Received: 20 July 2018 Revised and accepted: 28 November 2018

DOI: 10.1002/eat.23004

ORIGINAL ARTICLE

Psychiatric and medical correlates of DSM-5 eating disorders

in a nationally representative sample of adults in the
United States
Tomoko Udo Ph.D.1 | Carlos M. Grilo Ph.D.2

1
Department of Health Policy, Management,
and Behavior, School of Public Health,
University at Albany, State University of
New York, Albany, New York
2
Department of Psychiatry, Yale University
School of Medicine, New Haven, Connecticut
Correspondence
Tomoko Udo, 1 University Place, Rensselaer,
NY 12144.
Email: tschaller@albany.edu
Abstract
Objective: To examine psychiatric and somatic correlates of DSM-5 eating disorders (EDs)—
anorexia nervosa (AN), bulimia nervosa (BN), and binge-eating disorder (BED)—in a nationally
representative sample of adults in the United States.
Method: A national sample of 36,309 adult participants in the national epidemiologic survey
on alcohol and related conditions III (NESARC-III) completed structured diagnostic interviews
(AUDADIS-5) to determine psychiatric disorders, including EDs, and reported 12-month diag-
nosis of chronic somatic conditions. Prevalence of lifetime psychiatric disorders and somatic
conditions were calculated across the AN, BN, and BED groups and a fourth group without
specific ED; multiple logistic regression models compared the likelihood of psychiatric/somatic
conditions with each specific ED relative to the no-specific ED group.
Results: All three EDs were associated significantly with lifetime mood disorders, anxiety
disorders, alcohol and drug use disorders, and personality disorders. In all three EDs, major
depressive disorder was the most prevalent, followed by alcohol use disorder. AN was associ-
ated significantly with fibromyalgia, cancer, anemia, and osteoporosis, and BED with diabetes,
hypertension, high cholesterol, and triglycerides. BN was not associated significantly with any
somatic conditions.
Conclusions: This study examined lifetime psychiatric and somatic correlates of DSM-5 AN, BN,
and BED in a large representative sample of U.S. adults. Our findings on significant associations
with other psychiatric disorders and with current chronic somatic conditions indicate the serious
burdens of EDs. Our findings suggest important differences across specific EDs and indicate
some similarities and differences to previous smaller studies based on earlier diagnostic criteria.

KEYWORDS

chronic somatic conditions, DSM-5, eating disorders, epidemiology, national representative

sample, psychiatric comorbidity

1 | I N T RO D UC T I O N
The section on feeding and eating disorders (EDs) in the Diagnostic and
Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) (American
Psychiatric Association, 2013) includes three specific ED diagnoses:
anorexia nervosa (AN), bulimia nervosa (BN), and binge-eating disorder
(BED). There were several important changes in the diagnoses and
required criteria sets for EDs between DSM-IV-TR (American Psychiatric
Association, 2004) and DSM-5. In the DSM-5, BED is included as a
specific diagnostic category for the first time and defined with a lower
frequency of once weekly binge eating and with a shorter duration
requirement than previously specified in the “provisional research criteria
set” in the DSM-IV-TR; these frequency and duration changes were
intended to parallel changes made to the BN diagnosis. In the DSM-5,
BN is defined with a lower frequency of once weekly for both binge eat-
ing and extreme weight compensatory behaviors for 3 months. Addition-
ally, AN no longer requires amenorrhea and now defines significantly low
weight as a weight that is below the minimal normal threshold for age.

42 © 2019 Wiley Periodicals, Inc. wileyonlinelibrary.com/journal/eat Int J Eat Disord. 2019;52:42–50.

