ARTICLES

Prevalence and Treatment of Mental Disorders

Among US Children in the 2001–2004 NHANES
AUTHORS: Kathleen Ries Merikangas, PhD,a Jian-Ping He, WHAT’S KNOWN ON THIS SUBJECT: Although there have been
MSc,a Debra Brody, MPH,b Prudence W. Fisher, PhD,c several regional community surveys of mental disorders in the
Karen Bourdon, MA,d and Doreen S. Koretz, PhDe United States, there are no studies of DSM-IV– defined disorders
aGenetic Epidemiology Research Branch and dEpidemiology in a representative sample of US youths.
Branch, National Institute of Mental Health, Bethesda, Maryland;
bNational Center for Health Statistics, Centers for Disease
WHAT THIS STUDY ADDS: This study adds new information about
Control and Prevention, Hyattsville, Maryland; cNew York State
Psychiatric Institute, Columbia University, New York, New York; the prevalence of DSM-IV– defined mental disorders and service
and eOffice of the Provost, Harvard University, Cambridge, patterns in a general population sample in the United States.
Massachusetts
KEY WORDS
mental disorders, children, epidemiology, services, National
Health and Nutrition Examination Survey
ABBREVIATIONS
NHANES—National Health and Nutrition Examination Survey
abstract
NIMH—National Institute of Mental Health OBJECTIVE: This article presents the 12-month prevalence estimates
DISC—Diagnostic Interview Schedule for Children of specific mental disorders, their social and demographic correlates,
DSM-IV—Diagnostic and Statistical Manual of Mental Disorders,
and service use patterns in children and adolescents from the National
Fourth Edition
ADHD—attention-deficit/hyperactivity disorder Health and Nutrition Examination Survey, a nationally representative
GAD— generalized anxiety disorder probability sample of noninstitutionalized US civilians.
MDD—major depressive disorder
DD— dysthymic disorder METHODS: The sample includes 3042 participants 8 to 15 years of age
OR— odds ratio from cross-sectional surveys conducted from 2001 to 2004. Data on
CI— confidence interval Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition
PIR—poverty index ratio
criteria for mental disorders were derived from administration of se-
www.pediatrics.org/cgi/doi/10.1542/peds.2008-2598
lected modules of the National Institute of Mental Health Diagnostic
doi:10.1542/peds.2008-2598 Interview Schedule for Children, version IV, a structured diagnostic
Accepted for publication Jul 28, 2009 interview administered by lay interviewers to assess psychiatric diag-
Address correspondence to Kathleen Ries Merikangas, PhD, noses of children and adolescents.
National Institute of Mental Health, Genetic Epidemiology
Research Branch, Building 35, Room 1A201, 35 Convent Dr, MSC RESULTS: Twelve-month prevalence rates of Diagnostic and Statistical
3720, Bethesda, MD 20892. E-mail: kathleen.merikangas@nih.gov Manual of Mental Disorders, Fourth Edition– defined disorders in this
PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275). sample were 8.6% for attention-deficit/hyperactivity disorder, 3.7% for
Copyright © 2009 by the American Academy of Pediatrics mood disorders, 2.1% for conduct disorder, 0.7% for panic disorder or
FINANCIAL DISCLOSURE: The authors have indicated they have generalized anxiety disorder, and 0.1% for eating disorders. Boys had
no financial relationships relevant to this article to disclose. 2.1 times greater prevalence of attention-deficit/hyperactivity disorder
than girls, girls had twofold higher rates of mood disorders than
boys, and there were no gender differences in the rates of anxiety
disorders or conduct disorder. Only approximately one half of those
with one of the disorders assessed had sought treatment with a
mental health professional.
CONCLUSION: These data constitute a first step in building a national
database on mental health in children and adolescents. Pediatrics
2010;125:75–81

