: To estimate binge eating disorder (BED) prevalence according to DSM-5 and DSM-IV-TR criteria in US adults and to estimate the proportion of individuals meeting DSM-5 BED criteria who reported being formally diagnosed.

Methods: A representative sample of US adults who participated in the National Health and Wellness Survey were asked to respond to an Internet survey (conducted in October 2013). Assessments included 3-month, 12-month, and lifetime BED prevalence based on DSM-5 and DSM-IV-TR criteria and demographics, psychiatric comorbidities, and self-esteem (Rosenberg Self-Esteem Scale). Descriptive statistics are provided. Prevalence estimates were calculated using poststratification sampling weights.

Results: Of 22,397 respondents, 344 (women, n = 242; men, n = 102) self-reported symptoms consistent with DSM-5 BED symptom criteria. The 3-month, 12-month, and lifetime DSM-5 prevalence estimates (95% CIs) projected to the US population were 1.19% (1.04%-1.37%), 1.64% (1.45%-1.85%), and 2.03% (1.83%-2.26%), respectively. The 12-month and lifetime projected DSM-IV-TR prevalence estimates were 1.15% (1.00%-1.32%) and 1.52% (1.35%-1.70%), respectively. Of respondents meeting DSM-5 BED criteria in the past 12 months, 3.2% (11/344) reported receiving a formal diagnosis. Compared with non-BED respondents, respondents meeting DSM-5 BED criteria in the past 12 months were younger (mean ± SD age = 46.01 ± 14.32 vs 51.59 ± 15.80 years; P < .001), had a higher body mass index (mean ± SD = 33.71 ± 9.36 vs 27.96 ± 6.68 kg/m2; P < .001), and had lower self-esteem (mean ± SD score = 16.47 ± 6.99 vs 23.33 ± 6.06; P < .001).

Conclusions: DSM-5 BED criteria resulted in higher BED prevalence estimates than with DSM-IV-TR criteria. Most BED respondents did not report being formally diagnosed, indicating an unmet need in BED recognition and diagnosis.

J Clin Psychiatry 2016;77(8):e968-e974

dx.doi.org/10.4088/JCP.15m10059

aShire, Lexington, Massachusetts

bMerck Research Laboratories, North Wales, Pennsylvania

cKantar Health, Princeton, New Jersey

dEpi Excellence LLC, Garnet Valley, Pennsylvania, and Center for Clinical Epidemiology and Biostatistics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania

eKantar Health, Horsham, Pennsylvania

fCenter for Weight and Eating Disorders, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania

gM. H. Erder Health Economics, Inc, Livingston, New Jersey

Drs Cossrow, Pawaskar, and Erder and Ms Ming are no longer affiliated with Shire. Dr Victor is no longer affiliated with Kantar Health.

*Corresponding author: Barry K. Herman, MD, MMM, Shire, 300 Shire Way, Lexington, MA 02421 (bherman@shire.com).

Binge eating disorder (BED), the general symptomatology of which was recognized as early as 1959,1 was included in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) as a diagnosis for further study in the early 1990s.2-4 However, it did not receive official recognition as a distinct eating disorder until its inclusion in DSM-5 in 2013.5

According to DSM-5 diagnostic criteria for BED (Table 1), binge eating must occur at least once a week for ≥ 3 months. These episodes are characterized by the consumption, in a discrete period of time, of a larger amount of food than is typical for most people under similar circumstances, by a sense of lack of control over eating, and by marked distress about binge eating.5 Unlike bulimia nervosa (BN) and anorexia nervosa (AN), BED is not associated with recurrent inappropriate compensatory behaviors (eg, purging or excessive exercise).5 In the DSM-5, the binge episode frequency criterion was reduced from the DSM-IV Text Revision (DSM-IV-TR) criterion of ≥ 2 days per week for ≥ 6 months,6 but other criteria were unchanged. Therefore, individuals previously not meeting DSM-IV-TR criteria may now meet DSM-5 criteria.