UDO AND GRILO
New studies of prevalence and correlates of EDs are needed in
light of these changes in the DSM-5. A few epidemiological studies
have attempted to estimate the impact of changes between DSM-IV
and DSM-5 on the prevalence of AN, BN, and BED (Cossrow et al.,
2016; Hudson, Coit, Lalonde, & Pope Jr., 2012; Mohler-Kuo, Schny-
der, Dermota, Wei, & Milos, 2016; Trace et al., 2012). Those studies
suggested that DSM-5-based criteria for these three specific EDs
would yield higher prevalence estimates. Udo and Grilo (2018), in the first
study of lifetime and 12-month prevalence estimates of EDs in a large
nationally representative study with US adults, reported similar (slightly
higher) estimates for AN but lower estimates for BN and BED. To under-
stand a full scope of clinical and public health implications of changes in
diagnostic criteria for EDs, it is also important to evaluate their psychiatric
comorbidities and somatic correlates DSM-5 definitions. Previous
research has generally reported that EDs are frequently associated with
elevated rates of other psychiatric disorders, including mood disorders,
anxiety disorders, substance use disorders, impulse control disorders, and
personality disorders, across clinical, community, and epidemiological
samples (Grilo, White, Barnes, & Masheb, 2013; Hudson, Hiripi, Pope
Jr., & Kessler, 2007; Ulfvebrand, Birgegård, Norring, Högdahl, & von
Hausswolff-Juhlin, 2015; Welch et al., 2016). Recognizing and under-
standing associated psychiatric comorbidities is important for treatment
planning and informing research (Newman, Moffitt, Caspi, & Silva, 1998).
Similarly, understanding associations with medical or health problems can
inform improved screening, recognition, and medical management. Across
clinical and larger community or epidemiological studies, EDs have shown
significant, albeit highly variable, associations with various medical prob-
lems (Javaras et al., 2008; Kessler et al., 2013). Clinically, some of these
associations appear to partly reflect the symptomatic features (e.g., high
purging or laxative misuse associated with gastrointestinal issues)
(Brown & Mehler, 2015; Mehler & Rylander, 2015; Sato & Fukudo, 2015)
or associated features (e.g., excess weight and obesity associated with
BED also being associated with metabolic disturbances) (Kessler et al.,
2013). One recent study, based on the Swedish population registry,
reported that DSM-5 BED was associated with respiratory diseases,
endocrine system diseases, musculoskeletal system and connective tis-
sues diseases, and skin diseases (Thornton et al., 2017).
Thus, a little is known about psychiatric and somatic correlates of
DSM-5 EDs in the general population. The third wave of the national
epidemiologic survey on alcohol and related conditions (NESARC-III)
is the largest epidemiologic household survey of US adults that has
assessed EDs along with psychiatric and somatic correlates (Grant
et al., 2014). Thus, this study aimed to examine psychiatric and
somatic correlates of DSM-5 EDs in a nationally-representative sample
of US adults using data from the NESARC-III.
2 | METHOD
2.1 | Study sample
NESARC-III, designed originally to estimate the prevalence of alcohol
use and related conditions in adults, included 36,309 noninstitutiona-
lized US civilians 18 years and older (Grant et al., 2014; Grant et al.,
2016). The sample is composed of 56.3% women and, in terms of
43
ethnicity/race, 52.9% non-Hispanic White, 21.4% non-Hispanic Black,
19.4% Hispanic, and 6.4% “Other” groups; 22.4% of the respondents
were ages between 18 and 29, 27.9% were between ages 30 and
44, 26.7% were ages between 45 and 59, and 23.0% were above
60 years old (mean age = 45.6  17.5 years; see Grant et al., 2015,
for the weighted sample sociodemographic characteristics). Multi-
stage probability sampling was employed with counties or groups of
contiguous counties as primary sampling units, groups of Census-
defined blocks as secondary sampling units, and households within
secondary sampling units as tertiary sampling units. Eligible adults
were randomly selected from each household, but Hispanic, Black,
and Asian household members were oversampled (i.e., two respon-
dents from households with more than four eligible minority mem-
bers). NESARC had an overall response rate of 60.1% (72% household
response rate and 84% person-level response rate; see Grant et al.,
2016 for details). Between April 2012 and June 2013, participants
completed computer-assisted face-to-face personal interviews per-
formed by 970 trained lay-assessors who had, on average, 5 years of
experience performing health-related surveys. NERSAC-III received an
approval from the National Institute of Health (NIH) Institutional
Review Board (IRB) and participants provided oral informed consent
(Grant et al., 2016); IRB-exempt approval was obtained from Univer-
sity at Albany to perform the present analyses.
2.2 | Measures
2.2.1 | Diagnostic assessment of EDs and other
psychiatric disorders
A structured diagnostic interview, the NIAAA alcohol use disorder and
associated disabilities interview schedule-5 (AUDADIS-5; Grant et al.,
2011), was used to assess a range of DSM-5-defined psychiatric disor-
ders and their criteria, including specific information for AN, BN, and
BED diagnoses.
AUDADIS-5-generated lifetime psychiatric disorders as coded by
NESARC-III were used in this study. This included mood disorders
(major depressive episodes, persistent depression, and bipolar I), anxi-
ety disorders (specific phobia, social phobia, panic disorders, agora-
phobia, and generalized anxiety disorder), posttraumatic stress
disorder (PTSD), substance use disorders (SUD), alcohol use disorder
(AUD), drug use disorder, nicotine use disorder, personality disorders
(antisocial, borderline, and schizotypal), and conduct disorder. To
investigate the validity of the AUDADIS-5, it was compared with the
Psychiatric Research Interview for Substance and Mental Disorders,
DSM-5 version (PRISM-5) administered by clinicians blinded to
AUDADIS-5 findings; concordance was fair-to-moderate for diagnos-
tics (k = 0.24–0.72) and fair-to-excellent for dimensional measures of
SUD, mood disorders, anxiety disorders, and PTSD (ICC = 0.43–0.72;
Hasin et al., 2015; Hasin et al., 2015). Test–retest reliability of the
AUDADIS-5 for diagnosing specific disorders (non-ED categories) ranged
from fair-to-excellent (k = 0.35–0.87) for diagnosis of SUD, mood disor-
ders, anxiety disorders, PTSD, and personality disorders, and from good-
to-excellent for dimensional measures (ICC = 0.50–0.85; Grant
et al., 2015).
Unfortunately, the NESARC studies on validity and reliability of
the AUDADIS-5 did not include EDs. As we detailed in our previous
44
paper on the prevalence of EDs in NESARC-III (Udo & Grilo, 2018),
inspection of the NESARC dataset revealed various errors and mis-
classifications of ED diagnoses. We therefore did not utilize the ED
diagnosis variables provided by NESARC-III. Rather, we re-scored the
NESARC-III variable data to create DSM-5-based ED categories for
the analyses reported (see Supporting Information Table S1 for the
coding schemes).
2.2.2 | Chronic somatic conditions
NESARC-III interviews included questions about whether respondents
had been told by their doctor or other health professional that they
have any of 31 chronic somatic conditions during the past 12 months.
This study focused on the following chronic somatic conditions: arterio-
sclerosis, hypertension, diabetes, high cholesterol, high triglycerides,
myocardial infarction, other minor heart diseases (angina pectoris,
tachycardia, and other forms combined), stomach ulcer, stroke, arthritis,
cancer (liver, breast, mouth/tongue/throat/esophagus, and other types
of cancer combined), problems with falling or staying asleep, anemia,
fibromyalgia, irritable bowel syndrome or inflammatory bowel disease,
osteoporosis, lung problems (chronic bronchitis, emphysema, pneumo-
nia, or influenza), liver diseases (liver cirrhosis and other liver diseases
combined), and nerve problems (reflex sympathetic dystrophy, complex
regional pain syndrome, and other nerve problems in legs, arms or back
combined).
2.2.3 | Sociodemographic covariates
Respondents provided information about their sociodemographic sta-
tus, including age, sex, ethnicity/race (non-Hispanic White, non-
Hispanic Black, Hispanic, non-Hispanic Asian/Pacific Islander, and
American Indianan/Alaska Native), and education (categorized as less
than H.S., H.S. or GED, at least some college).
2.3 | Statistical analysis
Analyses were conducted with the Statistical Analysis System
(SAS) (release 9.4, 2002–2012), and accounted for NESARC-survey
design by using Proc Survey procedures with Taylor series variance
estimation method. For each ED, weighted means, frequencies and
cross-tabulations were computed for prevalence of comorbid psy-
chiatric and somatic conditions. Individuals without a lifetime his-
tory of AN, BN, or BED diagnosis served as the comparison group
(no-specific ED). For these inferential statistics comparing these
three specific ED groups and no-specific ED group, we followed
the well-established diagnostic “hierarchy” of AN>BN > BED
(i.e., lifetime BN excluded those with lifetime AN, lifetime BED
excluded those with lifetime AN/BN).
Multiple logistic regression models were used to calculate
adjusted odds ratios (AORs) to compare risk of lifetime diagnosis of
each psychiatric disorder and past 12-month diagnosis of somatic
conditions with adjustment for age, sex, race/ethnicity, and education
(Model 1). An additional adjustment for lifetime diagnosis of other
psychiatric disorder (included as Yes/No for any mood disorder, any
anxiety disorder, any SUD, PTSD, any personality or conduct disorder
[all lifetime diagnoses]) was used to be consistent with prior studies
on the epidemiology of various psychiatric disorders based on this
UDO AND GRILO
data set (Grant et al., 2015; Grant et al., 2016; Hasin et al., 2016;
Hasin et al., 2018), and to investigate the statistically independent
associations between three specific EDs and their psychiatric and
somatic correlates (Model 2).
3 | RE SU LT S
Lifetime prevalence of AN, BN, and BED was 0.80% (0.07%), 0.28%
(0.03%), and 0.85% (0.05%), respectively (see Udo & Grilo, 2018, for
more information on the prevalence estimates, including by sex and
by ethnicity/race).
3.1 | Lifetime psychiatric comorbidity
Table 1 summarizes lifetime prevalence estimates (%) and standard
errors of each psychiatric disorder across lifetime AN, BN, BED, and
the no-specific ED groups. 87.3% of respondents with AN, 94.4% with
BN, and 93.8% with BED met criteria for at least one additional
NESARC-III-assessed DSM-5 lifetime psychiatric disorder. The lifetime
prevalence rates for all psychiatric disorders were substantially higher
in the ED groups than the no-specific ED group. Overall, all three ED
groups had a significantly greater number of comorbid lifetime psychi-
atric conditions than the no-specific ED group; the lifetime BED group
met criteria for significantly greater number of lifetime psychiatric
diagnoses than the lifetime AN group (2.3 vs 1.7; t[113] = −3.19,
p = .001) after adjusting for sociodemographic variables.
As shown in Table 2, after adjusting for sociodemographic vari-
ables, the three EDs were significantly associated with all other
lifetime psychiatric disorders (except for BN with specific phobia).
When additionally adjusting for other lifetime psychiatric disorders
(Model 2), AN was significantly associated with any mood disor-
ders, major depressive disorder, persistent depression, panic disor-
der, agoraphobia, PTSD, any SUDs, AUD, any personality/conduct
disorder, borderline personality disorder, and conduct disorder. BN
was significantly associated with any mood disorders, major depressive
disorder, persistent depression, AUD, borderline personality disorder,
and conduct disorder. BED was significantly associated with any mood
disorders, major depressive disorder, persistent depression, any anxiety
disorders, all individual anxiety disorders (except for panic disorder),
PTSD, AUD, any personality or conduct disorders, and all individual per-
sonality and conduct disorders.
3.2 | Chronic somatic conditions
Table 3 summarizes prevalence estimates (%) and standard errors of
each chronic somatic condition across each specific ED and no-specific
ED groups. Across all three specific EDs, over half of respondents
reported having been diagnosed with at least one somatic condition in
the past 12 months (59.7  3.51% for AN, 54.5  5.06% for BN, and
68.6  3.03% for BED). The total number of diagnosed chronic somatic
conditions in the past 12 months was significantly greater for AN and
BED than the no-specific ED group.
As shown in Table 4, after adjusting for sociodemographic vari-
ables, AN was associated with significantly greater odds of reporting
UDO AND GRILO 45

