Please note: An erratum has been published for this issue. To view the erratum, please click here.

Morbidity and Mortality Weekly Report

Surveillance Summaries / Vol. 65 / No. 3 April 1, 2016

Prevalence and Characteristics of Autism Spectrum

Disorder Among Children Aged 8 Years Autism
and Developmental Disabilities Monitoring
Network, 11 Sites, United States, 2012

U.S. Department of Health and Human Services

Centers for Disease Control and Prevention
 Surveillance Summaries

CONTENTS
Introduction.............................................................................................................2
Methods.....................................................................................................................3
Results........................................................................................................................6
Discussion.................................................................................................................8
Limitations............................................................................................................. 11
Future Study Directions.................................................................................... 11
Conclusion............................................................................................................. 12
References.............................................................................................................. 12

The MMWR series of publications is published by the Center for Surveillance, Epidemiology, and Laboratory Services, Centers for Disease Control and Prevention (CDC),
U.S. Department of Health and Human Services, Atlanta, GA 30329-4027.
Suggested citation: [Author names; first three, then et al., if more than six.] [Title]. MMWR Surveill Summ 2016;65(No. SS-#):[inclusive page numbers].

Centers for Disease Control and Prevention

Thomas R. Frieden, MD, MPH, Director
Harold W. Jaffe, MD, MA, Associate Director for Science
Joanne Cono, MD, ScM, Director, Office of Science Quality
Chesley L. Richards, MD, MPH, Deputy Director for Public Health Scientific Services
Michael F. Iademarco, MD, MPH, Director, Center for Surveillance, Epidemiology, and Laboratory Services

MMWR Editorial and Production Staff (Serials)

Sonja A. Rasmussen, MD, MS, Editor-in-Chief Martha F. Boyd, Lead Visual Information Specialist
Charlotte K. Kent, PhD, MPH, Executive Editor Maureen A. Leahy, Julia C. Martinroe,
Christine G. Casey, MD, Editor Stephen R. Spriggs, Moua Yang, Tong Yang,
Teresa F. Rutledge, Managing Editor Visual Information Specialists
David C. Johnson, Lead Technical Writer-Editor Quang M. Doan, MBA, Phyllis H. King, Terraye M. Starr,
Jeffrey D. Sokolow, MA, Project Editor Information Technology Specialists

MMWR Editorial Board

Timothy F. Jones, MD, Chairman William E. Halperin, MD, DrPH, MPH Jeff Niederdeppe, PhD
Matthew L. Boulton, MD, MPH King K. Holmes, MD, PhD Patricia Quinlisk, MD, MPH
Virginia A. Caine, MD Robin Ikeda, MD, MPH Patrick L. Remington, MD, MPH
Katherine Lyon Daniel, PhD Rima F. Khabbaz, MD Carlos Roig, MS, MA
Jonathan E. Fielding, MD, MPH, MBA Phyllis Meadows, PhD, MSN, RN William L. Roper, MD, MPH
David W. Fleming, MD Jewel Mullen, MD, MPH, MPA William Schaffner, MD
 Surveillance Summaries

Prevalence and Characteristics of Autism Spectrum Disorder Among

Children Aged 8 Years Autism and Developmental Disabilities
Monitoring Network, 11 Sites, United States, 2012
Deborah L. Christensen, PhD1
Jon Baio, EdS1
Kim Van Naarden Braun, PhD1
Deborah Bilder, MD2
Jane Charles, MD3
John N. Constantino, MD4
Julie Daniels, PhD5
Maureen S. Durkin, PhD6
Robert T. Fitzgerald, PhD4
Margaret Kurzius-Spencer, PhD7
Li-Ching Lee, PhD8
Sydney Pettygrove, PhD7
Cordelia Robinson, PhD9
Eldon Schulz, MD10
Chris Wells, PhD11
Martha S. Wingate, DrPH12
Walter Zahorodny, PhD13
Marshalyn Yeargin-Allsopp, MD1
1Division of Congenital and Developmental Disorders, National Center on Birth Defects and Developmental Disabilities, CDC
2University of Utah, Salt Lake City
3Medical University of South Carolina, Charleston
4Washington University in St. Louis, Missouri
5University of North Carolina, Chapel Hill
6University of WisconsinMadison
7University of Arizona, Tucson
8Johns Hopkins University
9University of Colorado at Denver and Health Sciences Center
10University of Arkansas for Medical Sciences, Little Rock
11Colorado Department of Public Health and Environment, Denver
12University of Alabama at Birmingham
13Rutgers UniversityNew Jersey Medical School, Newark

Abstract
Problem/Condition: Autism spectrum disorder (ASD).
Period Covered: 2012.
Description of System: The Autism and Developmental Disabilities Monitoring (ADDM) Network is an active surveillance system
that provides estimates of the prevalence and characteristics of ASD among children aged 8 years whose parents or guardians reside
in 11 ADDM Network sites in the United States (Arkansas, Arizona, Colorado, Georgia, Maryland, Missouri, New Jersey, North
Carolina, South Carolina, Utah, and Wisconsin). Surveillance to determine ASD case status is conducted in two phases. The first
phase consists of screening and abstracting comprehensive evaluations performed by professional service providers in the community.
Data sources identified for record review are categorized as either 1) education source type, including developmental evaluations to
determine eligibility for special education services or 2) health care source type, including diagnostic and developmental evaluations.
The second phase involves the review of all abstracted evaluations by trained clinicians to determine ASD surveillance case status. A
child meets the surveillance case definition for ASD if one or more comprehensive evaluations of that child completed by a qualified
professional describes behaviors that are consistent with the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition,
Text Revision diagnostic criteria for any of the following conditions: autistic disorder, pervasive developmental disordernot otherwise
specified (including atypical autism), or Asperger disorder. This
report provides ASD prevalence estimates for children aged
Corresponding author: Deborah L. Christensen, Division of Congenital
and Developmental Disorders. Telephone: 404-498-3836; E-mail:
8 years living in catchment areas of the ADDM Network sites
dchristensen@cdc.gov. in 2012, overall and stratified by sex, race/ethnicity, and the type
of source records (education and health records versus health