Two large-scale studies assessed BED prevalence using DSM-IV-TR criteria.7,8 The 2001-2003 US National Comorbidity Survey Replication study (N = 9,282) reported 12-month and lifetime BED prevalence estimates of 1.2% and 2.6%, respectively.7,8 In a series of surveys across 14 countries (N = 24,124), another study reported 12-month and lifetime BED prevalence estimates of 0.8% (range, 0.1%-1.8%) and 1.9% (range, 0.2%-4.7%), respectively.8 Several studies have reported that BED is associated with increased rates of comorbid psychiatric disorders (eg, mood and anxiety disorders),7-11 personality disorders (eg, obsessive-compulsive disorder),12 and psychological features (eg, low self-esteem).12,13

Although studies have estimated the epidemiologic implications of the updated DSM-5 BED criteria,14-16 no large-scale surveys have used these criteria to assess BED prevalence or the demographic and clinical profiles of afflicted individuals.17 This study estimated and compared BED prevalence based on DSM-5 and DSM-IV-TR criteria in a representative US adult population. It also determined the proportion of individuals meeting DSM-5 BED criteria who reported receiving a formal diagnosis of BED. The demographic and clinical characteristics and self-esteem levels of respondents who met DSM-5 criteria for BED were also compared with those of respondents without BED.

METHODS

Sample and Procedures

The National Health and Wellness Survey (NHWS) is a self-administered Internet survey completed by approximately 75,000 US adults each year, which is sponsored by Kantar Health (Princeton, New Jersey). Potential NHWS respondents are identified through the general panel of Lightspeed Research (Warren, New Jersey). As previously described,18 adults (≥ 18 years of age) in the United States are eligible for this general panel and can join by responding to advertisements in e-newsletters and online banner advertisements. The demographic profile of NHWS respondents approximates the Current Population Survey of the US Census Bureau.19,20 Disease prevalence estimates of the NHWS are similar to those of the National Health Interview Survey.19

Among respondents to the 2012 NHWS (N = 71,157 from January 2012-December 2012) and 2013 NHWS (N = 75,000 from January 2013-September 2013), 69,972 were contacted to participate in this Internet survey (Validate Attitudes and Lifestyle Issues in Depression, ADHD and Troubles with Eating [VALIDATE]). Participation was cutoff when the desired sample size was met. The survey was completed October 9-29, 2013. It included questions assessing DSM-5 and DSM-IV-TR criteria for BED and questions related to demographic and clinical profiles, general health, self-esteem, and diagnosed psychiatric disorders. For their participation in the study, respondents were compensated with "panel points," which could be accumulated and exchanged for consumer goods (eg, electronics, clothing).

Table 1

Click figure to enlarge

clinical points

  • Twelve-month and lifetime binge eating disorder (BED) prevalence estimates based on DSM-5 criteria (1.64% and 2.03%, respectively) were higher than estimates based on DSM-IV-TR criteria (1.15% and 1.52%, respectively).
  • Among survey respondents who met DSM-5 criteria for BED, only 3.2% had ever received a formal diagnosis from a health care provider.
  • Survey respondents meeting DSM-5 criteria for BED were significantly more likely to report lifetime depression (adjusted odds ratio [OR] = 3.82), lifetime anxiety (adjusted OR = 3.17), lifetime bipolar disorder (adjusted OR = 3.92), and attention-deficit/hyperactivity disorder in the last 6 months (adjusted OR = 8.24) than respondents who did not meet diagnostic criteria for BED after controlling for age, sex, and body mass index.

A stratified random sample framework was implemented to ensure representation across sex, age, and ethnicity. The protocol for the survey was approved by an institutional review board (Sterling IRB #4509) before initiation.

Measures

Self-reported demographics, psychiatric symptoms, and psychological features were collected. All respondents were also asked whether they had received a diagnosis for several psychiatric and medical conditions, including attention-deficit/hyperactivity disorder (ADHD), anxiety, bipolar disorder, and depression.

Responses to questions representing DSM-5 or DSM-IV-TR BED symptom criteria (Supplementary eTable 1) were used to assess whether individuals could be considered to meet diagnostic criteria for BED across 3-month, 12-month, and "ever" time frames to capture the respective prevalence estimates (3-month prevalence was not assessed for DSM-IV-TR because the duration criterion was 6 months in DSM-IV-TR). For the duration criterion, respondents were asked how long they had experienced BED symptoms (< 1, 1-2, 3-4, 5-6, or 7-12 months). Those responding 5-6 or 7-12 months were classified as meeting the DSM-IV-TR criterion. Consistent with DSM-IV-TR and DSM-5 criteria, respondents who self-reported receiving a BN or AN diagnosis were not considered to meet BED symptom criteria.