TABLE 1 Lifetime prevalence (% [SE]) of comorbid DSM-5 psychiatric disorders across anorexia nervosa (AN), bulimia nervosa (BN), binge-eating
disorder (BED), and no-specific ED groups
AN BN BED No-specific ED
(n = 276) (n = 92) (n = 318) (n = 35,709)
Any mood disorders 54.2 (2.83) 79.6 (3.09) 69.9 (2.70) 23.6 (0.41)
Major depressive disorder 49.5 (2.69) 76.3 (3.16) 65.5 (2.72) 21.7 (0.39)
Persistent depression 22.4 (2.72) 34.8 (4.98) 32.8 (2.82) 6.0 (0.20)
Bipolar I 8.1 (1.99) 13.0 (3.53) 8.7 (1.76) 2.0 (0.1)
Any anxiety disorders 40.5 (3.21) 44.6 (5.02) 59.0 (2.72) 16.4 (0.31)
Panic disorder 21.0 (2.95) 17.8 (4.29) 22.5 (2.30) 5.0 (0.17)
Agoraphobia 10.8 (2.92) 9.7 (1.51) 13.4 (1.96) 1.8 (0.09)
Social anxiety disorder 8.7 (1.78) 14.1 (2.58) 20.6 (2.62) 3.5 (0.15)
Specific phobia 15.3 (2.51) 14.4 (2.91) 24.4 (2.35) 6.2 (0.15)
General anxiety disorder 21.9 (2.94) 25.8 (4.64) 33.0 (2.33) 7.4 (0.20)
Posttraumatic stress disorder 22.7 (3.25) 32.4 (5.54) 31.6 (2.50) 5.8 (0.21)
Any substance use disorder 60.3 (3.33) 67.0 (4.73) 67.7 (2.72) 42.8 (0.55)
Alcohol use disorder 49.2 (2.67) 61.0 (4.96) 52.0 (2.96) 28.7 (0.49)
Nicotine use disorder 38.5 (3.90) 42.5 (5.12) 40.2 (2.96) 27.7 (0.52)
Other drug use disorder 17.6 (2.35) 29.5 (4.29) 24.7 (2.25) 9.7 (0.27)
Any personality or conduct disorder 35.2 (3.39) 51.1 (5.61) 56.0 (3.04) 15.2 (0.40)
Antisocial 8.7 (2.18) 10.6 (2.41) 15.8 (2.19) 4.2 (0.16)
Borderline 30.3 (3.08) 48.0 (5.64) 49.2 (3.30) 11.0 (0.31)
Schizotypal 16.9 (2.38) 24.6 (4.32) 27.8 (2.53) 6.1 (0.23)
Conduct 10.4 (2.26) 13.1 (2.44) 16.0 (2.20) 4.7 (0.18)
Any disorder 83.7 (2.05) 94.4 (1.27) 93.8 (1.40) 57.6 (0.59)
Mean total # of comorbid psychiatric conditionsb 1.7 (0.15)* 2.1 (0.20)* 2.3 (0.14)a,* 0.6 (0.01)

Notes. CI = confidence interval. All analyses adjusted for complex survey design of NESARC-III. The total sample size added across four groups is 36,395
as some respondents met criteria for more than one ED. If we apply the hierarchical rule, the sample sizes of each group are: AN = 276, BN = 77 (excluded
those who met lifetime AN diagnosis), BED = 247 (excluded those who met lifetime diagnosis of AN or BN), and No-specific ED group = 35,709.
a
Significantly different from AN at p < .05 based on Tukey–Kramer post hoc test.
b
Adjusted for sociodemographic variables.
*Significantly different from No-specific ED at p < .05 based on Tukey–Kramer post hoc test.

minor heart conditions, stomach ulcer, sleep problems, cancer, anemia,
fibromyalgia, osteoporosis, and nerve problems. When additionally
adjusting for other psychiatric conditions, odds ratios for fibromyalgia,
cancer, anemia, and osteoporosis remained significant in AN. When
adjusting only for sociodemographic variables, BN was significantly
associated with increased odds of arteriosclerosis, minor heart condi-
tions, stomach ulcer, epilepsy or seizure, and sleep problems. All asso-
ciations became nonsignificant after additionally adjusting for other
psychiatric disorders, however, odds of fibromyalgia, became signifi-
cantly less for BN. After adjusting for sociodemographic variables,
BED was significantly associated with increased odds of diabetes,
hypertension, high cholesterol, high triglycerides, minor heart condi-
tions, stomach ulcer, arthritis, sleep problems, anemia, fibromyalgia,
bowel problems, osteoporosis, lung problems, liver diseases, and nerve
problems. Finally, after additionally adjusting for other psychiatric dis-
orders, odds ratios for diabetes, hypertension, high cholesterol, and
high triglycerides remained significant.
sample of adults in the US. Overall, we found that DSM-5 EDs were
significantly and highly associated with other psychiatric disorders.
While only 57.6% of individuals with no history of EDs met criteria for
at least one lifetime psychiatric diagnosis, among the ED diagnoses,
87.3% of AN, 94.4% of BN, and 93.8% BED met criteria for an addi-
tional psychiatric disorder. Overall, these findings are generally consis-
tent with previous, a smaller nationally-representative US study
(Hudson et al., 2007).
Overall the most frequently occurring comorbid diagnoses with
three specific EDs were mood disorders and substance use disorders
(both occurring in roughly half to two-thirds of cases), followed by
anxiety disorders (occurring in nearly half of cases). For all specific
EDs, odds of lifetime diagnosis were significantly increased for major
depressive disorder, persistent depression, AUD, borderline personality
disorder, and conduct disorder, even after adjusting for other psychiat-
ric disorders. BED had the greatest number of significant associations
with other psychiatric disorders and BN had the least.
We note some divergences in our observed psychiatric comorbid-