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 1
 Surveillance Summaries

records only). In addition, this report describes the proportion of children with ASD with a score consistent with intellectual disability
on a standardized intellectual ability test, the age at which the earliest known comprehensive evaluation was performed, the proportion
of children with a previous ASD diagnosis, the specific type of ASD diagnosis, and any special education eligibility classification.
Results: For 2012, the combined estimated prevalence of ASD among the 11 ADDM Network sites was 14.6 per 1,000 (one in 68)
children aged 8 years. Estimated prevalence was significantly higher among boys aged 8 years (23.6 per 1,000) than among girls
aged 8 years (5.3 per 1,000). Estimated ASD prevalence was significantly higher among non-Hispanic white children aged 8 years
(15.5 per 1,000) compared with non-Hispanic black children (13.2 per 1,000), and Hispanic (10.1 per 1,000) children aged 8 years.
Estimated prevalence varied widely among the 11 ADDM Network sites, ranging from 8.2 per 1,000 children aged 8 years (in the
area of the Maryland site where only health care records were reviewed) to 24.6 per 1,000 children aged 8 years (in New Jersey,
where both education and health care records were reviewed). Estimated prevalence was higher in surveillance sites where education
records and health records were reviewed compared with sites where health records only were reviewed (17.1 per 1,000 and 10.7
per 1,000 children aged 8 years, respectively; p<0.05). Among children identified with ASD by the ADDM Network, 82% had a
previous ASD diagnosis or educational classification; this did not vary by sex or between non-Hispanic white and non-Hispanic
black children. A lower percentage of Hispanic children (78%) had a previous ASD diagnosis or classification compared with non-
Hispanic white children (82%) and with non-Hispanic black children (84%). The median age at earliest known comprehensive
evaluation was 40 months, and 43% of children had received an earliest known comprehensive evaluation by age 36 months. The
percentage of children with an earliest known comprehensive evaluation by age 36 months was similar for boys and girls, but was
higher for non-Hispanic white children (45%) compared with non-Hispanic black children (40%) and Hispanic children (39%).
Interpretation: Overall estimated ASD prevalence was 14.6 per 1,000 children aged 8 years in the ADDM Network sites in 2012. The
higher estimated prevalence among sites that reviewed both education and health records suggests the role of special education systems in
providing comprehensive evaluations and services to children with developmental disabilities. Disparities by race/ethnicity in estimated
ASD prevalence, particularly for Hispanic children, as well as disparities in the age of earliest comprehensive evaluation and presence of
a previous ASD diagnosis or classification, suggest that access to treatment and services might be lacking or delayed for some children.
Public Health Action: The ADDM Network will continue to monitor the prevalence and characteristics of ASD among children
aged 8 years living in selected sites across the United States. Recommendations from the ADDM Network include enhancing
strategies to 1) lower the age of first evaluation of ASD by community providers in accordance with the Healthy People 2020
goal that children with ASD are evaluated by age 36 months and begin receiving community-based support and services by age
48 months; 2) reduce disparities by race/ethnicity in identified ASD prevalence, the age of first comprehensive evaluation, and
presence of a previous ASD diagnosis or classification; and 3) assess the effect on ASD prevalence of the revised ASD diagnostic
criteria published in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition.

Introduction five subtypes of autism, including autistic disorder, Asperger

disorder, pervasive developmental disordernot otherwise
Autism spectrum disorder (ASD) is a developmental disability specified (PDD-NOS), childhood disintegrative disorder, and
characterized by social and communication impairments and by Retts disorder (4). The first three subtypes comprise autism
restricted interests and repetitive behaviors (1). The first studies spectrum disorder (ASD), whereas the latter two conditions
of the prevalence of autism were published in the 1960s and belong to the wider category of pervasive developmental
1970s, when autism was thought to be a very severe condition, disorders. The fifth edition of DSM, which was published in
usually accompanied by intellectual disability (2). These studies 2013 (5), redefined ASD as a single disorder, along with other
reported the prevalence to be approximately four to five cases changes in the diagnostic classification of ASD. For this report,
per 10,000 children. Autism was first distinguished as a unique the evaluations contained in childrens records were conducted
clinical diagnosis by the American Psychiatric Association with no later than 2012, and therefore DSM-IV-TR diagnostic
the publication in 1980 of the third edition of the Diagnostic criteria were used in the ASD surveillance case definition to
and Statistical Manual of Mental Disorders (DSM-III) (3), which estimate ASD prevalence and characteristics.
provided diagnostic criteria for infantile autism and pervasive Substantial increases in the estimated prevalence of ASD
developmental disorder. Since that time, autism has become in the United States have been reported since the 1990s. Two
recognized as a spectrum of behavioral characteristics, which studies conducted in the late 1980s that used DSM-III screening
results in varying degrees of functional limitations. In 1994, and diagnostic criteria for pervasive developmental disorder
the Diagnostic and Statistical Manual of Mental Disorders, Fourth estimated prevalence as 3.3 cases per 10,000 children aged
Edition (DSM-IV) introduced revised diagnostic criteria and