PEDIATRICS Volume 125, Number 1, January 2010 75

Downloaded from www.aappublications.org/news by guest on November 17, 2018
Recent attention in the media to the METHODS and conduct disorder on parent re-
diagnosis,1,2 prevalence,3 and treat- ports alone, and those of MDD/DD
Sample
ment4 of mental illness and the conse- and eating disorders on either youth
quences of untreated mental illness5 in The sample for the present study in- or parent reports. Diagnostic algo-
children has generated increasing cluded 3024 children 8 to 15 years of rithms corresponding to DSM-IV cri-
public recognition of the serious im- age who were evaluated in person at teria were developed in SAS (SAS
pact of childhood mental disorders. In the mobile examination centers of the Institute, Cary, NC) by the DISC Group
fact, there have been more references 2001–2004 NHANES, a nationally rep- at the Division of Child and Adoles-
to child mental health in the news in resentative probability sample of cent Psychiatry, Columbia University
the past 2 years than in the entire pre- noninstitutionalized US civilians. The (New York, NY).20
ceding decade. However, the absence NHANES used a complex, stratified,
Four levels of impairment were de-
of empirical data on the magnitude, multistage, probability cluster design
rived for each disorder, on the basis
course, and treatment patterns of that oversampled low-income per-
of 6 impairment questions, as follows:
mental disorders in a nationally rep- sons, adolescents 12 to 19 years of
level A, intermediate or severe rating
resentative sample of US youths has age, persons ⬎60 years of age, black
on ⱖ1 question; level B, intermediate
impeded efforts essential for estab- persons, and Mexican American per-
or severe rating on ⱖ2 questions;
lishing mental health policy for US sons. The response rates for the youth
level C, severe rating on ⱖ1 question;
youths.6–11 sample ranged from 79.2% to 92.3%,
level D, either level B or C. These ques-
On the basis of the recommendations depending on the disorder and the
tions assessed impairment in 6 do-
of the landmark report of the Surgeon source of information. There were no
mains, including interference with the
General on mental health12 and those significant differences in demographic
respondent’s own life, family life, so-
of a subgroup of the National Institute characteristics between participants
cial life, peers, teachers, and school
of Mental Health (NIMH) National Advi- and nonparticipants. Additional details
of the NHANES methods are available performance.
sory Mental Health Council,13 the NIMH
elsewhere. Information on 12-month mental
established several research initia-
health service use was collected in
tives to address the lack of national
Measures each of the diagnostic modules. In
statistics on mental health in chil-
Information on mental disorders was each of the DISC diagnostic modules by
dren. This led to a collaboration be-
derived from the NIMH Diagnostic In- the following: “In the past year, have
tween the NIMH and the National Cen-
terview Schedule for Children (DISC), you been to see someone at a hospital
ter for Health Statistics to collect
version IV, a structured diagnostic in- or a clinic or at their office [for specific
population-based data on selected
terview administered by lay interview- symptoms of disorders]?”
mental disorders14–16 in the National
Health and Nutrition Examination Sur- ers to assess Diagnostic and Statisti-
Statistical Methods
vey (NHANES). This article reports the cal Manual of Mental Disorders,
first aggregate prevalence rates of Fourth Edition (DSM-IV) diagnostic cri- The sample weights used followed Na-
mental disorders in children and ado- teria for mental disorders in children tional Center for Health Statistics
lescents 8 to 15 years of age, from the and adolescents.14,15,19 Modules for guidelines and were calculated ac-
2001–2004 NHANES findings. Previous generalized anxiety disorder (GAD), cording to the base probabilities of se-
reports presented prevalence data panic disorder, eating disorders (an- lection, adjusted for nonresponse, and
on attention-deficit/hyperactivity dis- orexia nervosa and bulimia nervosa), poststratified to match population
order (ADHD)17 and conduct disorder.18 and major depressive disorder (MDD)/ control totals. Weighted prevalences
The goals of this report are to present dysthymic disorder (DD) were admin- and SEs were estimated according to
(1) estimates of the prevalence of spe- istered to youths, and those for MDD/ gender, age, race/ethnicity, and pov-
cific mental disorders and their social DD, eating disorders, ADHD, and erty index ratio (PIR) (family income/
and demographic correlates in chil- conduct disorder were administered family poverty threshold level ratio, on
dren and adolescents in the United to the primary caretakers, via tele- the basis of family size) by using a
States and (2) patterns and correlates phone, within 4 to 28 days after the 4-year weight variable for pooled data
of service use among children with youth interviews. The diagnoses of GAD analysis, using one half of the 2-year
these mental health problems in the and panic disorder were based on medical examination center weight in
United States. youth reports alone, those of ADHD the 2001–2002 NHANES and one half of