Self-esteem was assessed with the Rosenberg Self-Esteem Scale (RSE), a 10-item self-report scale measuring both positive and negative feelings about oneself.21 The RSE has acceptable reliability and validity in measuring global self-worth in adult populations.22

The Patient Health Questionnaire-9 (PHQ-9)23 assessed major depressive disorder (MDD) symptoms in the past 2 weeks using 2 scoring methods. For each of the PHQ-9 items, scores range from 0 ("not at all") to 3 ("nearly every day"). When scored as a continuous variable, PHQ-9 total score ranges from 0 to 27 (0-4 = minimal depression; 5-9 = mild depression; 10-14 = moderate depression; 15-19 = moderately severe depression; 20-27 = severe depression).23 When each item is scored in a dichotomized fashion,23 PHQ-9 total score ranges from 0 to 9. Individuals are classified as "consider major depressive disorder" if their total score is ≥ 5 and they report "little interest or pleasure in doing things" and/or "feeling down, depressed, or hopeless" on "more than half the days" or "nearly every day." The Mood Disorder Questionnaire (MDQ)24 was administered as part of the NHWS. Respondents were omitted from being categorized as having MDD via the PHQ-9 if they screened positive for bipolar disorder on the MDQ or reported a prior diagnosis of bipolar disorder or schizophrenia.

Respondents were assessed for ADHD in the past 6 months using the 18-item Adult ADHD Self-Report Scale (ASRS) version 1.1.25 Each item is scored on 5-point Likert scales that range from "never" to "very often."25,26 For assessment of a DSM-5 ADHD diagnosis, all items were scored dichotomously (1 or 0) and the total score was obtained by summing the individual item scores.26 Scores ≥ 5 among the 9 inattentive symptoms or the 9 hyperactive/impulsive symptoms were considered to fulfill DSM-5 ADHD symptom criteria.

Data Analysis and Statistics

Prevalence estimates for BED are based on the full sample and calculated using poststratification sampling weights based on population totals for joint strata of sex (male/female), age (18-29, 30-39, 40-49, 50-64, ≥ 65 years), and race (white, black, other) derived from the September 2013 Current Population Survey.27 Prevalence estimates are provided as percentages with 95% CI calculated using Taylor Series linearization. Significance testing for comparisons of prevalence estimates based on DSM-IV-TR and DSM-5 criteria were not conducted because the statistical comparison of the estimates would violate the assumption of independence.

Individuals were classified as meeting criteria for a DSM-5 BED diagnosis in the past 12 months (referred to herein as BED respondents) or as not meeting those criteria (referred to herein as non-BED respondents). Respondents who self-reported a lifetime diagnosis of AN or BN were excluded from the BED respondent group. These criteria for the categorization of BED respondents and non-BED respondents were used because they are generally in keeping with DSM-5 diagnostic criteria for BED, hence providing the most conservative BED prevalence estimate.

Non-BED respondents may or may not have reported periods during which they ate large amounts of food (Criterion A1); however, these periods were not accompanied by loss of control over eating (Criterion A2). Respondents were excluded from the bivariate analyses (BED respondents vs non-BED respondents) if they reported binge eating in the past 12 months (periods when large amounts of food were consumed [Criterion A1] and feelings of loss of control were present [Criterion A2]), but other BED criteria were not met.

Score distributions on the RSE and PHQ-9 and psychiatric comorbidity prevalence among BED respondents were compared with those of non-BED respondents using t tests for continuous variables (least significant difference test with Bonferroni corrections to adjust for multiple comparisons) or χ2 tests for categorical variables. Unadjusted odds ratios for psychiatric comorbidities were calculated from 2 ×— 2 frequencies tables (BED vs non-BED respondents by comorbidity vs no comorbidity). Adjusted odds ratios were calculated via logistic regression models controlling for age, sex, and body mass index (BMI).

RESULTS

Respondent Classification and Demographics

The survey was completed by 22,397 respondents within 3 weeks or less of being invited to participate (Figure 1), all of whom were included in the analyses. Based on the number of e-mail invitations sent, the participation rate was 32%. Among all respondents, 344 (women, n = 242; men, n = 102) met the DSM-5 BED criteria in the past 12 months (BED respondents); 20,437 individuals did not meet these criteria in the past 12 months (non-BED respondents). The remaining 1,616 respondents were excluded because they met some, but not all, BED criteria in the past 12 months. Of respondents who self-reported a lifetime diagnosis of AN (n = 144) or BN (n = 104), 108 who self-reported AN and 58 who self-reported BN were included in the non-BED respondent group. The remainder (AN, n = 36; BN, n = 46) were part of the overall 1,616 respondents who were excluded for meeting some, but not all, BED criteria in the past 12 months.