2 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

318 years (6) and 3.6 cases per 10,000 children aged 812 years Methods
(7). Since then, increases in estimated ASD prevalence have
been measured, using data from special education and other ADDM Network Sites
administrative records (810), national surveys (1114), and
The ADDM Network is an active surveillance system that
active public health surveillance conducted through CDCs
provides estimates of the prevalence and characteristics of
Metropolitan Atlanta Developmental Disabilities Surveillance
ASD among children aged 8 years whose parents or guardians
Program (MADDSP) and its extended surveillance network, the
reside within 11 ADDM sites in the United States (selected
Autism and Developmental Disabilities Monitoring (ADDM)
counties or parts of counties in Arkansas, Arizona, Colorado,
Network. MADDSP first estimated ASD prevalence among
Georgia, Maryland, Missouri, New Jersey, North Carolina,
children aged 310 years in 1996 to be 3.4 per 1,000 children
South Carolina, Utah, and Wisconsin). The ADDM Network
aged 310 years (15). Subsequently, the larger ADDM Network
uses multisite, multiple-source surveillance based on a review
estimated prevalence across multiple U.S. sites every 2 years
of behavioral descriptions and ASD diagnoses documented in
during 20002010. The most recent prevalence estimate from
comprehensive developmental evaluations present in childrens
the ADDM Network for children aged 8 years was 14.7 per
health and education records, using a model developed by
1,000 children in 2010 (16), compared with 11.3 in 2008 (17),
CDCs MADDSP (15,23). The ADDM Network sites were
9.0 in 2006 (18), 6.6 in 2002 (19), and 6.7 in 2000 (20).
selected through a competitive process, with preference for a
The American Academy of Pediatrics recommends that
diverse population in terms of race/ethnicity. Each ADDM site
pediatric health care providers administer two ASD screenings,
functions as a public health authority under the Health Insurance
at ages 18 and 24 months, using a valid and reliable screening
Portability and Accountability Act of 1996 Privacy Rule and
tool (21). Children whose screening results are concerning
meets applicable local Institutional Review Board, privacy, and
should subsequently receive a comprehensive developmental
confidentiality requirements under 45 CFR 46 (24).
evaluation from a general or developmental pediatrician, child
neurologist, child psychiatrist, or child psychologist, which can
be obtained privately or through the Part C (ages 0<3 years) Case Ascertainment
or Part B (ages 321 years) programs of the Individuals with Children eligible for case ascertainment were born in 2004,
Disabilities Education Act (http://idea.ed.gov/explore/home) and their parents or guardians lived in site-specific ADDM
supported by each state. To support and measure progress in Network surveillance catchment areas at some time during
early identification, the Healthy People 2020 initiative includes a 2012. At each site, surveillance data were linked to the states
goal to increase the percentage of children with ASD who receive 2004 birth certificate records to obtain data on race/ethnicity
their first comprehensive evaluation by age 36 months by 10%, and other demographic characteristics. If a birth certificate
from the baseline of 42.7% in 2006 to the goal of 47.0% in 2020 match was not made, the child was assumed to have been born
(22). ADDM Network ASD surveillance data for children aged outside the state. No clinical examinations of children were
8 years are used to evaluate progress toward this goal. performed by ADDM Network staff.
This report describes estimated ASD prevalence and ADDM Network investigators use a two-phase surveillance
characteristics among children aged 8 years in the ADDM approach to ascertain potential ASD cases. The first phase
Network in 2012. This includes 1) total estimated ASD involves screening and abstracting records from multiple
prevalence as well as prevalence by surveillance site, sex, and data sources in the community, including special education
race/ethnicity and 2) characteristics of children with ASD, programs and health care providers who evaluate and treat
including presence of intellectual disability, age at earliest known children with developmental disabilities. Agreements to
comprehensive evaluation, presence of a previous ASD diagnosis access records are made at the institutional level in the form
or educational classification, age at previous ASD diagnosis and of contracts, memoranda, or other formal agreements. In the
diagnosis subtype, and special education eligibility classification. second phase, all abstracted evaluations are compiled and
The intended audience for this report includes health care reviewed by clinicians with specialized training in the evaluation
providers, early intervention service providers, therapists, school and diagnosis of ASD, including physicians, psychologists, and
psychologists, educators, researchers, policymakers, and program speech/language pathologists. These clinician reviewers follow
administrators seeking to understand and provide for the needs the ADDM surveillance protocol to determine ASD case status
of persons with ASD and their families. These data can be used to and to maintain reliability.
help plan for service needs, stimulate research into etiology and Data sources identified for record review are categorized
risk factors, and initiate and implement policies that promote as either 1) education source type, including developmental
optimal outcomes in health care and education. evaluations to determine eligibility for special education

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 3
 Surveillance Summaries

services or 2) health care source type, including diagnostic and deidentified and reviewed systematically by clinicians who have
developmental evaluations. All ADDM Network sites have undergone standardized training to determine ASD case status
agreements in place to access records at health care sources. using a coding scheme based on the DSM-IV-TR (1) criteria for
For the 2012 surveillance year, six sites (Arizona, Georgia, ASD. A child meets the surveillance case definition for ASD if
New Jersey, North Carolina, South Carolina, and Utah) also behaviors described in the composite record are consistent with
reviewed education records in all or most of the surveillance the DSM-IV-TR diagnostic criteria for any of the following
area. In the Maryland site, education records were reviewed in conditions: autistic disorder, PDD-NOS (including atypical
one of the six participating counties, and in the Colorado site, autism), or Asperger disorder. ASD surveillance case criteria were
education records were reviewed for some of the public school based on DSM-IV-TR because surveillance was conducted using
districts in one of the seven counties in the surveillance area. records generated before or during 2012, prior to publication of
For these two surveillance sites, only health care source type new diagnostic criteria in the Diagnostic and Statistical Manual
records were reviewed in the remaining counties. Three sites of Mental Disorders, Fifth Edition (DSM-5) (5). For the majority
(Arkansas, Missouri, and Wisconsin) reviewed records only at of children, one clinician reviews the composite record. If a child
health care sources. meets the surveillance case criteria, but the reviewer is uncertain
In the first phase of surveillance, ADDM Network sites whether ASD is the most appropriate classification, a second,
identify source records to review on the basis of a childs independent review is done. Following the second review, the two
year of birth and either 1) eligibility classifications in special reviewers meet and come to a consensus on the childs case status.
education or 2) International Classification of Diseases, Ninth
Revision, Clinical Modification (ICD-9-CM) billing codes for Descriptive Characteristics
select childhood disabilities or conditions. Childrens records
are screened to confirm year of birth and whether the parent The diagnostic summaries from each evaluation record were
or guardian of the child lived in the surveillance area at any abstracted for each child, including age at and subtype of any
time during the surveillance year. For children meeting age previous ASD diagnoses. Children were considered to have a
and residency requirements, the source files are screened previously documented ASD classification if the child received
for specific behavioral or diagnostic descriptions defined by a DSM-IV-TR diagnosis of autistic disorder, Asperger disorder,
ADDM as triggers for abstraction. These triggers include a PDD-NOS, or ASD-NOS, or if ASD was documented by an
documented ASD classification (either a diagnosis of ASD ICD-9-CM billing code at any time from birth through the
or a special education placement category of ASD) and/or year when the child reached age 8 years, or if the child received
descriptions of behaviors consistent with an ASD diagnosis special education services under an autism eligibility during
(e.g., limited or no interaction with other children or prefers the surveillance year. Information was collected on childrens
objects over persons). If abstraction triggers are found, available functional abilities, including scores on standardized tests of
information from birth through the current surveillance year intellectual ability. The most recently recorded scores from
is abstracted, including: 1) information on demographic tests of a childs intellectual ability were used to categorize
characteristics; 2) other medical conditions; 3) evaluation the child in the intellectual disability range if the intelligence
dates and verbatim descriptions of behaviors consistent with quotient (IQ) score was 70, in the borderline range if the
ASD from comprehensive developmental evaluations by score was 7185, and in the average to above-average range if
community professionals; 4) community professional type and the score was >85. The childs age at the earliest comprehensive
degree (e.g., MD [neurologist, psychiatrist, or developmental evaluation was documented and is reported as the median age
pediatrician], PhD [psychologist], or EdS [education specialist]); at the earliest comprehensive evaluation in months and as the
5) developmental history, including statements about parental percentage of children with an earliest known comprehensive
or professional concerns that the childs development was evaluation performed by age 36 months. Information also was
atypical; 6) special education service category; 7) scores from recorded about the age at which developmental concerns were
intelligence quotient (IQ), adaptive, and autism diagnostic documented in the records. Analyses of the age at earliest known
tests; and 8) evaluation conclusions. The most recent eligibility comprehensive evaluation and the age at which developmental
classification for receiving special education services (e.g., autism concerns first were documented were restricted to children who
or learning disability) is collected from special education records. were born in the state in which they resided at age 8 years. This
For all abstracted evaluations, information from multiple sources restriction was imposed to reduce bias that might have resulted
is combined into one composite summary record for each child. from the unavailability of evaluations performed early in life
In the second phase of surveillance, referred to as clinician when the child was residing in a state other than the state in
review, the abstracted composite evaluation records are which the ADDM Network site was located.