76 MERIKANGAS et al
Downloaded from www.aappublications.org/news by guest on November 17, 2018
 ARTICLES

TABLE 1 Prevalence of 12-Month, DSM-IV–Defined Disorders According to Gender and Age in US Children 8 to 15 Years of Age
DSM-IV–Defined Prevalence, Estimate ⫾ SE, %
Disorder
Disorder Without Impairment Disorder With Severe Impairment (Level D)a
Gender Age Total Gender Age Total
(N ⫽ 3042) (N ⫽ 3042)
Male Female 8–11 y 12–15 y Male Female 8–11 y 12–15 y
(N ⫽ 1492) (N ⫽ 1550) (N ⫽ 1148) (N ⫽ 1894) (N ⫽ 1492) (N ⫽ 1550) (N ⫽ 1148) (N ⫽ 1894)
ADHD, all 11.6 ⫾ 1.0 5.4 ⫾ 0.6 9.9 ⫾ 1.0 7.4 ⫾ 1.0 8.6 ⫾ 0.7 10.8 ⫾ 0.9 4.7 ⫾ 0.7 9.1 ⫾ 1.0 6.7 ⫾ 0.8 7.8 ⫾ 0.7
␹12 ⫽ 45.18, P ⬍.001 ␹12 ⫽ 3.23, P ⫽ .082 ␹12 ⫽ 46.86, P ⬍.001 ␹12 ⫽ 4.29, P ⫽ .047
Attention deficit 5.4 ⫾ 0.9 3.1 ⫾ 0.5 4.6 ⫾ 0.8 4.0 ⫾ 0.8 4.3 ⫾ 0.6
␹12 ⫽ 5.59, P ⫽ .025 ␹12 ⫽ 0.43, P ⫽ .517
Hyperactivity 2.8 ⫾ 0.7 1.2 ⫾ 0.3 2.8 ⫾ 0.7 1.3 ⫾ 0.3 2.0 ⫾ 0.4
␹12 ⫽ 4.56, P ⫽ .041 ␹12 ⫽ 3.85, P ⫽ .059
Combined 3.4 ⫾ 0.4 1.1 ⫾ 0.2 2.4 ⫾ 0.5 2.1 ⫾ 0.3 2.2 ⫾ 0.2
␹12 ⫽ 20.99, P ⫽ ⬍.001 ␹12 ⫽ 0.27, P ⫽ .610
Conduct disorder 2.3 ⫾ 0.3 1.9 ⫾ 0.5 1.5 ⫾ 0.3 2.7 ⫾ 0.5 2.1 ⫾ 0.3 2.0 ⫾ 0.3 1.4 ⫾ 0.4 1.2 ⫾ 0.2 2.2 ⫾ 0.5 1.7 ⫾ 0.3
␹12 ⫽ 0.71, P ⫽ .406 ␹12 ⫽ 5.76, P ⫽ .023 ␹12 ⫽ 1.26, P ⫽ .271 ␹12 ⫽ 3.90, P ⫽ .058
Anxiety disorder 0.4 ⫾ 0.2 0.9 ⫾ 0.3 0.4 ⫾ 0.2 0.8 ⫾ 0.3 0.7 ⫾ 0.2 0.4 ⫾ 0.2 0.4 ⫾ 0.2 0.3 ⫾ 0.2 0.5 ⫾ 0.2 0.4 ⫾ 0.1
␹12 ⫽ 1.74, P ⫽ .197 ␹12 ⫽ 1.04, P ⫽ .317 ␹12 ⫽ 0.004, P ⫽ .948 ␹12 ⫽ 0.20, P ⫽ .656
Generalized anxiety 0.3 ⫾ 0.2 0.4 ⫾ 0.2 0.1 ⫾ 0.1 0.7 ⫾ 0.3 0.3 ⫾ 0.1 0.3 ⫾ 0.2 0.1 ⫾ 0.1 0.0 ⫾ 0.0 0.4 ⫾ 0.2 0.2 ⫾ 0.1
␹12 ⫽ 0.33, P ⫽ .569 ␹12 ⫽ 3.54, P ⫽ .070 ␹12 ⫽ 0.33, P ⫽ .571 ␹12 ⫽ 3.29, P ⫽ .081
Panic disorder 0.2 ⫾ 0.1 0.6 ⫾ 0.2 0.4 ⫾ 0.2 0.4 ⫾ 0.2 0.4 ⫾ 0.1 0.2 ⫾ 0.1 0.4 ⫾ 0.2 0.3 ⫾ 0.2 0.2 ⫾ 0.1 0.3 ⫾ 0.1
␹12 ⫽ 2.51, P ⫽ .124 ␹12 ⫽ 0.003, P ⫽ .955 ␹12 ⫽ 0.71, P ⫽ .406 ␹12 ⫽ 0.172, P ⫽ .681
Eating disorder 0.1 ⫾ 0.0 0.2 ⫾ 0.1 0.1 ⫾ 0.1 0.2 ⫾ 0.1 0.1 ⫾ 0.1 0.0 ⫾ 0.0 0.1 ⫾ 0.0 0.0 ⫾ 0.0 0.1 ⫾ 0.0 0.03 ⫾ 0.01
␹12 ⫽ 2.01, P ⫽ .167 ␹12 ⫽ 1.07, P ⫽ .309 ␹12 ⫽ 2.86, P ⫽ .101 ␹12 ⫽ 2.85, P ⫽ .102
Mood disorder 2.5 ⫾ 0.7 4.9 ⫾ 0.9 2.5 ⫾ 0.7 4.8 ⫾ 0.9 3.7 ⫾ 0.6 1.7 ⫾ 0.5 4.1 ⫾ 0.8 1.8 ⫾ 0.5 3.9 ⫾ 0.8 2.9 ⫾ 0.5
␹12 ⫽ 6.64, P ⫽ .015 ␹12 ⫽ 7.08, P ⫽ .012 ␹12 ⫽ 7.37, P ⫽ .011 ␹12 ⫽ 7.26, P ⫽ .011
Major depression 1.8 ⫾ 0.6 3.7 ⫾ 0.8 1.6 ⫾ 0.5 3.8 ⫾ 0.8 2.7 ⫾ 0.6 1.6 ⫾ 0.5 3.2 ⫾ 0.7 1.4 ⫾ 0.4 3.2 ⫾ 0.7 2.4 ⫾ 0.5
␹12 ⫽ 4.65, P ⫽ .039 ␹12 ⫽ 10.00, P ⫽ .004 ␹12 ⫽ 3.90, P ⫽ .058 ␹12 ⫽ 7.65, P ⫽ .010
Dysthymia 0.7 ⫾ 0.3 1.2 ⫾ 0.4 0.8 ⫾ 0.4 1.1 ⫾ 0.3 1.0 ⫾ 0.3 0.1 ⫾ 0.1 0.9 ⫾ 0.4 0.4 ⫾ 0.2 0.7 ⫾ 0.3 0.5 ⫾ 0.2
␹12 ⫽ 1.53, P ⫽ .225 ␹12 ⫽ 0.28, P ⫽ .601 ␹12 ⫽ 4.73, P ⫽ .038 ␹12 ⫽ 0.91, P ⫽ .348
Any of above 14.5 ⫾ 1.0 11.6 ⫾ 1.1 12.8 ⫾ 1.3 13.4 ⫾ 1.2 13.1 ⫾ 0.9 13.0 ⫾ 0.9 9.4 ⫾ 1.2 11.0 ⫾ 1.1 11.5 ⫾ 1.3 11.3 ⫾ 0.9
␹12 ⫽ 6.47, P ⫽ .016 ␹12 ⫽ 0.12, P ⫽ .731 ␹12 ⫽ 9.73, P ⫽ .004 ␹12 ⫽ 0.10, P ⫽ .758
a Impairment level D indicates ⱖ2 intermediate or 1 severe rating on the 6 impairment questions regarding personal distress and social (at home or with peers) or academic difficulties.