Figure 1

Click figure to enlarge

In the overall population, most respondents were female (54.4% [12,182/22,397]), white (82.67% [18,515/22,397]), and ≥ 40 years old (72.84% [16,315/22,397]). Mean ± SD age was 51.12 ± 15.83 years. BED respondents were significantly younger than non-BED respondents (46.01 ± 14.32 vs 51.59 ± 15.80 years; P < .001). Compared with non-BED respondents, more BED respondents were female (53.7% [10,968/20,437] vs 70.3% [242/344]). The majority of respondents had a BMI ≥ 25 kg/m2 (62.9% [14,096/22,397]), and nearly one-third had a BMI ≥ 30 kg/m2 (31.2% [6,991/22,397]). Mean ± SD BMI was 28.27 ± 6.91 kg/m2 overall and was significantly greater in BED respondents than non-BED respondents (33.71 ± 9.36 vs 27.96 ± 6.68; P < .001). The overall demographic profile of survey respondents was roughly comparable to the general US adult population based on the September 2013 Current Population Survey.27

Prevalence of Binge Eating Disorder

Prevalence estimates (95% CI) for BED projected to the US adult population based on DSM-5 and DSM-IV-TR criteria are summarized in Table 2. The overall 3-month, 12-month, and lifetime prevalence estimates (95% CIs) based on DSM-5 criteria were 1.19% (1.04%-1.37%), 1.64% (1.45%-1.85%), and 2.03% (1.83%-2.26%), respectively. The 12-month and lifetime projected prevalence estimates using DSM-IV-TR criteria were 1.15% (1.00%-1.32%) and 1.52% (1.35%-1.70%), respectively (Table 2).

Table 2

Click figure to enlarge

Among BED respondents, a minority (11 [3.2%]) reported ever being diagnosed with BED by a health care provider.

Rosenberg Self-Esteem Scale

Among BED respondents, 34.9% (120/344) met criteria for low self-esteem (based on RSE scores < 15) compared with 8.7% (1,778/20,437) of non-BED respondents (Figure 2A and 2B). Mean ± SD RSE total scores were 16.47 ± 6.99 in BED respondents and 23.33 ± 6.06 in non-BED respondents (P < .001).

Figure 2

Click figure to enlarge

Patient Health Questionnaire-9

Mean ± SD PHQ-9 total score based on continuous scoring was 11.99 ± 7.27 in BED respondents and 3.19 ± 4.59 in non-BED respondents (P < .001). Based on dichotomized PHQ-9 scoring, 5.8% [1,295/22,397] of all respondents met criteria for possible MDD (BED respondents: 31.1% [107/344]; non-BED: 4.6% [948/20,437]) (Figure 2C and 2D).

For BED respondents meeting criteria for MDD, the mean ± SD age at onset for BED symptoms was 30.55 ± 16.49 years (women, 30.61 ± 15.31; men, 30.46 ± 18.29) and for depressive symptoms was 35.01 ± 15.64 years (women, 33.93 ± 13.93; men, 36.58 ± 17.91). Based on the difference between the respondents’ year of birth and the year in which symptoms were first experienced, 44.8% (39/87) of respondents with both BED and MDD symptoms reported that the age at BED symptom onset was earlier than the age at MDD symptom onset, and 25.3% (22/87) reported that the age at MDD symptom onset was earlier than at BED symptom onset. In the remaining respondents, BED and MDD age at symptom onset coincided.

Comorbidities Associated With Binge Eating Disorder

Lifetime psychiatric comorbidity prevalence based on self-report of having been diagnosed with a specific disorder was higher among BED respondents than non-BED respondents for depression (52.6% [181/344] vs 17.2% [3,518/20,437]; P < .001), anxiety (43.3% [149/344] vs 15.2% [3,098/20,437]; P < .001), and bipolar disorder (10.5% [36/344] vs 1.7% [356/20,437]; P < .001). A total of 41.3% (142/344) of BED respondents met criteria for comorbid ADHD in the last 6 months compared with 6.9% (1,400/20,437) of non-BED respondents. Adjusted odds ratios (BED relative to non-BED respondents) for these comorbidities were lower than unadjusted odds ratios, but they remained significant after controlling for age, sex, and BMI (Table 3).