4 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

Quality Assurance Hispanic). Other race/ethnicity groups were represented by too

few children to generate stable estimates of ASD prevalence at all
All ADDM Network sites follow the same quality assurance
surveillance sites. The race/ethnicity of each child whose records
protocol. In the first phase of case ascertainment, screening
were abstracted was determined from information contained in
and abstraction of source records are monitored for accuracy
source records or, if not found in the source records, from birth
by means of a 10% random sample of records to check the
certificates (when available). Hispanic refers to all children who
accuracy of data collection as well as the appropriate selection
are of Hispanic ethnicity, regardless of race. Overall prevalence
of the record for abstraction. Initial interrater reliability
estimates included all children identified with ASD regardless of
among ASD clinician reviewers was established to a minimum
sex, race/ethnicity, or intellectual ability and therefore were not
standard of 90% agreement on their decision about whether
limited to children with available data on these characteristics.
the child meets the ASD surveillance case definition defined in
Statistical tests were used and confidence interval (CI) estimates
the ADDM study protocol prior to the beginning of the second
were calculated following the assumption that the observed counts
phase of case ascertainment. Subsequently, interrater reliability
of ASD surveillance cases were drawn from an underlying Poisson
for clinician reviewers is monitored on an ongoing basis using
sampling distribution. Pearson chi-square tests and prevalence
a blinded, random 10% sample of abstracted records that are
ratios (PR) were used to examine the association between ASD
scored independently by two reviewers. The final interrater
prevalence estimates and characteristics of children with ASD by
agreement for determining surveillance ASD case status (ASD
surveillance site, record source type, sex, and race/ethnicity. Exact
case versus not ASD case) was 90.4% when reliability samples
tests were used when the number of children was fewer than five.
from all ADDM Network sites were combined (K = 0.8).
The nonparametric median test was used to determine differences
in median age at first evaluation for ASD and earliest known ASD
Analytic Methods diagnosis by sex and race/ethnicity. Statistical significance was set
The prevalence estimate of ASD among children in the at p<0.05. All analyses were performed by using SAS statistical
ADDM Network was calculated as the number of children aged software (SAS Institute, Cary, North Carolina).
8 years who met the surveillance ASD case definition across the
11 ADDM Network sites in 2012 divided by the number of Evaluation Methods
children aged 8 years residing in the counties comprising the
Some children who were identified for screening could not
11 surveillance sites. Population denominators used were obtained
be included in the ADDM Network ASD case determination
from CDCs National Center for Health Statistics (NCHS)
process because some or all of the education and health records
vintage 2014 postcensal bridged-race population estimates for
could not be found for review. Therefore, an analysis was
2012 (http://www.cdc.gov/nchs). In the Arizona site, only part of a
performed to determine the potential effect of these missing
county was included in the surveillance catchment area. Therefore,
records on ASD prevalence estimates. All children initially
the number of children in this county who lived within the
identified for screening were first stratified by two factors closely
surveillance area was estimated in order to obtain the appropriate
associated with final case status: information source (health
denominator. This was done by obtaining the third-grade school
source type only, education source type only, or both source
enrollment counts for the years 20122013 for the public school
types) and the presence or absence of either an autism special
districts included in the surveillance area from the National Center
education eligibility or an ICD-9-CM code for ASD, collectively
for Education Statistics (https://nces.ed.gov). The number of
forming six strata. The potential number of cases that might
third-grade students enrolled in the public school districts included
have been identified if missing records had been included was
in the surveillance area was divided by the number of third-grade
estimated by assuming that within each of these six strata, the
students enrolled in all of the public school districts in the county
proportion of children with ASD in each stratum (with and
to obtain the proportion of students enrolled in participating
without missing records) would be similar. Subsequently, the
school districts. This proportion was then applied to the NCHS
proportion of children meeting the ASD surveillance case
vintage 2014 postcensal bridged-race population estimate for
definition was applied to the number of children with missing
each county in 2012 to obtain the relevant denominator. The
records in the same stratum to estimate the number of missed
bridged-race categories used in this report include non-Hispanic
cases and the corresponding increase in prevalence. Estimates
white, non-Hispanic black, Hispanic, American Indian/Alaska
from all six strata were added to produce the total for each
Native, and Asian/Pacific Islander. Data from all ADDM sites
site. The analysis of the potential effect of missing records was
were pooled to produce combined ASD prevalence estimates.
performed for evaluation purposes, and the prevalence estimates
Prevalence estimates were stratified by surveillance site, sex, and
presented in this report do not reflect this adjustment.
race/ethnicity (i.e., non-Hispanic white, non-Hispanic black, and