the 2-year medical examination center portional to the probability of selection yielded a weighted prevalence of 1.8%
weight in the 2003–2004 NHANES. The into the sample and are interpreted as (SE: 0.3%). Total rates of specific disor-
weight represents the number of in- the number of individuals in the target ders were 8.6% for ADHD (4.3% for at-
dividuals in the target population population each sample participant is tention deficit, 2.0% for hyperactivity,
each sample participant is esti- estimated to represent. and 2.2% for attention deficit and hy-
mated to represent. peractivity combined type), 2.1% for
Logistic regression models were used RESULTS conduct disorder, 0.7% for anxiety dis-
to assess the association between Table 1 presents the 12-month preva- orders (0.3% for GAD and 0.4% for
mental disorders with service use and lence of specific disorders, with or panic disorder), 0.1% for eating disor-
comorbidity across disorders. Data without impairment, as assessed in ders (0.1% for anorexia and 0.1% for
were analyzed by using SUDAAN 9 (RTI the NHANES, according to child gender bulimia), and 3.7% for mood disorders
International, Research Triangle Park, and age group. One of 8 children 8 to 15 (2.7% for MDD and 1.0% for DD).
NC) procedures, which apply Taylor se- years of age met 12-month criteria for Boys had significantly higher rates of
ries linearization methods to accom- ⱖ1 of the 6 DSM-IV– defined disorders, any 12-month disorder than did girls,
modate sampling weights to account which yielded a weighted prevalence primarily because of the high rates
for stratification and clustering of the of 13.1% (SE: 0.9%). Approximately 14% of male-predominant ADHD. Girls had
multistage NHANES sampling design in of children with one 12-month disor- higher rates of mood disorders (␹21 ⫽
the calculation of SEs and test statistics. der met the criteria for ⱖ2 of the 6.64; P ⫽ .015), particularly MDD
The sampling weights are inversely pro- 6 DSM-IV– defined disorders, which (␹21 ⫽ 4.65; P ⫽ .039). Rates of conduct

PEDIATRICS Volume 125, Number 1, January 2010 77

Downloaded from www.aappublications.org/news by guest on November 17, 2018
disorder, mood disorders, and MDD TABLE 2 Past-Year Mental Health Service Use Among Children 8 to 15 Years of Age With 12-Month,
DSM-IV–Defined Disorder
were higher in older children (12–15
12-mo Disorder All Cases Cases With Severe Impairmenta
years of age) than in younger ones,
whereas ADHD (hyperactivity subtype) n Proportion, n Proportion,
Estimate ⫾ SE, % Estimate ⫾ SE, %
was marginally significantly greater in
Any disorder 366 50.6 ⫾ 3.4 305 52.8 ⫾ 3.7
younger children (␹21 ⫽ 3.85; P ⫽ .059). ADHD 218 47.7 ⫾ 4.4 195 48.5 ⫾ 4.5
Compared with non-Hispanic white Conduct disorder 68 46.4 ⫾ 8.0 63 44.2 ⫾ 8.4
youths, Mexican American youths had Anxiety disorders (GAD 21 32.2 ⫾ 14.3 12 33.9 ⫾ 19.0
or panic disorder)
significantly lower rates of 12-month Mood disorders 128 43.8 ⫾ 6.0 95 50.7 ⫾ 6.0
ADHD (hyperactivity subtype) (␹23 ⫽ Service use indicated the respondent saw someone at a hospital, clinic, or office because of 关specific disorder兴 in the
28.20; P ⬍ .001) and higher rates of DD previous 12 months.
a Impairment level D.
(␹23 ⫽ 11.08; P ⫽ .022). Youths with low
PIR were more likely to report any 12-
TABLE 3 Demographic Correlates of Mental Health Service Use Among Subjects With 12-Month
month disorder, ADHD, and its inatten- Mental Disorders, With and Without Severe Impairment
tion subtype, whereas those with high Demographic Correlates Among Subjects With Among Subjects With Any
PIR were more likely to report an anxiety Mental Disorder Impairment Level
(N ⫽ 366) D Mental Disorder (N ⫽ 305)
disorder (␹3 ⫽ 8.75; P ⫽ .050). PIR was
2

not associated with other 12-month, DSM-IV– n Adjusted OR (95% CI) n Adjusted OR (95% CI)

defined disorders (race/ethnicity and Gender

Male 186 Reference 163 Reference
PIR data not shown in Table 1). Female 180 0.5 (0.3–1.0) 142 0.6 (0.3–1.2)
Table 1 shows the prevalence of disor- ␹2 4.8 2.4
P .029 .120
ders with impairment level D (defined Age
by ⱖ2 intermediate or 1 severe rating 8–11 y 153 Reference 130 Reference
on the 6 impairment questions regard- 12–15 y 213 1.9 (1.0–2.8) 175 1.6 (0.7–2.6)
␹2 4.1 3.4
ing personal distress and social [at P .042 .067
home or with peers] and academic diffi- Race/ethnicity
culties), referred to as severe impair- White, non-Hispanic 128 Reference 111 Reference
Black, non-Hispanic 122 0.8 (0.5–1.3) 102 0.9 (0.5–1.5)
ment. For most disorders, there was
Mexican American 90 0.5 (0.2–1.0) 69 0.5 (0.2–1.1)
minimal reduction in rates with applica- Other 26 1.1 (0.4–2.7) 23 0.8 (0.3–1.9)
tion of impairment criterion D. Rates of ␹2 4.2 3.5
ADHD decreased from 8.6% to 7.8% (9%) P .240 .319
PIRa
and rates of conduct disorder from 2.1% ⬎2 131 Reference 113 Reference
to 1.7%. There were greater reductions 1–2 97 0.8 (0.4–1.5) 79 0.9 (0.4–1.8)
in mood disorders (from 3.7% to 2.9%), ⬍1 (poor) 130 0.8 (0.4–1.8) 105 0.9 (0.4–2.2)
␹2 0.7 0.2
panic disorders or GAD (from 0.7% to P .690 .917
0.4%), and eating disorders (from 1.0% Health insurance
to 0.5%). However, the only significant re- Not covered 48 Reference 40 Reference
Covered 318 2.7 (0.7–10.2) 265 2.8 (0.7–10.5)
duction occurred for mood disorders ␹2 2.4 2.5
(odds ratio [OR]: 4.4 [95% confidence in- P .126 .113
terval [CI]: 1.4 –14.1]; P ⫽ .0145). Service use indicated the respondent saw someone at a hospital, clinic, or office because of 关specific disorder兴 in the
previous 12 months. Adjusted ORs were from a model containing all variables shown in the table.
We also examined patterns of comor- a Data for subjects with unknown PIRs are not presented in the table.