4 | DISCUSSION ity patterns from those reported previously with DSM-IV (Hudson
et al., 2007). We found a higher rate of psychiatric comorbidity for
This study reports psychiatric comorbidities and somatic conditions DSM-5-defined AN and BN than the previous NCS-R study (Hudson
associated with DSM-5-defined EDs in a nationally-representative et al., 2007) and our rate of substance use disorders across all EDs
 46

TABLE 2 Lifetime comorbidity (adjusted odds ratios) of DSM-5 eating disorders with other psychiatric disorders adjusting for Sociodemographic variables (Model 1) plus other psychiatric disorders
(Model 2)
AN BN BED
Model 1 Model 2 Model 1 Model 2 Model 1 Model 2
Any mood disorders 2.66 (2.09–3.38) 1.61* (1.17–2.34) 9.30* (5.34–16.18) 5.39* (2.82–10.29) 6.36* (4.75–8.52) 2.51* (1.76–3.60)
Major depressive disorder 2.40* (1.90–3.04) 1.45** (1.02–2.06) 8.49* (4.96–14.53) 4.93* (2.65–9.18) 5.73* (4.33–7.60) 2.36* (1.66–3.26)
Persistent depression 3.42* (2.33–5.02) 1.87** (1.12–3.12) 6.50a (3.64–11.61) 3.14** (1.30–7.63) 6.50a (4.78–9.01) 2.17a (1.49–3.16)
a a *
Bipolar I 4.22 (2.35–7.57) 1.61 (0.89–2.92) 6.51 (3.09–13.72) 1.93 (0.85–4.38) 4.58 (2.93–7.16) 1.08 (0.65–1.79)
Any anxiety disorders 2.44* (1.75–3.40) 1.34 (0.91–1.99) 3.00* (1.69–5.34) 0.94 (0.42–2.11) 6.21* (4.77–8.06) 2.33* (1.74–3.12)
* ** * *
Panic disorder 3.35 (2.28–4.93) 1.77 (1.14–2.75) 2.72 (1.42–5.23) 0.83 (0.39–1.78) 4.37 (3.18–5.99) 1.44 (1.00–2.08)
Agoraphobia 4.52* (2.38–8.59) 2.15** (1.13–4.10) 3.92* (1.81–8.49) 1.14 (0.47–2.76) 6.76* (4.29–10.67) 2.08* (1.25–3.45)
* * *
Social anxiety disorder 2.00 (1.24–3.22) 0.95 (0.57–1.57) 3.47 (1.76–6.85) 1.10 (0.50–2.41) 6.29 (4.46–8.89) 2.17 (1.47–3.20)
Specific phobia 1.92* (1.27–2.91) 1.17 (0.78–1.74) 1.81 (0.94–3.49) 0.78 (0.38–1.61) 3.99* (2.97–5.37) 1.81* (1.32–2.50)
* * *
General anxiety disorder 2.59 (1.76–3.81) 1.28 (0.83–1.98) 3.41 (1.82–6.36) 1.03 (0.45–2.39) 5.28 (4.05–6.88) 1.70* (1.23–2.35)
Posttraumatic stress disorder 3.46* (2.28–5.25) 1.78* (1.18–2.68) 5.54* (3.04–10.12) 1.96 (0.99–3.88) 6.24* (4.66–8.35) 1.74* (1.21–2.50)
* ** * *
Any substance use disorder 2.29 (1.65–3.18) 1.55 (1.08–2.22) 2.79 (1.61–4.82) 1.34 (0.70–2.54) 3.08 (2.29–4.13) 1.31 (0.93–1.85)
Alcohol use disorder 2.69* (2.02–3.59) 1.87* (1.37–2.54) 3.95* (2.29–6.80) 2.07** (1.12–3.81) 3.00* (2.29–3.92) 1.37** (1.02–1.85)
* ** *
Nicotine use disorder 1.87 (1.32–2.65) 1.27 (0.85–1.89) 2.11 (1.19–3.75) 1.06 (0.55–2.06) 1.90 (1.42–2.54) 0.85 (0.62–1.18)
Other drug use disorder 2.22* (1.51–3.27) 1.17 (0.80–1.71) 3.86* (2.24–6.68) 1.48 (0.80–2.76) 3.36* (2.56–4.41) 1.14 (0.82–1.58)
* ** * *
Any personality or conduct disorder 3.18 (2.29–4.40) 1.69 (1.08–2.66) 5.56 (3.36–9.20) 1.56 (0.79–3.08) 7.40 (5.64–9.71) 2.68* (1.93–3.72)
* * *
Antisocial 3.48 (1.95–6.22) 1.58 (0.85–2.93) 3.54 (1.78–7.03) 1.26 (0.62–2.58) 5.68 (3.85–8.37) 2.22* (1.43–3.45)
* ** * * *
Borderline 3.34 (2.38–4.67) 1.68 (1.07–2.34) 6.59 (4.01–10.82) 2.56 (1.52–4.33) 7.75 (5.83–10.31) 2.60* (1.83–9.71)
Schizotypal 3.25* (2.23–4.72) 1.43 (0.92–2.23) 4.63* (2.61–8.20) 1.57 (0.76–3.24) 6.06* (4.52–8.13) 1.86* (1.26–2.63)
* ** * * *
Conduct 3.73 (2.26–6.14) 1.80 (1.08–3.02) 4.05 (2.15–7.62) 2.16 (1.28–3.64) 5.09 (3.45–7.50) 2.06* (1.34–3.16)

Notes. AOR = adjusted odds ratio. CI = 95% confidence interval. Model 1 = adjusting for age, sex, race/ethnicity, and education. Model 2 = adjusting for sociodemographic variables and other psychiatric conditions
(any mood disorders, any anxiety disorders, any substance use disorders, any personality/conduct disorders, and PTSD). All analyses adjusted for complex survey design of NESARC-III.
*Significant at p < .01.; **Significant at p < .05.
Significant results are highlighted by bold letters.
UDO AND GRILO
UDO AND GRILO 47