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 5
 Surveillance Summaries

Comparison of Surveillance Sites between Overall ASD Prevalence Estimates

2010 and 2012 Overall estimated ASD prevalence for the 2012 surveillance
For eight sites (Arizona, Colorado, Georgia, Missouri, New year was 14.6 per 1,000 (one in 68) children aged 8 years,
Jersey, North Carolina, Utah, and Wisconsin), the geographic on the basis of pooled data from 11 ADDM sites (range: 8.2
area covered and record source type reviewed were the same [Maryland, health records only reviewed] to 24.6 [in New
in 2012 and 2010. Therefore, these eight sites were included Jersey]) (Table 2). Estimated ASD prevalence was highest in
in analyses comparing estimated ASD prevalence in 2010 New Jersey (24.6), Maryland (education and health records
and 2012. For two sites (Arkansas and Maryland), there reviewed, 18.2), Utah (17.3), and North Carolina (16.9). The
was a change in geographic area and/or record source type. seven areas (Arizona, Georgia, Maryland [education and health
South Carolina contributed data in 2012 but not in 2010. records reviewed], New Jersey, North Carolina, South Carolina,
An ADDM Network site located in Alabama conducted and Utah) with access to both education and health care sources
ASD surveillance for part of the 2012 surveillance year, but had higher estimated ASD prevalence compared with the five
because of the loss of access to health care data sources, data areas (Arkansas, Colorado, Maryland [health records only
from Alabama were not complete for the 2012 surveillance reviewed], Missouri, and Wisconsin) with limited or no access to
year and are not included in this report. education records (17.1 and 10.7 per 1,000 children aged 8 years,
respectively; PR: 1.6; 95% CI: 1.51.7; p<0.001) (Figure 1).

Results Prevalence by Sex and Race/Ethnicity

Across the 11 ADDM Network sites, estimated ASD
Population Characteristics
prevalence among children aged 8 years was 23.6 per 1,000
The geographic surveillance area for the 11 ADDM (one in 42) boys and 5.3 per 1,000 (one in 189) girls; site-
Network sites in 2012 included 346,978 children aged specific ASD estimates for boys ranged from 13.9 per 1,000 (in
8 years, which comprised 8.5% of the U.S. population Maryland [health records only reviewed]) to 39.1 per 1,000 (in
of children aged 8 years for that year (Table 1). The New Jersey), and for girls from 2.2 per 1,000 (in Maryland) to
population distribution of children by race/ethnicity 9.3 per 1,000 (in New Jersey) (Table 2). The overall prevalence
varied by study site. In the pooled data, the population ratio for boys compared with girls was 4.5 (95% CI: 4.24.8;
was 53.3% non-Hispanic white, 21.4% non-Hispanic p<0.001); site-specific male-to-female prevalence ratios ranged
black, 19.9% Hispanic, 4.8% Asian/Pacific Islander, and from 4.1 (in Colorado) to 6.3 (in Maryland [health records
0.6% American Indian/Alaska Native. only reviewed]) and were all statistically significant (Table 2).
Estimated prevalence among non-Hispanic white children
Record Review (15.5 per 1,000) was significantly higher than it was among
A total of 48,304 records for 38,038 children aged 8 years non-Hispanic black children (13.2 per 1,000; PR: 1.2, 95% CI:
were reviewed from education and health care sources. Among 1.11.3; p<0.001), Asian/Pacific Islander children (11.3 per
these, the source records of 9,629 (19%) children met the 1,000; PR: 1.4, 95% CI: 1.21.6; p<0.001), and Hispanic
criteria for abstraction and subsequently were reviewed by children (10.1 per 1,000; PR: 1.5, 95% CI: 1.41.7; p<0.001)
clinicians. Of these 9,629 children, 5,063 (53%) met the ASD (Table 3). Prevalence ratios by sex and race/ethnicity were similar
surveillance case criteria. between the areas that reviewed education and health records
and the areas that reviewed health records only (Tables 2 and 3).
Birth Certificate Linkage
Intellectual Ability
Of the 5,063 children meeting the ASD surveillance case
criteria, 3,881 children (77%) were born in the state where the Nine ADDM Network areas (Arizona, Arkansas, Colorado,
ADDM Network surveillance site is located, as confirmed by Georgia, Maryland [education and health records review area],
a match to a birth certificate from that state. This percentage New Jersey, North Carolina, South Carolina, and Utah) had
ranged from 68% (South Carolina) to 86% (Missouri). The data on intellectual ability for 70% of ASD cases (range:
percentage of children who were matched to a birth certificate 70% [Arkansas and New Jersey]92% [in North Carolina]).
did not vary by sex or race/ethnicity. In these nine areas, 3,390 (80%) of 4,234 children with
ASD had data on intellectual ability; for most areas, this
percentage did not vary by sex or race/ethnicity, with the