bidity across the 12-month disorders

(data not shown). After adjustment
for age, gender, ethnicity, and pov- and anxiety disorder (OR: 29.5 [95% ADHD and those with conduct disorder
erty level, significant associations CI: 9.4 –92.3]). had the greatest treatment rates (47.7%
were found between 12-month ADHD Table 2 presents the rates of mental and 46.4%, respectively), whereas those
and conduct disorder (OR: 7.6 [95% health service use for those with each with GAD or panic disorder had the low-
CI: 4.0 –14.7]), between ADHD and of the mental disorders assessed in est treatment rates (32.2%). Treatment
mood disorder (OR: 3.4 [95% CI: 1.8 – this study, for all cases and for those rates were slightly greater for those
6.4]), and between mood disorder with severe impairment. Youths with with severe impairment.

78 MERIKANGAS et al
Downloaded from www.aappublications.org/news by guest on November 17, 2018
 ARTICLES

Table 3 presents the demographic pre- lence rates revealed that the rates of disorders (43.8%) reported service con-
dictors associated with mental health mood disorders were higher among tacts for those conditions.43 There was a
service for those with any 12-month, older children, compared with younger moderate range of treatment rates for
DSM-IV– defined disorder and for children, whereas the rates of conduct specific subtypes of disorders, ranging
those with severe impairment. Few of disorder were higher among early ad- from 47.7% mental health service use
the demographic correlates were as- olescents (12–15 years of age), com- among those with ADHD to 32.2% among
sociated with service use among those pared with younger children. In gen- those with GAD or panic disorder. Treat-
with either a 12-month disorder or a eral, there were few ethnic differences ment rates increased substantially for
severe 12-month disorder. Boys were in disorders studied in this survey; those with panic disorder/GAD, eating
more likely to seek treatment than however, Mexican American youths disorders, and/or mood disorders when
were girls, and older youths (12–15 had significantly higher rates of mood the sample was restricted to youths with
years of age) reported more mental disorders than did either white or severe impairment. Despite the rela-
health service use than did younger black youths. Finally, poor children had tively large proportions of youths with
youths (8 –11 years of age). Neither higher rates of any disorder and ADHD ADHD, conduct disorder, and mood dis-
race/ethnicity nor poverty level was and lower rates of anxiety disorders orders who sought mental health ser-
associated with professional service than did their wealthier counterparts. vices specifically for those disorders,
use specifically for ⱖ1 of the disor- Comorbidity was less common among only a minority (32.2%) of youths with
ders assessed in this survey. youths in this survey than in compara- anxiety disorders, even those with se-
ble studies of adults, where the major- vere impairment, did so. This confirms
DISCUSSION ity of studies found that few individuals the consistent finding from previous
The findings of this article provide the in the population had only 1 disorder. studies that up to 80% of youths with
first estimates of the prevalence of In the present study, only 1.8% (SE: anxiety disorders do not use health ser-
DSM-IV– defined mental disorders in a 0.3%) of youths 8 to 15 years of age had vices.44 Increased education about the
broad age range of children in the gen- ⬎1 disorder. Similar to previous com- availability of effective treatments
eral population of the United States. munity studies,21 the disorders with for anxiety disorders are also may be
The most-common 12-month disorder the highest rates of cooccurrence
warranted.
was ADHD (8.7%),17 followed by mood were conduct disorder and ADHD. Pre-
disorders (3.7%), conduct disorder vious prevalence studies of children This study has several strengths. It
(2.1%),18 and panic disorder/GAD and adolescents yielded far higher provides the first estimates of the
(0.7%). Eating disorders were very rates of comorbidity, particularly prevalence of specific DSM-IV– defined
rare (0.1%). With the exception of those that assessed lifetime disor- mental disorders in the US population
ADHD, these rates are substantially ders.21,30,36,37 The lower rates of comor- of children and adolescents. The reli-
lower than those reported in other bidity are likely attributable to the lim- ability of the prevalence rates is en-
prevalence studies of mental disor- ited number of disorders assessed in hanced by the use of repeated surveys
ders in children.21–23 However, they are the current study, compared with the over 4 years, which increases the pre-
quite comparable to findings from full range of disorders assessed in cision of the estimates. The use of a
other US studies that used similar di- other studies. structured interview and standardized
agnostic methods and criteria24–28 and With respect to service use, we found diagnostic criteria facilitate compari-
are strikingly similar to those re- that about one half of those who had sons of these findings with those of
ported for a community survey con- 1 of the mental disorders examined in other local studies in the United States
ducted at the same time in Houston, this survey during the past year had and in international settings. Informa-
Texas.26 sought treatment in the mental health tion that supplements the diagnostic
The sociodemographic correlates of sector. Approximately one half of those data in this study includes ratings of
mental disorders in this survey also with ADHD had sought mental health the clinical significance of the disor-
confirmed those of previous commu- treatment. This finding confirms the ders through inclusion of systematic
nity studies of youths in the United continuing increase in service use for information on different levels of func-
States.25–35 The prevalence of mood dis- childhood mental disorders, particu- tional impairment; data on service-
orders was greater in girls, whereas larly ADHD, in recent decades.38–42 seeking for specific disorders, as well
there was a male preponderance of Nearly equal proportions of those with as for mental health services in gen-
ADHD. Inspection of age-specific preva- conduct disorder (46.4%) and mood eral; and a comprehensive series of