Table 3

Click figure to enlarge

DISCUSSION

This study is the first to estimate BED prevalence using DSM-5 criteria in a representative population of US adults. The 12-month prevalence of BED was 1.64% based on DSM-5 criteria and 1.15% based on DSM-IV-TR criteria. Lifetime prevalence was 2.03% based on DSM-5 criteria and 1.52% based on DSM-IV-TR criteria. These changes are attributable to increases in the estimated prevalence of BED in women and men (from 1.53% to 2.00% and from 0.75% to 1.24%, respectively, for 12-month prevalence). Hudson et al14 previously estimated that DSM-5 criteria would increase lifetime BED prevalence in the United States from 3.5% to 3.6% in women and from 2.0% to 2.1% in men. However, Trace et al15 estimated that lifetime prevalence in women would double when the BED frequency and duration criteria shifted from ≥ 8 times per month for 6 months (approximate DSM-IV-TR criteria) to ≥ 4 times per month for 3 months (approximate DSM-5 criteria). These discrepancies in prevalence estimates and in the changes associated with the DSM-5 criteria may partially be due to differences in study populations and in the methods used to estimate prevalence. Hudson et al14 included individuals aged 18 to 70 years who were first-degree relatives of an individual with BED or of an overweight individual without BED and gathered data by face-to-face or telephone interviews via trained psychiatrist interviewers. Trace et al15 included only female twins aged 20 to 47 years and gathered data via telephone or Internet surveys. Only the current study used a population that is representative of the general US population.

The vast majority of BED respondents (96.8%) had not been formally diagnosed with BED by a health care provider. This may be due to the fact that BED was not a distinct disorder in DSM-IV-TR6 and would have been diagnosed as eating disorder not otherwise specified. Nonetheless, this low diagnosis rate emphasizes the need to improve awareness and recognition of BED among patients and health care providers, including primary care physicians. Given that so few individuals were formally diagnosed with BED, a limitation of this study is that it was not possible to conduct comparative statistical analyses to assess differences between these individuals and those meeting BED diagnostic criteria who had not received a formal BED diagnosis. However, in future studies, it would be informative to more closely examine the clinical characteristics of these individuals, including determining differences in their severity of illness.

The reported 12-month DSM-IV-TR prevalence estimate (1.15%) is roughly comparable to the 12-month DSM-IV-TR US estimate reported by Hudson et al (1.2%)7 and higher than the multinational estimate reported by Kessler et al (0.8%),8 which included the US estimate of Hudson et al.7 The lifetime prevalence in the current study (1.52%) was lower than the lifetime prevalence reported by Hudson et al and Kessler et al (2.6% and 1.9%, respectively).7,8 These discrepancies may also be attributable to different methods and operational definitions. The present study used an Internet survey, used self-reported symptom assessment, and directly asked about "loss of control." In contrast, Hudson et al7 and Kessler et al8 used face-to-face semistructured interviews and indirect measures of loss of control based on questions from the World Health Organization Composite International Diagnostic Interview. Furthermore, Hudson et al7 and Kessler et al8 asked whether individuals experienced 3 months of symptoms as opposed to the 6-month minimum required by DSM-IV-TR diagnostic criteria. As such, individuals experiencing symptoms for > 3 months but < 6 months were classified as meeting DSM-IV-TR BED criteria.

BED respondents also reported more psychiatric comorbidities than non-BED respondents; this statistically significant association remained after controlling for demographic factors. These findings are consistent with the published literature, which indicates that approximately 30%-80% of individuals who met DSM-IV-TR criteria for BED had lifetime comorbid mood or anxiety disorders.7-9,11,13,28 Similarly, BED respondents reported lower self-esteem compared with non-BED respondents, which is also consistent with previous reports.13 Although beyond the scope of the current study, in the future it will be important to examine any symptomatic or clinical differences between those individuals who meet criteria for DSM-5-defined BED and DSM-IV-TR-defined BED.

The key strength of this study is that it used a large, representative sample of US adults with respect to age, sex, and race/ethnicity. To our knowledge, no other study to date has assessed DSM-5 diagnostic criteria for BED in such a large nonclinical sample. Although direct comparisons of NHWS participants who responded to the survey to those who did not respond were not made in the current study, the overall NHWS population and respondents to the current study are both representative of the general US population.