6 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

exception of Georgia, where the percentage of ASD cases concerns at age 36 months was significantly higher for
with data on intellectual ability was significantly higher children with ASD and intellectual disability compared with
for boys compared with girls (87% and 79%, respectively; children with ASD without intellectual disability (95% and
p<0.05). Among all 3,390 children, 31.6% were classified 84%, respectively; p<0.001).
in the range of intellectual disability (IQ score 70 or the Using combined data from all sites for children meeting
presence of an examiners statement of intellectual disability), the ASD surveillance case criteria and restricting to children
24.5% were classified in the borderline range (IQ: 7185), born in the state where the ADDM Network surveillance
and 43.9% were classified in the average or above average site was located, the earliest known comprehensive
range (IQ >85 or the presence of an examiners statement of evaluation occurred at age 36 months for 43% of children,
average or above average intellectual ability) (Figure 2). The between 37 and 48 months for 20% of children, and after
percentage of children classified in the intellectual disability 48 months for the remaining 38% of children (Table 4).
range varied widely across the nine areas, ranging from 20% This percentage did not vary between boys and girls (42%
(in Utah) to 50% (in Arkansas). The percentage of ASD cases and 45%, respectively; p>0.05), but was significantly higher
classified in the intellectual disability range was significantly for non-Hispanic white children (45%) compared with non-
higher among girls compared with boys in all nine areas Hispanic black children (40%; p<0.05) and with Hispanic
combined (37% and 30%, respectively; p<0.01). children (39%; p<0.05) (data not shown). Children with
Combining data from all nine sites, the estimated prevalence ASD and intellectual disability were more likely to have an
of ASD with intellectual disability was 4.0 per 1,000 and earliest known comprehensive evaluation by age 36 months
ranged from 1.8 per 1,000 (in Colorado) to 5.3 per 1,000 (in compared with children with ASD without intellectual
North Carolina) (Figure 3). The estimated prevalence of ASD disability (55% and 39%, respectively; p<0.001) (data not
without intellectual disability was 8.7 per 1,000 and ranged shown). The median age at earliest known comprehensive
from 4.2 per 1,000 (in Arkansas) to 12.2 per 1,000 (in New evaluation was 40 months, ranging from 30 months (North
Jersey) (Figure 3). There was a greater male-to-female prevalence Carolina) to 48 months (Arkansas) (data not shown).
ratio for ASD without intellectual disability (PR: 5.1; 95%
CI: 4.65.7; p<0.001) than for ASD with intellectual disability Earliest Known ASD Diagnosis and
(PR: 3.7; 95% CI: 3.24.3; p<0.001) (Figure 4). The estimated
prevalence of ASD with intellectual disability was significantly
Diagnosis Category
lower for non-Hispanic white children (3.3 per 1,000) compared On the basis of pooled data from all ADDM Network sites,
with non-Hispanic black children (5.8 per 1,000; PR: 0.6; 95% 74% of children identified with ASD had an earliest known
CI: 0.50.7; p<0.001) (Figure 4). DSM-IV-TR ASD diagnosis of autistic disorder (46%), ASD-
NOS/PDD-NOS (44%), or Asperger disorder (10%) given
by a community provider (Table 5). The median age at the
Early Developmental Concerns and earliest known diagnosis was 50 months overall and was lower
Earliest Comprehensive Evaluation for autistic disorder (46 months) compared with ASD-NOS/
Analyses of the presence of early developmental concerns and PDD-NOS (49 months; p<0.01) and with Asperger disorder
earliest comprehensive evaluation were restricted to children (74 months; p<0.001) (Table 5). Within each specific diagnosis
born in the state where the ADDM Network surveillance subtype, there were no differences in median age at earliest
site was located in order to reduce bias associated with the known diagnosis by sex or race/ethnicity (data not shown).
inability to review early evaluations for children who moved
from their state of birth prior to ascertainment by the ADDM Special Education Eligibility
Network at age 8 years. Across all ADDM Network sites, 87%
of children meeting the ASD surveillance case criteria had The seven ADDM Network areas that reviewed records at
documentation of developmental concerns at age 36 months education sources obtained data on the eligibility categories
in a health or education record (Table 4). This percentage was through which children with ASD were served in the public
similar for areas that reviewed education and health records school special education system. Combined data from these
compared with areas that reviewed health records only (87% seven areas indicate that 74% of children with ASD had
and 88%, respectively); the percentage was significantly higher special education records; this percentage ranged from 55%
for non-Hispanic black children (91%) and for Hispanic (Utah) to 92% (Arizona). Among these children, more than
children (89%) compared with non-Hispanic white children half had a primary special education eligibility classification of
(86%; p<0.05). The percentage of children with developmental autism (range: 53% [in Utah]70% [in Maryland education

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 7
 Surveillance Summaries

and health records review area]) (Table 6). Combining data Although overall estimated ASD prevalence between 2010
from all seven areas, the percentage of children with an and 2012 was similar across the eight sites that were comparable
autism eligibility classification did not vary between boys between these 2 years, the same was not true of all of the
(61%) and girls (57%; p>0.05) or between non-Hispanic individual surveillance sites, three of which had significantly
white (56%) and Hispanic children (56%; p>0.05) but was different prevalence estimates in 2012 compared with 2010.
greater for non-Hispanic black children (65%) compared Between these two surveillance years, ASD prevalence increased
with non-Hispanic white (p<0.01) and Hispanic (p<0.01) by 16% in Wisconsin and by 12% in New Jersey and decreased
children (data not shown). by 19% in Missouri.

Previously Documented ASD Classification Evaluation of the Effect of Missing Records

Across the 11 ADDM Network sites, 82% of children who An evaluation of the effect of missing records suggested
met the ASD surveillance case criteria had either a previous that estimated ASD prevalence might have increased by 0.1%
diagnosis of ASD or a documented eligibility classification of (in Wisconsin) to 3.3% (in Utah) if missing records had
autism in the special education system, 9% had a suspicion been available for review. Across all 11 sites, estimated ASD
of ASD documented in an evaluation, and the remaining prevalence might have increased by <1% in four sites (Arizona,
9% had no mention of ASD in the records (Figure 5). At Colorado, Missouri, and Wisconsin), by 1%<2% in three sites
individual ADDM Network sites, the percentage of children (Georgia, North Carolina, and New Jersey) and by 2.0%3.3%
with a previous ASD diagnosis or eligibility classification in four sites (Arkansas, Maryland, South Carolina, and Utah).
ranged from 68% (Colorado) to 93% (Missouri) (Figure 5).
The percentage of children with a previous ASD diagnosis or
eligibility classification was the same for boys and girls (82%) Discussion
and similar for non-Hispanic white and non-Hispanic black Estimated ASD prevalence among children aged 8 years
children (82% and 84%, respectively), but was significantly in the ADDM Network in 2012 was 14.6 per 1,000, or one
lower for Hispanic children (78%) compared with non- in 68. Estimated prevalence was four and a half times higher
Hispanic white children (p<0.01) and non-Hispanic black among boys than among girls; estimated ASD prevalence was
children (p<0.001) (data not shown). 23.6 per 1,000 boys (one in 42 boys) and 5.3 per 1,000 girls
(one in 189 girls). Prevalence estimates varied widely among the
Comparison of ASD Prevalence Estimates 11 ADDM Network sites, ranging from 8.2 per 1,000 children
Between 2010 and 2012 aged 8 years (in the area of the Maryland site where health
records only were reviewed) to 24.6 per 1,000 (in New Jersey,
For eight sites (Arizona, Colorado, Georgia, Missouri, New
where both education and health records were reviewed). The
Jersey, North Carolina, Utah, and Wisconsin), the geographic
estimated prevalence of ASD with intellectual disability was 4.0
areas covered and record source types reviewed were the same
per 1,000 overall and ranged from 1.8 per 1,000 (in Colorado)
for 2010 and 2012. On the basis of combined data from these
to 5.3 per 1,000 (in North Carolina). The estimated prevalence
eight sites in each respective year, estimated ASD prevalence
of ASD without intellectual disability was 8.7 per 1,000 overall,
was 15.1 and 15.2 per 1,000 children aged 8 years in 2010 and
ranged from 4.2 per 1,000 (in Arkansas) to 12.2 per 1,000
2012, respectively (p>0.05) (Table 7). Estimated ASD prevalence
(in New Jersey), and exceeded the estimated prevalence of
for male, female, non-Hispanic white, non-Hispanic black, or
ASD with intellectual disability in all sites. Across all ADDM
Hispanic children did not differ significantly between 2010 and
Network sites, estimated ASD prevalence was 20% higher
2012. Five of these eight sites collected data on intellectual ability
among non-Hispanic white compared with non-Hispanic
for 70% of the children identified with ASD. On the basis
black children, 40% higher among non-Hispanic white
of combined data from these five sites in each respective year,
compared with Asian/Pacific Islander children, 50% higher
estimated ASD prevalence was 17.5 and 17.6 per 1,000 children
among non-Hispanic white compared with Hispanic children,
aged 8 years in 2010 and 2012, respectively (p>0.05) and was
and 30% higher among non-Hispanic black compared with
similar between 2010 and 2012 for male, female, non-Hispanic
Hispanic children.
white, non-Hispanic black, and Hispanic children (Table 8).
The overall prevalence estimate for 2012 was nearly
Prevalence estimates were similar for 2010 and 2012 for children
identical to the reported estimate for the ADDM Network in
with ASD with intellectual disability and for children with ASD
2010 of 14.7 per 1,000, or one in 68 children aged 8 years.
without intellectual disability.
However, because of differences between 2010 and 2012 in the