TABLE 3 Prevalence (% [SE]) of comorbid somatic conditions across DSM-5 lifetime anorexia nervosa (AN), bulimia nervosa (BN), binge-eating
disorder (BED), and no-specific ED groups
AN BN BED No-specific ED
(n = 276) (n = 92) (n = 318) (n = 35,709)
Arteriosclerosis 0.5 (0.03) 2.0 (1.01) 1.0 (0.51) 1.4 (0.09)
Diabetes 7.2 (2.03) 4.6 (1.11) 13.6 (2.2) 9.3 (0.24)
Hypertension 18.4 (2.18) 11.6 (2.22) 31.2 (3.01) 25.0 (0.42)
High cholesterol 17.1 (2.79) 11.2 (3.07) 27.2 (2.84) 19.9 (0.38)
High triglycerides 6.7 (2.45) 6.5 (1.56) 14.5 (2.10) 8.4 (0.25)
Myocardial infarction 0.5 (0.03) 0.0 (0.0) 0.8 (0.64) 0.8 (0.06)
Other heart conditionsa 13.6 (2.16) 13.6 (6.76) 17.2 (2.19) 9.8 (0.25)
Stomach ulcer 6.5 (1.89) 8.8 (4.16) 5.2 (1.23) 2.4 (0.10)
Epilepsy or seizure 1.2 (0.37) 3.0 (1.00) 1.0 (0.59) 0.8 (0.05)
Arthritis 19.7 (3.05) 13.1 (4.10) 24.0 (2.57) 18.7 (0.34)
Stroke 0.9 (0.34) 0.0 (0.00) 0.2 (0.15) 0.8 (0.05)
Sleep problems 17.4 (2.32) 17.8 (2.48) 21.3 (2.74) 7.4 (0.23)
b
Cancer 6.3 (1.64) 2.5 (0.16) 4.4 (0.90) 4.1 (0.16)
Anemia 13.1 (2.14) 9.4 (3.33) 10.6 (1.70) 4.8 (0.14)
Fibromyalgia 8.1 (2.05) 0.5 (0.03) 5.3 (1.35) 2.0 (0.09)
Bowel problemsc 8.0 (1.54) 8.4 (1.84) 11.9 (2.00) 3.7 (0.13)
Osteoporosis 11.1 (2.23) 1.9 (1.34) 6.1 (1.54) 3.8 (0.13)
Lung problems 4.6 (0.93) 6.1 (2.49) 11.0 (1.77) 5.2 (0.20)
Liver diseasesd 2.2 (0.87) 2.2 (0.93) 2.6 (1.00) 1.2 (0.08)
Nerve problemse 14.2 (2.57) 10.4 (3.06) 16.0 (2.84) 10.2 (0.30)
f
Mean total # of chronic somatic conditions 1.9 (0.19)* 1.7 (0.25) 2.3 (0.18)* 1.4 (0.02)

Notes. All analyses adjusted for complex survey design of NESARC-III. The total sample size added across four groups is 36,395 as some respondents met
criteria for more than one ED. If we apply the hierarchical rule, the sample sizes of each group are: AN = 276, BN = 77 (excluded those who met lifetime
AN diagnosis), BED = 247 (excluded those who met lifetime diagnosis of AN or BN), and no-specific ED group = 35,709.
a
Other heart conditions included angina pectoris, tachycardia, and other forms.
b
Cancer included liver, breast, mouth/tongue/throat/esophagus, other.
c
Bowel problems included inflammatory bowel disease and irritable bowel syndrome.
d
Liver diseases included liver cirrhosis and other liver diseases.
e
Nerve problems included reflex sympathetic dystrophy, complex regional pain syndrome and other nerve problems in legs, arms or back.
f
Adjusted for sociodemographic variables.
*Significantly different from no-specific ED group at p < .05 based on Tukey–Kramer post hoc test.

was greater. Differences in patterns of associations with previous epi-
demiological estimates (Hudson et al., 2007; Ulfvebrand et al., 2015)
may be the result of changes in the DSM-5 diagnostic criteria not only
for EDs but also for other psychiatric disorders (e.g., removal of abuse
vs. dependence distinction in substance use disorders; Bartoli, Carrà,
Crocamo, & Clerici, 2015; Hasin et al., 2013). Particularly for SUD,
changes in the strength of associations may be attributable partly to
general increases in its prevalence (e.g., Grant et al., 2017; Hasin et al.,
2016). Differences might also be partly due to methodological differ-
ences, such as analytic approaches (i.e., we included other psychiatric
comorbidities as covariates) and diagnostic instruments (World Health
Organization Composite International Diagnostic Interview [CIDI; Kess-
ler & Ustun, 2004] vs. AUDADIS-5 [Grant et al., 2011]).
DSM-5 EDs were also associated with high rates of chronic somatic
conditions. Lifetime diagnoses of AN and BED were associated with
significantly greater number of chronic somatic conditions than those
without lifetime diagnosis of three specific EDs. After adjusting for
sociodemographic characteristics, AN was associated with significantly
increased odds of minor heart conditions, stomach ulcer, sleep prob-
lems, anemia, cancer, fibromyalgia, osteoporosis, and nerve problems.
Even when additionally adjusting for other psychiatric disorders, only
fibromyalgia, cancer, anemia, and osteoporosis remained significantly
associated with AN. We emphasize that the cross-sectional nature of
the data analyses precludes any firm interpretations. We cautiously
note that some of these findings might fit clinical reports. For example,
our finding regarding osteoporosis in AN is well-documented
(Robinson, Aldridge, Clark, Misra, & Micali, 2016). The findings of a pos-
sible association between cancer and AN has been reported inconsis-
tently in the literature with most studies focusing specifically on breast
cancer (Mellemkjaer et al., 2001; Michels & Ekbom, 2004; O'Brien,
Whelan, Sandler, & Weinberg, 2017). A study pooled population medi-
cal registry data from Finland, Denmark, and Sweden suggested signifi-
cantly increased incidence rate ratios, particularly for esophageal, lung,
and liver cancer (Mellemkjaer et al., 2015).
BED was associated with significantly increased odds of many
somatic conditions including minor heart conditions, stomach ulcer,
diabetes, hypertension, high cholesterol, high triglycerides, arthritis,
anemia, fibromyalgia, bowel problems, osteoporosis, lung problems,
liver diseases, and nerve problems. These findings are generally
consistent with, but expand upon, those of the WHO Mental Health
Survey (Kessler et al., 2013) and with medical findings highlighted in a
recent review (Olguin et al., 2016) which concluded that type-2
48 UDO AND GRILO

TABLE 4 Adjusted odds ratios of comorbid somatic conditions across DSM-5 lifetime eating disorders