A limitation of this study was that all survey data were self-reported. Consequently, BED diagnoses and associated psychiatric comorbidities cannot be clinically confirmed, and potential biases that can be inherently associated with self-reported measures, including both overestimation and underestimation, need to be considered. Additionally, because there is no validated screener for BED, self-reported BED symptoms were assessed using only survey questions based on the DSM-5 and DSM-IV-TR symptom criteria (Supplementary eTable 1), and formal psychiatric validation of the survey questions relative to structured clinical interviews was not performed. As such, Criterion E (exclusion for recurrent use of compensatory behavior) was operationalized as having self-reported a lifetime diagnosis of BN or AN. However, few respondents were identified as having BN or AN in the current study. Further, whether binge eating occurred in such respondents outside of the course of BN or AN or whether compensatory behavior occurred in these respondents was not assessed. Furthermore, non-symptom-based criteria (eg, functional impairment) were not employed in this analysis. In future studies, it would be of interest to conduct sensitivity analyses, including those with lifetime AN or BN, or in those meeting some but not all BED criteria.

The calculated participation rate was 32%, which may limit the generalizability of these findings. It is important to note that this level of participation may be an underestimation of the true rate. The participation rate reported in the current study is based on the number of invitations sent via e-mail, and the authors do not know how many invitees received the e-mail request because a proportion of the requests may have been automatically filtered by e-mail programs (eg, sent to spam or junk mail folders).29 If it were possible to calculate how many invitees actually opened the invitation, the overall participation rate may have been higher. Additionally, the current study recruited participants by sending a large number of survey invitations to potential respondents and then ended the ability to participate once the desired sample size was reached. Invitees attempting to participate after the desired sample size was reached were told the survey was no longer available. This method has the advantage of recruiting a large number of respondents in a short time. However, a consequence of this approach compared with an approach in which a smaller number of participants are recruited and allowed many weeks or months to respond is that traditional response rate calculations (ie, responses received divided by total invitations sent) will be lower than what might be expected. If recruitment were not cut off when the desired sample size was reached, and more time were allotted for those invited to respond, the participation rate for this study would likely have been higher than the rate reported in this article and perhaps closer to what would typically be expected of this kind of study. Importantly, the participation rate observed in this study is roughly comparable to that reported in a meta-analysis of Internet-based surveys30 and of a cell phone-based study from the Centers for Disease Control and Prevention-assisted health data collection project.31 Furthermore, the percentages of respondents with obesity and with MDD (per the PHQ-9) in this study are similar to national estimates of obesity and depression,32,33 suggesting the current sample was representative of the overall population.

In summary, the 12-month BED prevalence estimate based on DSM-5 criteria in the current study was 1.64% (2.00% and 1.24% in women and men, respectively). As expected, this estimate was higher than the estimate based on DSM-IV-TR criteria (1.15%), owing to the reduction in the duration and frequency criteria for binge eating episodes. The increase was particularly pronounced in men. Consistent with previous reports,7-9,11,13,28 higher psychiatric comorbidity rates were associated with BED in this study. Importantly, approximately 97% of BED respondents had never received a formal diagnosis of BED from a health care provider. However, BED is a new diagnosis in DSM-5, so the opportunity for physicians to make the diagnosis has been limited, which could partially account for the low level of diagnosis in study respondents. This low rate of diagnosis emphasizes the need to improve awareness and recognition of BED among patients and health care providers.

Submitted: April 17, 2015; accepted March 14, 2016.

Online first: May 24, 2016.

Potential conflicts of interest: Dr Cossrow was an employee of Shire at the time this study was conducted and holds stock and/or stock options in Shire. Dr Pawaskar was an employee of Shire at the time the study was conducted and holds stock and/or stock options in Shire and Merck. Dr Herman is an employee of Shire and holds stock and/or stock options in Shire. Dr Witt is an employee of Kantar Health, which conducted the data collection for the VALIDATE survey and analyzed the data on behalf of and with funding from Shire Development LLC. Dr Victor was an employee of Kantar Health at the time the study was conducted. Ms Ming was an employee of Shire at the time this study was conducted. Dr Wadden is an employee of the Perelman School of Medicine at the University of Pennsylvania; serves on advisory boards for Novo Nordisk, Nutrisystem, Orexigen, and Shire Pharmaceuticals; has received grant support, on behalf of the University of Pennsylvania, from the first three of these organizations and from Weight Watchers International; served as a consultant to Boehringer Ingelheim and as an advisor to Johns Hopkins University on a proprietary weight loss program (Innergy); and receives royalties from Guilford Press. Dr Erder was an employee of Shire at the time the study was conducted and held stock and/or stock options in Shire.