8 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

geographic area covered and record source types reviewed for (NSCH) is a nationally representative survey of households
some individual ADDM Network sites, comparing the overall with children aged 017 years in the United States. Both
prevalence estimates might be misleading. For this reason, surveys base ASD prevalence estimates on parent or caregiver
comparisons of ASD prevalence estimates between 2010 and report of being told by a doctor or other health care provider
2012 were restricted to the eight sites for which the geographic that the child had ASD. Both NHIS and NSCH ask if the
surveillance area and record source type reviewed were parent/caregiver was ever told that the child has ASD (ever
comparable between the two surveillance years, including five ASD); NSCH also includes a follow-up question asking
sites with sufficient information on intellectual ability among whether the child currently has ASD (current ASD). A
children with ASD for both years. When results were restricted previous analysis showed that 13% of parents who reported ever
to these eight sites, combined ASD prevalence estimates were being told that the child had ASD also reported that the child
similar for 2010 and 2012, including in subgroups defined by did not currently have ASD; most of these parents attributed
sex and race/ethnicity. In the five sites with data on intellectual the lack of a current ASD diagnosis to new information,
ability, the estimated prevalence of ASD with and without suggesting that basing prevalence estimates on ever ASD might
intellectual disability was unchanged in 2012 compared with overestimate prevalence compared with current ASD (25). For
2010. This is notable given that the increase in estimated ASD the 2014 NHIS, the prevalence of parent or caregiver-reported
prevalence that has occurred since 2002 has been accompanied ever ASD was 22.4 per 1,000 children aged 317 years (26).
by a greater increase in ASD without intellectual disability than For the 20112012 NSCH, the prevalence of parent or
ASD with intellectual disability. caregiver-reported current ASD was 20 per 1,000 children
Despite the similar findings when the population was aged 617 years (11). The 2012 ADDM Network overall
restricted to these eight sites for 2010 and 2012 (15.1 and 15.2 ASD prevalence estimate of 14.6 per 1,000 is lower than the
per 1,000, respectively), there were significant differences in overall estimates reported in these surveys; however, differences
ASD prevalence estimates between 2010 and 2012 for three in the sample population and methodology should be taken
of these sites. Significantly increased ASD prevalence estimates into account when comparing results for these three studies.
were observed in New Jersey (12%) and Wisconsin (16%). In The 20112012 NSCH included children aged 617 years;
Missouri, estimated ASD prevalence decreased significantly, when further stratified by age, ASD prevalence was 18.2 per
by 19%, and at the remaining five sites (Arizona, Colorado, 1,000 children aged 69 years and 23.9 per 1,000 children
Georgia, North Carolina, and Utah), ASD prevalence estimates aged 1013 years. Although the difference in ASD prevalence
did not change. The factors underlying the prevalence estimate between these two age groups in the NSCH was not statistically
changes at individual ADDM Network sites are not clear. The significant, the estimate for children aged 69 years (18.2 per
two sites with the greatest change from 2010 to 2012 (Missouri 1,000) is closer to the 2012 ADDM Network overall ASD
and Wisconsin) both reviewed only health source type records prevalence estimate for children aged 8 years (14.6 per 1,000)
for 2010 and 2012. The ability to obtain a comprehensive and similar to the estimate for the 2012 ADDM Network sites
developmental evaluation through the health care system might that reviewed education and health care records (17.1 per 1,000).
be subject to greater local variation compared with evaluations The ASD prevalence estimate from the 2007 NSCH (11.6
performed through the education system because of changes in per 1,000 children aged 617 years) (13) was similar to 2008
the number and availability of providers, changes in insurance ADDM Network prevalence estimate (11.3 per 1,000 children
coverage policies, or other factors. In addition, changes in aged 8 years) (17). Taken as a whole, studies using different
record retention associated with migration to electronic health methodologies and in different populations have reported
records could limit the availability of historical evaluations at converging estimates for ASD prevalence in the United States.
some sources. The wide range of ASD prevalence estimates Future studies by the ADDM Network will incorporate DSM-5
reported by sites participating in the 2012 ADDM Network diagnostic criteria, and ongoing ADDM Network surveillance
coupled with the prevalence estimate increases at some sites will provide information regarding ASD prevalence trends using
suggest the need for caution in interpreting the similarity DSM-IV-TR and DSM-5 diagnostic criteria.
of overall estimated ASD prevalence between 2010 and Consistent with previous years of ADDM Network surveillance
2012. Data from additional surveillance years are needed to (1620), the overall male-to-female ASD prevalence ratio was
understand the trajectory of ASD prevalence. 4.5 in 2012 and has remained largely unchanged across recent
Population-based estimates of ASD prevalence in the United surveillance years: 4.5 in 2004 (18), 2006 (18), and 2010 (16)
States also are reported by two CDC surveys. The National and 4.6 in 2008 (17). A similar male-to-female ASD prevalence
Health Interview Study (NHIS) is a nationally representative ratio was found among school-age children in data from the
household survey, and the National Survey of Childrens Health 20102011 NSCH (11). Observed differences in estimated