PEDIATRICS Volume 125, Number 1, January 2010 79

Downloaded from www.aappublications.org/news by guest on November 17, 2018
laboratory measurements and assess- The rates are based solely on reports generated 12-month estimates of men-
ments of numerous physical disorders of the diagnostic criteria by either the tal disorders in children on the basis
and health behaviors. Future reports parent or the child, rather than clini- of DISC and DSM-IV criteria demon-
on this survey will address these po- cians or teachers, who may be more strated the reliability of the methods
tential correlates of mental health. knowledgeable about impairment and used here. The large variation between
There are several methodologic fea- disability either in the individual our estimates and those of studies us-
tures of this study that should be con- youths or among comparable children ing other diagnostic interviews, preva-
sidered as possible sources of differ- as a frame of reference. Fourth, there lence periods, and constellations of
ences from the findings of previous were no systematic clinical evalua- disorders also highlights the need for
studies. First, it is likely that these find- tions to validate the diagnoses.16,47,48 research designed to validate the diag-
ings are underestimates of the true The generally low base rates of most nostic classification of mental disor-
population base rates, because the disorders in this study should be inter- ders in children.7–9,11,21,49–51
most-common disorders in children, preted in the context of the methods
This study has begun to address the
including separation anxiety and pho- used in this study. Reviews of the ag-
gap in knowledge regarding national
bic states, were not included in the gregate data from population-based
patterns of mental health problems in
survey.21,45 Second, the rates of most studies have shown that different
disorders in this study were based on methods of ascertaining data on diag- children. When combined with the re-
reports of only 1 informant, rather nostic criteria (eg, symptom rating sults of several regional epidemiological
than information from both parent and scales, structured and semistructured studies of mental disorders in children11
child informants, which has been diagnostic interviews, and objective and other national, population-based
shown to provide the most-valid infor- measures) can yield a wide range of studies now underway (including the Na-
mation on disorders in youths.46 Third, prevalence estimates.11,21 It will be im- tional Comorbidity Survey Adolescent
the parent report was obtained by tele- portant to consider a range of preva- Supplement,52,53), these data will provide
phone after completion of the direct lence estimates, depending on the a valuable empirical basis for the devel-
assessment of the child. This would be specific goals of the application of opment of health policies designed to
expected to yield different rates, com- morbidity rates. Nevertheless, the sim- maximize prevention efforts and to min-
pared with studies that assess both in- ilarity between our findings and those imize the consequences of these condi-
formants in the same direct interview. of previous community surveys that tions in US youths.

REFERENCES
1. Gaviria M. The Medicated Child [Frontline pro- life-long history of mental illness. USA To- update review: the epidemiology of child and ad-
duction]. Boston, MA: WGBH Boston; 2008 day. Available at: www.usatoday.com/ olescent psychiatric disorders, part I: methods
2. Allen S. Backlash on bipolar diagnoses in news/nation/2007-08-29-vt-report_n.htm. and public health burden. J Am Acad Child Ado-
children: MGH psychiatrist’s work stirs de- Accessed November 29, 2007 lesc Psychiatry. 2005;44(10):972–986
bate. Boston Globe. Available at: www. 6. Achenbach TM. Advancing assessment of 12. US Surgeon General. US Surgeon General’s
boston.com/yourlife/health/diseases/ children and adolescents: commentary on Report on Mental Health. Washington, DC:
articles/2007/06/17/backlash_on_bipolar_ evidence-based assessment of child and US Office of Surgeon General; 2001
diagnoses_in_children/. Accessed June 17, adolescent disorders. J Clin Child Adolesc 13. National Institute of Mental Health. An NIMH
2007 Psychol. 2005;34(3):541–547 Oversight Board Report to the National Ad-
3. Dooren JC. Autism rate is still rising despite 7. Achenbach TM, Edelbrock CS. Psychopathol- visory Mental Health Council. Bethesda, MD:
vaccine change. Wall Street Journal. Avail- ogy of childhood. Annu Rev Psychol. 1984;35: National Institute of Mental Health; 1998
able at: http://online.wsj.com/article_ 227–256 14. Shaffer D, Fisher P, Dulcan MK, et al. The
email/SB119973993129672867-1MyQj 8. Hudziak JJ, Achenbach TM, Althoff RR, Pine NIMH Diagnostic Interview Schedule for
AxMDE4OTA5NzcwMzc5Wj.html. Accessed DS. A dimensional approach to developmen- Children Version 2.3 (DISC-2.3): description,
January 7, 2008 tal psychopathology. Int J Methods Psychi- acceptability, prevalence rates, and perfor-
4. Steenhuysen J. Effect of antidepressant atr Res. 2007;16(suppl 1):S16 –S23 mance in the MECA Study: Methods for the
warnings moderate US study. Reuters. 9. Mash EJ, Hunsley J. Evidence-based assess- Epidemiology of Child and Adolescent Men-
Available at: www.reuters.com/article/ ment of child and adolescent disorders: is- tal Disorders Study. J Am Acad Child Ado-
healthNews/idUSN0734236420080107? sues and challenges. J Clin Child Adolesc lesc Psychiatry. 1996;35(7):865– 877
feedType⫽RSS&FeedName⫽healthNews& Psychol. 2005;34(3):362–379 15. Shaffer D, Fisher P, Lucas CP, Dulcan MK,
pageNumber⫽2&virtualBrandChannel⫽0. 10. Szatmari P, March JS. Clinical practice Schwab-Stone ME. NIMH Diagnostic Inter-
Accessed January 7, 2008 guidelines. J Am Acad Child Adolesc Psychi- view Schedule for Children Version IV (NIMH
5. Report details red flags preceding Tech atry. 2007;46(8):939 –940 DISC-IV): description, differences from pre-
tragedy: report examines student shooter’s 11. Costello EJ, Egger H, Angold A. 10-year research vious versions, and reliability of some com-