Funding/support: This research was conducted by Kantar Health (Horsham, Pennsylvania), with funding provided by the sponsor, Shire Development LLC (Lexington, Massachusetts). Shire Development LLC provided funding to Complete Healthcare Communications, LLC (Chadds Ford, Pennsylvania) for support in writing and editing this manuscript.

Role of the sponsor: The sponsor was involved in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, and approval of the manuscript. Although the sponsor was involved in the design, collection, analysis, interpretation, and fact checking of information, the content of this manuscript, the ultimate interpretation, and the decision to submit it for publication in The Journal of Clinical Psychiatry were made by the authors independently.

Previous presentation: Presented at the Annual Meeting of the American Psychiatric Association, August 7-10, 2014.

Acknowledgments: The authors would like to acknowledge Leo J. Russo, PhD, a former employee of Shire, for his contribution to this study. Prior to his departure from Shire, Dr Russo was involved in the design and conduct of the study. Under the direction of the authors, Stefan Kolata, PhD, and Craig Slawecki, PhD, employees of Complete Healthcare Communications, LLC, provided writing and formatting assistance for the manuscript.

Supplementary material: See accompanying pages.

REFERENCES

1. Stunkard AJ. Eating patterns and obesity. Psychiatr Q. 1959;33(2):284-295. PubMed doi:10.1007/BF01575455

2. Spitzer RL, Stunkard A, Yanovski S, et al. Binge eating disorder should be included in DSM-IV: a reply to Fairburn et al.’s "The classification of recurrent overeating: the binge eating disorder proposal." Int J Eat Disord. 1993;13(2):161-169. PubMed doi:10.1002/1098-108X(199303)13:2<161::AID-EAT2260130204>3.0.CO;2-R

3. Spitzer RL. Nonpurging bulimia nervosa and binge eating disorder. Am J Psychiatry. 1991;148(8):1097-1098. PubMed doi:10.1176/ajp.148.8.1097-a

4. Spitzer RL, Yanovski S, Wadden T, et al. Binge eating disorder: its further validation in a multisite study. Int J Eat Disord. 1993;13(2):137-153. PubMed

5. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Fifth Edition. Washington, DC: American Psychiatric Association; 2013.

6. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Fourth Edition, Text Revision. Washington, DC: American Psychiatric Association; 2000.

7. Hudson JI, Hiripi E, Pope HG Jr, et al. The prevalence and correlates of eating disorders in the National Comorbidity Survey Replication. Biol Psychiatry. 2007;61(3):348-358. PubMed doi:10.1016/j.biopsych.2006.03.040

8. Kessler RC, Berglund PA, Chiu WT, et al. The prevalence and correlates of binge eating disorder in the World Health Organization World Mental Health Surveys. Biol Psychiatry. 2013;73(9):904-914. PubMed doi:10.1016/j.biopsych.2012.11.020

9. Grilo CM, White MA, Masheb RM. DSM-IV psychiatric disorder comorbidity and its correlates in binge eating disorder. Int J Eat Disord. 2009;42(3):228-234. PubMed doi:10.1002/eat.20599

10. Grucza RA, Przybeck TR, Cloninger CR. Prevalence and correlates of binge eating disorder in a community sample. Compr Psychiatry. 2007;48(2):124-131. PubMed doi:10.1016/j.comppsych.2006.08.002

11. Javaras KN, Pope HG, Lalonde JK, et al. Co-occurrence of binge eating disorder with psychiatric and medical disorders. J Clin Psychiatry. 2008;69(2):266-273. PubMed doi:10.4088/JCP.v69n0213

12. Friborg O, Martinussen M, Kaiser S, et al. Personality disorders in eating disorder not otherwise specified and binge eating disorder: a meta-analysis of comorbidity studies. J Nerv Ment Dis. 2014;202(2):119-125. PubMed doi:10.1097/NMD.0000000000000080

13. Jones-Corneille LR, Wadden TA, Sarwer DB, et al. Axis I psychopathology in bariatric surgery candidates with and without binge eating disorder: results of structured clinical interviews. Obes Surg. 2012;22(3):389-397. PubMed doi:10.1007/s11695-010-0322-9

14. Hudson JI, Coit CE, Lalonde JK, et al. By how much will the proposed new DSM-5 criteria increase the prevalence of binge eating disorder? Int J Eat Disord. 2012;45(1):139-141. PubMed doi:10.1002/eat.20890