AN BN BED
Model 1 Model 2 Model 1 Model 2 Model 1 Model 2
Arteriosclerosis 0.57 (0.08–4.18) 0.57 (0.08–4.18) 4.52* (1.07–19.15) 2.71 (0.59–12.50) 1.35 (0.49–3.70) 0.81 (0.28–2.32)
Diabetes 1.24 (0.63–2.46) 1.04 (0.55–1.96) 0.92 (0.39–2.17) 0.67 (0.29–1.62) 2.18** (1.52–3.13) 1.59* (1.11–2.29)
Hypertension 1.13 (0.77–1.64) 0.92 (0.63–1.35) 0.78 (0.43–1.43) 0.58 (0.33–1.03) 2.03** (1.55–2.67) 1.44* (1.08–1.93)
**
High cholesterol 1.16 (0.72–1.87) 0.96 (0.60–1.52) 0.87 (0.37–2.04) 0.64 (0.28–1.48) 2.01 (1.48–2.75) 1.43* (1.04–1.95)
**
High triglycerides 0.92 (0.41–2.05) 0.79 (0.36–1.74) 1.18 (0.49–2.82) 1.31 (0.35–1.91) 2.39 (1.62–3.51) 1.54* (1.03–2.30)
Myocardial infarction 1.06 (0.14–7.84) 0.82 (0.11–5.90) — — 1.47 (0.29–7.37) 0.97 (0.19–5.04)
a ** * **
Other heart conditions 1.82 (1.19–2.78) 1.25 (0.81–1.92) 2.14 (1.03–4.45) 1.26 (0.63–2.51) 2.25 (1.62–3.13) 1.18 (0.84–1.66)
Stomach ulcer 2.83** (1.46–5.47) 1.75 (0.89–3.43) 4.13** (1.42–12.05) 2.13 (0.64–7.02) 2.18** (1.30–3.66) 0.98 (0.58–1.68)
Epilepsy or seizure 1.34 (0.48–3.73) 0.85 (0.30–2.42) 3.63* (1.14–11.58) 1.81 (0.53–6.16) 1.16 (0.36–3.68) 0.53 (0.16–1.75)
Arthritis 1.44 (0.90–2.32) 1.06 (0.68–1.65) 1.18 (0.62–2.25) 0.75 (0.39–1.43) 1.78** (1.32–2.41) 1.01 (0.74–1.38)
Stroke 2.00 (0.55–7.26) 1.39 (0.39–4.98) — — 0.27 (0.04–1.96) 0.14 (0.02–1.02)
Sleep problems 2.43** (1.67–3.56) 1.49 (1.03–2.16) 2.79** (1.69–4.59) 1.29 (0.77–2.16) 3.28** (2.35–4.57) 1.38 (0.95–2.00)
Cancerb 2.41** (1.25–4.66) 2.00* (1.04–3.87) 1.14 (0.37–3.50) 0.94 (0.31–2.84) 1.41 (0.70–2.84) 1.00 (0.51–1.97)
**
Anemia 2.10 (1.39–3.16) 1.49* (1.03–2.15) 1.47 (0.65–3.32) 0.86 (0.40–1.86) 1.85** (1.28–2.68) 0.97 (0.66–1.43)
** * * *
Fibromyalgia 2.79 (1.57–4.95) 1.82 (1.04–3.17) 0.17 (0.02–1.24) 0.09 (0.01–0.63) 2.08 (1.14–3.79) 0.90 (0.48–1.69)
Bowel problemsc 1.69* (1.04–2.74) 1.09 (0.66–1.78) 2.05 (0.97–4.35) 1.07 (0.49–2.31) 2.99** (1.98–4.50) 1.35 (0.88–2.07)
Osteoporosis 3.87** (2.14–7.01) 3.09** (1.76–5.43) 0.79 (0.17–3.75) 0.50 (0.11–2.27) 1.89* (1.13–3.17) 1.08 (0.65–1.81)
Lung problems 0.90 (0.51–1.58) 0.57** (0.32–1.00) 1.44 (0.50–4.17) 0.80 (0.29–2.22) 2.37** (1.59–3.52) 1.18 (0.77–1.81)
d *
Liver diseases 2.37 (0.92–6.13) 1.34 (0.51–3.56) 2.61 (0.88–7.77) 1.28 (0.43–3.78) 2.56 (1.13–5.79) 1.15 (0.49–2.69)
Nerve problemse 1.67* (1.05–2.66) 1.09 (0.68–1.75) 1.33 (0.65–2.73) 0.71 (0.34–1.48) 1.84** (1.18–2.89) 0.88 (0.56–1.41)

Notes. AOR = adjusted odds ratio. CI = confidence interval. Models were not valid for myocardial infarction and stroke in BN due to no positive cases.
Model 1 = adjusting for age, sex, race/ethnicity, educational level. Model 2 = additionally adjusting for other psychiatric disorders (any mood disorders,
any anxiety disorders, any substance use disorders, any personality/conduct disorders, and PTSD). All analyses adjusted for complex survey design of
NESARC-III.
a
Minor heart conditions included angina pectoris, tachycardia, and other forms.
b
Cancer included liver, breast, mouth/tongue/throat/esophagus, other.
c
Bowel problems included inflammatory bowel disease and irritable bowel syndrome.
d
Liver diseases included liver cirrhosis and other liver diseases.
e
Nerve problems included reflex sympathetic dystrophy, complex regional pain syndrome, and other nerve problems in legs, arms or back.
*Significant at p < .05.; **Significant at p < .01.