80 MERIKANGAS et al
Downloaded from www.aappublications.org/news by guest on November 17, 2018
 ARTICLES

mon diagnoses. J Am Acad Child Adolesc ders in Puerto Rico: prevalence, correlates, 42. Hoagwood K, Kelleher KJ, Feil M, Comer DM.
Psychiatry. 2000;39(1):28 –38 service use, and the effects of impairment. Treatment services for children with ADHD:
16. Schwab-Stone ME, Shaffer D, Dulcan MK, Arch Gen Psychiatry. 2004;61(1):85–93 a national perspective. J Am Acad Child Ado-
et al. Criterion validity of the NIMH Diagnos- 29. Earls F. Epidemiology and child psychiatry: lesc Psychiatry. 2000;39(2):198 –206
tic Interview Schedule for Children Version historical and conceptual development. 43. Wu P, Hoven CW, Bird HR, et al. Depressive
2.3 (DISC-2.3). J Am Acad Child Adolesc Psy- Compr Psychiatry. 1979;20(3):256 –269 and disruptive disorders and mental health
chiatry. 1996;35(7):878 – 888 30. Costello J, Angold A, Burns BJ, et al. The service utilization in children and adoles-
17. Froehlich TE, Lanphear BP, Epstein JN, Bar- Great Smoky Mountains Study of Youths: cents. J Am Acad Child Adolesc Psychiatry.
baresi WJ, Katusic SK, Kahn RS. Prevalence, goals, design, methods, and the prevalence 1999;38(9):1081–1090
recognition, and treatment of attention- of DSM-III-R disorders. Arch Gen Psychiatry. 44. Essau CA. Frequency and patterns of mental
deficit/hyperactivity disorder in a national 1996;53(12):1129 –1136 health services utilization among adolescents
sample of US children. Arch Pediatr Adolesc 31. Garrison CZ, Waller JL, Cuffe SP, McKeown with anxiety and depressive disorders. De-
Med. 2007;161(9):857– 864 RE, Addy CL, Jackson KL. Incidence of major press Anxiety. 2005;22(3):130 –137
18. Braun JM, Froehlich TE, Daniels JL, et al. Associ- depressive disorder and dysthymia in 45. Costello EJ, Foley DL, Angold A. 10-year research
ation of environmental toxicants and conduct young adolescents. J Am Acad Child Adolesc update review: the epidemiology of child and ad-
disorder in US children: NHANES 2001–2004. En- Psychiatry. 1997;36(4):458 – 465 olescent psychiatric disorders, part II: develop-
viron Health Perspect. 2008;116(7):956–962 32. Kashani J, Orvaschel H, Rosenberg T, Reid mental epidemiology. J Am Acad Child Adolesc
19. American Psychiatric Association. Diagnos- J. Psychopathology in a community sample Psychiatry. 2006;45(1):8–25
tic and Statistical Manual of Mental Disor- of children and adolescents: a developmen- 46. Jensen PS, Rubio-Stipec M, Canino G, et al.
ders. 4th ed. Arlington, VA: American Psychi- tal perspective. J Am Acad Child Adolescent Parent and child contributions to diagnosis
atric Association; 1994 Psychiatry. 1989;28(5):701–706 of mental disorder: are both informants al-
20. Lahey BB, Flagg EW, Bird HR, et al. The NIMH 33. Lewinsohn PM, Hops H, Roberts RE, Seeley ways necessary? J Am Acad Child Adolesc
Methods for the Epidemiology of Child and Ad- JR, Andrews JA. Adolescent psychopathol- Psychiatry. 1999;38(12):1569 –1579
olescent Mental Disorders (MECA) Study: ogy, part I: prevalence and incidence of de- 47. Hodges B, Regehr G, Hanson M, McNaughton
background and methodology. J Am Acad pression and other DSM-III-R disorders in N. Validation of an objective structured clin-
Child Adolesc Psychiatry. 1996;35(7):855– 864 high school students. J Abnorm Psychol. ical examination in psychiatry. Acad Med.
21. Costello E, Mustillo S, Keeler G, Angold A. Prev- 1993;102(1):133–144 1998;73(8):910 –912
alence of psychiatric disorders in childhood 34. Reinherz HZ, Giaconia RM, Lefkowitz ES, Pa- 48. Cohen P, O’Connor P, Lewis S, Velez CN, Mala-
and adolescence. In: Levin B, Petrila J, Hen- kiz B, Frost A. Prevalence of psychiatric dis- chowski B. Comparison of DISC and
nessy K, eds. Mental Health Services: A Public orders in a community population of older K-SADS-P interviews of an epidemiological
Health Perspective. New York, NY: Oxford Uni- adolescents. J Am Acad Child Adolesc Psy- sample of children. J Am Acad Child Adolesc
versity Press; 2004:111–128 chiatry. 1993;32(2):369 –377 Psychiatry. 1987;26(5):662– 667
22. Meltzer H, Gatward R, Goodman R, Ford T. 35. Whitaker A, Johnson J, Shaffer D, et al. Un- 49. Angst J, Gamma A, Benazzi F, Ajdacic V, Eich
Mental health of children and adolescents common troubles in young people: preva- D, Rossler W. Toward a re-definition of sub-
in Great Britain. Int Rev Psychiatry. 2003; lence estimates of selected psychiatric dis- threshold bipolarity: epidemiology and pro-
15(1–2):185–187 orders in a nonreferred population. Arch posed criteria for bipolar-II, minor bipolar
23. Maughan B, Collishaw S, Meltzer H, Good- Gen Psychiatry. 1990;47(5):487– 496 disorders and hypomania. J Affect Disord.
man R. Recent trends in UK child and ado- 36. Kessler RC, Avenevoli S, Ries Merikangas K. 2003;73(1–2):133–146
lescent mental health. Soc Psychiatry Psy- Mood disorders in children and adolescents: 50. Angst J, Merikangas KR, Preisig M. Sub-
chiatr Epidemiol. 2008;43(4):305–310 an epidemiologic perspective. Biol Psychiatry. threshold syndromes of depression and
24. Bird HR, Canino G, Rubino-Stipec M, et al. Esti- 2001;49(12):1002–1014 anxiety in the community. J Clin Psychiatry.
mates of the prevalence of childhood malad- 37. Angold A, Costello EJ, Erkanli A. Comorbidity. 1997;58(suppl 8):6 –10
justment in a community survey in Puerto J Child Psychol Psychiatry. 1999;40(1):57– 87 51. Buka SL, Monuteaux M, Earls F. The epidemi-
Rico: the use of combined measures. Arch Gen 38. Wu P, Hoven CW, Cohen P, et al. Factors as- ology of child and adolescent mental disor-
Psychiatry. 1988;45(12):1120 –1126 sociated with use of mental health services ders. In: Tsuang MT, Tohen M, eds. Textbook
25. Cohen P, Cohen J, Kasen S, et al. An epide- for depression by children and adolescents. in Psychiatric Epidemiology. 2nd ed. New
miological study of disorders in late child- Psychiatr Serv. 2001;52(2):189 –195 York, NY: Wiley; 2002:629 – 655
hood and adolescence, part I: age- and 39. Angold A, Erkanli A, Farmer EM, et al. Psychi- 52. Merikangas K, Avenevoli S, Costello J, Koretz
gender-specific prevalence. J Child Psychol atric disorder, impairment, and service use D, Kessler RC. National comorbidity survey
Psychiatry. 1993;34(6):851– 867 in rural African American and white youth. replication adolescent supplement (NCS-A):
26. Roberts RE, Roberts CR, Xing Y. Rates of Arch Gen Psychiatry. 2002;59(10):893–901 I. Background and measures. J Am Acad
DSM-IV psychiatric disorders among ado- 40. Elster A, Jarosik J, VanGeest J, Fleming M. Racial Child Adolesc Psychiatry. 2009;48(4):
lescents in a large metropolitan area. J Psy- and ethnic disparities in health care for 367–369
chiatr Res. 2007;41(11):959 –967 adolescents: a systematic review of the litera- 53. Kessler RC, Avenevoli S, Costello EJ, Green
27. Velez C, Johnson J, Cohen P. A longitudinal ture. Arch Pediatr Adolesc Med. 2003;157(9): JG, Gruber MJ, Heeringa S, Merikangas KR,
analysis of selected risk factors for child- 867–874 Pennell BE, Sampson NA, Zaslavsky AM. De-
hood psychopathology. J Am Acad Child Ad- 41. Owens PL, Hoagwood K, Horwitz SM, et al. sign and field procedures in the US National
olescent Psychiatry. 1989;28(6):861– 864 Barriers to children’s mental health ser- Comorbidity Survey Replication Adolescent
28. Canino G, Shrout PE, Rubio-Stipec M, et al. The vices. J Am Acad Child Adolesc Psychiatry. Supplement (NCS-A). Int J Methods Psychi-
DSM-IV rates of child and adolescent disor- 2002;41(6):731–738 atr Epidemiol. 2009;18(2):69 – 83