15. Trace SE, Thornton LM, Root TL, et al. Effects of reducing the frequency and duration criteria for binge eating on lifetime prevalence of bulimia nervosa and binge eating disorder: implications for DSM-5. Int J Eat Disord. 2012;45(4):531-536. PubMed doi:10.1002/eat.20955

16. Stice E, Marti CN, Rohde P. Prevalence, incidence, impairment, and course of the proposed DSM-5 eating disorder diagnoses in an 8-year prospective community study of young women. J Abnorm Psychol. 2013;122(2):445-457. PubMed doi:10.1037/a0030679

17. Smink FR, van Hoeken D, Hoek HW. Epidemiology, course, and outcome of eating disorders. Curr Opin Psychiatry. 2013;26(6):543-548. PubMed doi:10.1097/YCO.0b013e328365a24f

18. Annunziata K, Rak A, Del Buono H, et al. Vaccination rates among the general adult population and high-risk groups in the United States. PLoS ONE. 2012;7(11):e50553. PubMed doi:10.1371/journal.pone.0050553

19. Bolge SC, Doan JF, Kannan H, et al. Association of insomnia with quality of life, work productivity, and activity impairment. Qual Life Res. 2009;18(4):415-422. PubMed doi:10.1007/s11136-009-9462-6

20. DiBonaventura MD, Wagner JS, Yuan Y, et al. Humanistic and economic impacts of hepatitis C infection in the United States. J Med Econ. 2010;13(4):709-718. PubMed doi:10.3111/13696998.2010.535576

21. Rosenberg M. Society and the Adolescent Self-Image. Princeton, NJ: Princeton University Press; 1965.

22. Gray-Little B, Williams VS, Hancock TD. An item response theory analysis of the Rosenberg Self-Esteem Scale. Pers Soc Psychol Bull. 1997;23(5):443-451. doi:10.1177/0146167297235001

23. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606-613. PubMed doi:10.1046/j.1525-1497.2001.016009606.x

24. Hirschfeld RM. The Mood Disorder Questionnaire: a simple, patient-rated screening instrument for bipolar disorder. Prim Care Companion J Clin Psychiatry. 2002;4(1):9-11. PubMed doi:10.4088/PCC.v04n0104

25. Kessler RC, Adler LA, Gruber MJ, et al. Validity of the World Health Organization Adult ADHD Self-Report Scale (ASRS) Screener in a representative sample of health plan members. Int J Methods Psychiatr Res. 2007;16(2):52-65. PubMed doi:10.1002/mpr.208

26. Adult ADHD Self-Report Scale-V1.1 (ASRS-v1.1): Symptom Checklist. Harvard Medical School Web site. http://www.hcp.med.harvard.edu/ncs/ftpdir/adhd/18Q_ASRS_English.pdf. Updated 2003. Accessed November 14, 2014.

27. Current Population Survey. United States Census Bureau Web site. http://thedataweb.rm.census.gov/ftp/cps_ftp.html. Updated December 9, 2015. Accessed November 14, 2014.

28. Guerdjikova AI, McElroy SL, Kotwal R, et al. Comparison of obese men and women with binge eating disorder seeking weight management. Eat Weight Disord. 2007;12(1):e19-e23. PubMed doi:10.1007/BF03327777

29. Standard Definitions Report. American Association for Public Opinion Research Web site. http://www.aapor.org/AAPOR_Main/media/publications/Standard-Definitions2015_8theditionwithchanges_April2015_logo.pdf. Accessed April 2015

30. Cook C, Heath F, Thompson RL. A meta-analysis of response rates in web- or Internet-based surveys. Educ Psychol Meas. 2000;60(6):821-836. doi:10.1177/00131640021970934

31. Behavioral Risk Factor Surveillance System: 2013 Summary Data Quality Report. Centers for Disease Control and Prevention Web site. http://www.cdc.gov/brfss/annual_data/2013/pdf/2013_DQR.pdf. Updated August 15, 2014. Accessed February 13, 2015.

32. Pratt LA, Brody DJ. Depression and obesity in the US adult household population, 2005-2010. NCHS Data Brief. 2014;(167):1-8. PubMed

33. Greenberg PE, Fournier AA, Sisitsky T, et al. The economic burden of adults with major depressive disorder in the United States (2005 and 2010). J Clin Psychiatry. 2015;76(2):155-162. PubMed doi:10.4088/JCP.14m09298

J Clin Psychiatry 2016;77(8):e968-e974

Related Articles

Volume: 77

Quick Links: Eating

Continue Reading…