diabetes, hypertension, dyslipidemias, sleep problems, pain conditions,
and gastrointestinal symptoms are commonly comorbid somatic con-
ditions in BED. In addition, Olguin et al. (2016) suggested that more
studies are needed to determine whether the associations between
BED and these somatic conditions are above and beyond the effects
of obesity or other psychiatric comorbidity. Importantly, our current
analyses suggest that many somatic comorbidities in BED exist after
adjusting for other psychiatric conditions. Moreover, as we emphasize
below, the causal nature of these associations is unclear.
Strengths of the current study include the use of a large epidemi-
ological data set with a representative sample of 36,309 US adults
and broad coverage of comorbid psychiatric and chronic somatic con-
ditions. Our large sample provided greater precision (i.e., leading to
smaller confidence intervals and greater ability to detect smaller
effects) than previous studies and allowed reasonably for adjusting for
other psychiatric disorders in addition to sociodemographic factors.
The study also has potential limitations. The AUDADIS-5 structured
instrument was administered by lay, albeit experienced, interviewers,
not by clinicians; standardized training, however, may offset this limi-
tation to some degree. The AUDADIS-5 has not been evaluated for
reliability and validity of ED diagnosis, although it has been validated
for other psychiatric conditions (Grant, Goldstein, Saha, et al., 2015;
Hasin et al., 2015); it is therefore important that the AUDADIS-5 will
be validated for DSM-5 ED diagnoses. The skip-out rules and the lack
of specific details in the questions within ED sections in the
AUDADIS-5 might have resulted in possible underestimation of the
prevalence rates due to missing data, and precluded us from investi-
gating subthreshold BN and BED, as well as other specified feeding or
eating disorder (OSFSD) (Supporting Information Tables S1 and S2).
Collectively, these limitations highlight the complexity of assessment
issue in epidemiological research that are not limited to the AUDADIS-
5, and represent an important challenge for the field and provide critical
context for interpreting results (Udo & Grilo, in press).
The NESARC-III relied on self-report for diagnoses of chronic
somatic conditions rather than medical records; this may introduce
some degree of bias although it is a standard approach in large-scale
studies. Research should examine validity and accuracy of such self-
reported diagnoses of somatic conditions. Importantly, the causal
nature of the observed associations is unclear. Future research should
examine temporal aspects or sequencing for the ED and psychiatric
conditions even though such analyses would be limited in how they
might speak to causality (e.g., they might provide clues for hypotheses
about potential direct effects but not about other common causes of
both disorders). Due to limited sample sizes for men and racial minori-
ties with positive AN, BN, or BED diagnosis, the study was unable to
investigate sex and racial/ethnic differences in the relationships
UDO AND GRILO
between three specific EDs, other psychiatric disorders, and chronic
somatic conditions, and thus further epidemiological research to inves-
tigate this question is warranted. Although the NESARC-III is a carefully
designed study and allows adjustment of population rates and other
complex survey designs in calculation of prevalence, the sample may
not be perfectly representative of the US adult population for the limi-
tations commonly reported in many census studies (e.g., focus on non-
institutionalized individuals, possible differences between those who
are and are not willing to complete a lengthy survey).
In conclusion, our findings for DSM-5-defined AN, BN, and BED,
based on a large national sample of US adults, indicate that these dis-
orders are associated with significantly elevated rates of other major
psychiatric disorders and somatic conditions. The elevated rates of
comorbidities were observed even after adjusting for sociodemo-
graphic characteristics and other psychiatric disorders. EDs are preva-
lent across different sociodemographic groups, and are associated
with serious impairment in psychosocial functioning (Udo & Grilo,
2018). Our findings underscore the importance of screening of addi-
tional psychiatric and somatic conditions in those identified with EDs
to reduce the disease burdens due to medical comorbidities.
ACKNOWLEDGMENTS
This manuscript was prepared using a limited access dataset obtained
from the National Institute on Alcohol Abuse and Alcoholism (NIAAA).
This manuscript does not reflect the opinions or views of the NIDDK,
NIAAA, or the US government. The authors report no biomedical
financial interests or relevant direct or indirect conflicts of interest.
ORCID
Tomoko Udo https://orcid.org/0000-0002-5319-2689
Carlos M. Grilo https://orcid.org/0000-0003-0245-3444
RE FE R ENC E S
American Psychiatric Association. (2004). Diagnostic and statistical manual
of mental disorders (4th ed.). D. C: Washington.
American Psychiatric Association. (2013). Diagnostic and statistical manual
of mental disorders (5th ed.). D. C: Washington.
Bartoli, F., Carrà, G., Crocamo, C., & Clerici, M. (2015). From DSM-IV to
DSM-5 alcohol use disorder: An overview of epidemiological data.
Addictive Behaviors, 41, 46–50. https://doi.org/10.1016/j.addbeh.
2014.09.029
Brown, C., & Mehler, P. S. (2015). Medical complications of anorexia ner-
vosa and their treatments: An update on some critical aspects. Eating
and Weight Disorders, 20(4), 419–425. https://doi.org/10.1007/
s40519-015-0202-3
Cossrow, N., Pawaskar, M., Witt, E. A., Ming, E. E., Victor, T. W.,
Herman, B. K., … Erder, M. H. (2016). Estimating the prevalence of
binge eating disorder in a community sample from the United States:
Comparing DSM-IV-TR and DSM-5 criteria. Journal of Clinical Psychia-
try, 77(8), e968–e974. https://doi.org/10.4088/JCP.15m10059
Grant, B. F., Chou, S. P., Saha, T. D., Pickering, R. P., Kerridge, B. T.,
Ruan, W. J., … Hasin, D. S. (2017). Prevalence of 12-month alcohol use,
high-risk drinking, and DSM-IV alcohol use disorder in the United
States, 2001-2002 to 2012-2013: Results from the National Epidemio-
logic Survey on alcohol and related conditions. JAMA Psychiatry, 74(9),
911–923. https://doi.org/10.1001/jamapsychiatry.2017.2161
Grant, B. F., Chu, A., Sigman, R. A. M., Kali, J., Sugawara, Y., Jiao, R., …
Goldstein, R. (2014). Sources and accuracy statement: National Epidemi-
ologic Survey on alcohol and related conditions-III (NESARC-III). Rockville,
49
MD: National Institute on Alcohol Abuse and Alcoholism. Retrieved
from http://www.niaaa.nih.gov/sites/default/files/NESARC_Final_Report_
FINAL_1_8_15.pdf
Grant, B. F., Goldstein, R. B., Chou, S. P., Saha, T. D., Ruan, W. J.,
Huang, B., … Hasin, D. S. (2011). The alcohol use disorder and associated
disabilities interview schedule-diagnostic and statistical manual of mental
disorders, fifth edition version (AUDADIS-5). Rockville, MD: National
Institute on Alcohol Abuse and Alcoholism.
Grant, B. F., Goldstein, R. B., Saha, T. D., Chou, S. P., Jung, J., Zhang, H., …
Hasin, D. S. (2015). Epidemiology of DSM-5 alcohol use disorder:
Results from the national epidemiologic survey on alcohol and related
conditions III. JAMA Psychiatry, 72(8), 757–766. https://doi.org/10.
1001/jamapsychiatry.2015.0584
Grant, B. F., Goldstein, R. B., Smith, S. M., Jung, J., Zhang, H., Chou, S. P., …
Hasin, D. S. (2015). The alcohol use disorder and associated disabilities
interview Schedule-5 (AUDADIS-5): Reliability of substance use
and psychiatric disorder modules in a general population sample.
Drug and Alcohol Dependence, 148, 27–33. https://doi.org/10.