PEDIATRICS Volume 125, Number 1, January 2010 81

Downloaded from www.aappublications.org/news by guest on November 17, 2018
Prevalence and Treatment of Mental Disorders Among US Children in the 2001−
2004 NHANES
Kathleen Ries Merikangas, Jian-Ping He, Debra Brody, Prudence W. Fisher, Karen
Bourdon and Doreen S. Koretz
Pediatrics 2010;125;75
DOI: 10.1542/peds.2008-2598 originally published online December 14, 2009;

Updated Information & including high resolution figures, can be found at:
Services http://pediatrics.aappublications.org/content/125/1/75
References This article cites 43 articles, 0 of which you can access for free at:
http://pediatrics.aappublications.org/content/125/1/75#BIBL
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Developmental/Behavioral Pediatrics
http://www.aappublications.org/cgi/collection/development:behavior
al_issues_sub
Permissions & Licensing Information about reproducing this article in parts (figures, tables) or
in its entirety can be found online at:
http://www.aappublications.org/site/misc/Permissions.xhtml
Reprints Information about ordering reprints can be found online:
http://www.aappublications.org/site/misc/reprints.xhtml

Downloaded from www.aappublications.org/news by guest on November 17, 2018

Prevalence and Treatment of Mental Disorders Among US Children in the 2001−
2004 NHANES
Kathleen Ries Merikangas, Jian-Ping He, Debra Brody, Prudence W. Fisher, Karen
Bourdon and Doreen S. Koretz
Pediatrics 2010;125;75
DOI: 10.1542/peds.2008-2598 originally published online December 14, 2009;

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://pediatrics.aappublications.org/content/125/1/75

Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it
has been published continuously since 1948. Pediatrics is owned, published, and trademarked by
the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,
60007. Copyright © 2010 by the American Academy of Pediatrics. All rights reserved. Print ISSN:
1073-0397.

Downloaded from www.aappublications.org/news by guest on November 17, 2018