1016/j.drugalcdep.2014.11.026
Grant, B. F., Saha, T. D., Ruan, W. J., Goldstein, R. B., Chou, S. P., Jung, J.,
… Hasin, D. S. (2016). Epidemiology of DSM-5 drug use disorder:
Results from the National Epidemiologic Survey on alcohol and related
conditions-III. JAMA Psychiatry, 73(1), 39–47. https://doi.org/10.1001/
jamapsychiatry.2015.2132
Grilo, C. M., White, M. A., Barnes, R. D., & Masheb, R. M. (2013). Psychiat-
ric disorder co-morbidity and correlates in an ethnically diverse sample
of obese patients with binge eating disorder in primary care settings.
Comprehensive Psychiatry, 54, 209–216. https://doi.org/10.1016/j.
comppsych.2012.07.012
Hasin, D. S., Greenstein, E., Aivadyan, C., Stohl, M., Aharonovich, E.,
Saha, T., … Grant, B. F. (2015). The alcohol use disorder and associated
disabilities interview Schedule-5 (AUDADIS-5): Procedural validity of
substance use disorders modules through clinical re-appraisal in a gen-
eral population sample. Drug and Alcohol Dependence, 148, 40–46.
https://doi.org/10.1016/j.drugalcdep.2014.12.011
Hasin, D. S., Kerridge, B. T., Saha, T. D., Huang, B., Pickering, R.,
Smith, S. M., … Grant, B. F. (2016). Prevalence and correlates of DSM-
5 cannabis use disorder, 2012-2013: Findings from the national epide-
miologic survey on alcohol and related conditions-III. American Journal
of Psychiatry, 173(6), 588–599. https://doi.org/10.1176/appi.ajp.2015.
15070907
Hasin, D. S., O'Brien, C. P., Auriacombe, M., Borges, G., Bucholz, K.,
Budney, A., … Grant, B. F. (2013). DSM-5 criteria for substance use dis-
orders: Recommendations and rationale. American Journal of Psychiatry,
170(8), 834–851. https://doi.org/10.1176/appi.ajp.2013.12060782
Hasin, D. S., Sarvet, A. L., Meyers, J. L., Saha, T. D., Ruan, W. J.,
Stohl, M., & Grant, B. F. (2018). Epidemiology of adult DSM-5 major
depressive disorder and its specifiers in the United States. JAMA Psy-
chiatry, 75(4), 336–346. https://doi.org/10.1001/jamapsychiatry.2017.
4602
Hasin, D. S., Shmulewitz, D., Stohl, M., Greenstein, E., Aivadyan, C.,
Morita, K., … Grant, B. F. (2015). Procedural validity of the AUDADIS-5
depression, anxiety and post-traumatic stress disorder modules: Sub-
stance abusers and others in the general population. Drug and Alcohol
Dependence, 152, 246–256. https://doi.org/10.1016/j.drugalcdep.
2015.03.027
Hudson, J. I., Coit, C. E., Lalonde, J. K., & Pope, H. G., Jr. (2012). By how
much will the proposed new DSM-5 criteria increase the prevalence of
binge eating disorder? International Journal of Eating Disorders, 45(1),
139–141. https://doi.org/10.1002/eat.20890
Hudson, J. I., Hiripi, E., Pope, H. G., Jr., & Kessler, R. C. (2007). The preva-
lence and correlates of eating disorders in the national comorbidity
survey replication. Biological Psychiatry, 61(3), 348–358. https://doi.
org/10.1016/j.biopsych.2006.03.040
Javaras, K. N., Pope, H. G., Lalonde, J. K., Roberts, J. L., Nillni, Y. I.,
Laird, N. M., … Hudson, J. I. (2008). Co-occurrence of binge eating dis-
order with psychiatric and medical disorders. Journal of Clinical Psychia-
try, 69(2), 266–273.
Kessler, R. C., Berglund, P. A., Chiu, W. T., Deitz, A. C., Hudson, J. I.,
Shahly, V., … Xavier, M. (2013). The prevalence and correlates of binge
eating disorder in the World Health Organization world mental health
50
surveys. Biological Psychiatry, 73(9), 904–914. https://doi.org/10.
1016/j.biopsych.2012.11.020
Kessler, R. C., & Ustun, T. B. (2004). The world mental health (WMH) sur-
vey initiative version of the World Health Organization (WHO) com-
posite international diagnostic interview (CIDI). International Journal of
Methods in Psychiatric Research, 13(2), 93–121.
Mehler, P. S., & Rylander, M. (2015). Bulimia Nervosa - medical complica-
tions. Journal of Eating Disorders, 3, 12. https://doi.org/10.1186/s40337-
015-0044-4
Mellemkjaer, L., Emborg, C., Gridley, G., Munk-Jorgensen, P., Johansen, C.,
Tjonneland, A., … Olsen, J. H. (2001). Anorexia nervosa and cancer risk.
Cancer Causes & Control, 12(2), 173–177.
Mellemkjaer, L., Papadopoulos, F. C., Pukkala, E., Ekbom, A., Gissler, M.,
Christensen, J., & Olsen, J. H. (2015). Cancer incidence among patients
with anorexia nervosa from Sweden, Denmark and Finland. PLoS One,
10(5), e0128018. https://doi.org/10.1371/journal.pone.0128018
Michels, K. B., & Ekbom, A. (2004). Caloric restriction and incidence of
breast cancer. JAMA, 291(10), 1226–1230. https://doi.org/10.1001/
jama.291.10.1226
Mohler-Kuo, M., Schnyder, U., Dermota, P., Wei, W., & Milos, G. (2016).
The prevalence, correlates, and help-seeking of eating disorders in
Switzerland. Psychological Medicine, 46(13), 2749–2758. https://doi.
org/10.1017/S0033291716001136
Newman, D. L., Moffitt, T. E., Caspi, A., & Silva, P. A. (1998). Comorbid
mental disorders: Implications for treatment and sample selection.
Journal of Abnormal Psychology, 107(2), 305–311.
O'Brien, K. M., Whelan, D. R., Sandler, D. P., & Weinberg, C. R. (2017). Eating
disorders and breast cancer. Cancer Epidemiology, Biomarkers & Prevention,
26(2), 206–211. https://doi.org/10.1158/1055-9965.epi-16-0587
Olguin, P., Fuentes, M., Gabler, G., Guerdjikova, A. I., Keck, P. E., Jr., &
McElroy, S. L. (2016). Medical comorbidity of binge eating disorder.
Eating and Weight Disorders., 22, 13–26. https://doi.org/10.1007/
s40519-016-0313-5
Robinson, L., Aldridge, V., Clark, E. M., Misra, M., & Micali, N. (2016). A sys-
tematic review and meta-analysis of the association between eating dis-
orders and bone density. Osteoporosis International, 27(6), 1953–1966.
https://doi.org/10.1007/s00198-015-3468-4
Sato, Y., & Fukudo, S. (2015). Gastrointestinal symptoms and disorders in
patients with eating disorders. Clinical Journal of Gastroenterology, 8(5),
255–263. https://doi.org/10.1007/s12328-015-0611-x
UDO AND GRILO
Thornton, L. M., Watson, H. J., Jangmo, A., Welch, E., Wiklund, C.,
von Hausswolff-Juhlin, Y., … Bulik, C. M. (2017). Binge-eating disorder
in the Swedish national registers: Somatic comorbidity. International
Journal of Eating Disorders, 50, 58–65. https://doi.org/10.1002/eat.
22624
Trace, S. E., Thornton, L. M., Root, T. L., Mazzeo, S. E., Lichtenstein, P.,
Pedersen, N. L., & Bulik, C. M. (2012). Effects of reducing the fre-
quency and duration criteria for binge eating on lifetime prevalence of
bulimia nervosa and binge eating disorder: Implications for DSM-5.
International Journal of Eating Disorders, 45(4), 531–536. https://doi.
org/10.1002/eat.20955
Udo, T., & Grilo, C. M. (2018). Prevalence and correlates of DSM-5-defined
eating disorders in a nationally representative sample of U.S. adults.
Biological Psychiatry, 84(5), 345–354. https://doi.org/10.1016/j.
biopsych.2018.03.014
Udo, T., & Grilo, C. M. (in press). Assessment challenges in identifying eating
disorders, psychiatric disorders, and impairment. Biological Psychiatry.
Ulfvebrand, S., Birgegård, A., Norring, C., Högdahl, L., & von Hausswolff-
Juhlin, Y. (2015). Psychiatric comorbidity in women and men with eat-
ing disorders results from a large clinical database. Psychiatry Research,
230(2), 294–299.
Welch, E., Jangmo, A., Thornton, L. M., Norring, C., von Hausswolff-Juhlin, Y.,
Herman, B. K., … Bulik, C. M. (2016). Treatment-seeking patients with
binge-eating disorder in the Swedish national registers: Clinical course and
psychiatric comorbidity. BMC Psychiatry, 16, 163. https://doi.org/10.
1186/s12888-016-0840-7
SUPPOR TI NG I NFORMATION
Additional supporting information may be found online in the Sup-
porting Information section at the end of the article.
How to cite this article: Udo T, Grilo CM. Psychiatric and
medical correlates of DSM-5 eating disorders in a nationally
representative sample of adults in the United States. Int J Eat
Disord. 2019;52:42–50. https://doi.org/10.1002/eat.23004