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 9
 Surveillance Summaries

ASD prevalence by child characteristics such as sex and race/ children indicates that not all non-Hispanic black and Hispanic
ethnicity might indicate areas where ASD identification children with ASD are being evaluated and/or diagnosed in
is incomplete and can provide data to inform policies and the community, the children who are not identified might not
efforts to improve identification of ASD among subgroups, receive ASD-related services and supports, including school
particularly female and nonwhite children who have historically supports to facilitate educational progress. Targeted strategies
had lower identified prevalence compared with male and non- are needed to increase awareness and identification of ASD in
Hispanic white children. The higher estimated prevalence minority communities.
among boys might result from sex-specific differences in ASD The consistently greater ASD prevalence estimated with
risk (27,28) or differences in identification of girls with ASD data from sites that reviewed education and health source
arising from less well-recognized symptom profiles (29), or type records underscores the role that public schools play
both. The lower male-to-female prevalence ratio for ASD with in the equitable provision of comprehensive evaluations
intellectual disability (PR: 3.7) compared with ASD without to children with developmental concerns. The Individuals
intellectual disability (PR: 5.1) is consistent with data from with Disabilities Education Act mandates that states and
previous ADDM Network surveillance years. Continued school districts identify, locate, and evaluate all children
attention should be paid to ensuring that all children with with disabilities at no cost to the family, so comprehensive
ASD are identified, regardless of functional status. evaluations provided through school systems might be
Results from ADDM Network ASD surveillance in 2012 more accessible and affordable compared with evaluations
continue to indicate disparities in estimated ASD prevalence performed through the health care system. However, results
by race/ethnicity. Across all sites, estimated ASD prevalence from these evaluations might not be reported to the health
among non-Hispanic white children was 20% higher compared care provider or included in the health care provider records.
with non-Hispanic black children, 40% higher compared with Parents and caregivers should be encouraged to share the
Asian/Pacific Islander children, and 50% higher compared results of comprehensive evaluations performed through the
with Hispanic children. In addition, a lower percentage of school system with the childs health care provider to improve
non-Hispanic black and Hispanic children had an earliest continuity of care and ensure that the health care provider can
comprehensive evaluation by age 36 months compared with make recommendations that are based on the childs needs.
non-Hispanic white children. Observed prevalence differences The early identification of ASD is a priority of the American
by race/ethnicity might reflect differences in awareness of ASD Academy of Pediatrics, which recommends universal ASD
or access to specialty diagnostic services (30). For the Hispanic screening at ages 18 and 24 months, and by the U.S.
population, studies have identified lack of awareness of ASD, Department of Health and Human Services through the
stigma associated with disability, lack of access to health care Healthy People 2020 goal of a 10% increase in the percentage
services due to noncitizenship or low income, and language of children with ASD who receive their first evaluation by age
barriers as factors that might reduce the identification of 36 months (22). ADDM Network data are used to measure the
ASD among Hispanic children (3135). In the 20092010 goal that 47% of children with ASD have a first evaluation by
National Survey of Children with Special Health Care Needs age 36 months; the baseline percentage for this goal is 42.7%,
(NSCSHCN), estimated ASD prevalence was nearly 50% as measured by ADDM Network data in 2006. Lowering the
higher for non-Hispanic white children (15.3 per 1,000) age at first evaluation is important because when impairments
compared with non-Hispanic black children (10.4 per 1,000) are identified through a comprehensive evaluation, referrals
and nearly 300% higher for non-Hispanic white children for specific services can be made, often without a formal
compared with Hispanic children living in households where diagnosis. On the basis of evidence linking early treatment
the primary language was not English (5.2 per 1,000). In to improved outcomes (3639), it is important that children
contrast, estimated ASD prevalence was similar for non- with developmental concerns be evaluated and referred to
Hispanic white children compared with Hispanic children services as soon as possible. In 2010, the percentage of children
living in households where the primary language was English aged 8 years with ASD residing the ADDM catchment area with
(14.3 per 1,000) (32). Language differences could affect the an earliest known comprehensive evaluation by age 36 months
administration and interpretation of developmental screening was 43.8% (16), and the 2012 percentage was similar at 42.8%.
and monitoring, impede communication of parental concerns Although several years remain before determination of whether
about a childs development or a health care providers the goal was achieved, the lack of progress from the baseline
recommendation for further evaluation, and limit access to measured in 2006 through 2012 is disappointing. Of note, the
programs and campaigns aimed at increasing awareness of age cohort represented here was born in 2004 and therefore the
ASD. If lower prevalence in non-Hispanic black and Hispanic findings regarding the percentage of children with an earliest

10 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

known evaluation by age 36 months reflect practices during estimates for the intervening years between decennial census
20042007. Continued surveillance is necessary to monitor counts increases with increasing years beyond the decennial
progress towards the Healthy People 2020 goal, particularly in census (in this case, 2010) (42). Sixth, the analysis of age
light of the 2006 AAP screening recommendations, and to at first comprehensive evaluation was restricted to children
identify factors associated with later age at first evaluation so for whom linkage was made to birth certificates for the
that strategies to improve early referral and evaluation can be state where the ADDM Network site was located in an
developed. ADDM Network surveillance of ASD prevalence attempt to reduce bias resulting from the unavailability
and characteristics among children aged 4 years, which began of early evaluations for children who moved after birth.
in 2010, can help to provide more timely data on early However, a child might have moved out and back into this
identification of children with ASD (40). state between birth and ascertainment, so this restriction
The availability of records containing developmental might not have completely eliminated this potential source
evaluations conducted to determine eligibility for special of bias. Finally, race and ethnicity were defined broadly for
education services as well as those conducted through this surveillance population, and results for a specific race
the health care system in response to concerns about a or ethnic group might not be representative of results for
childs development forms the basis for the public health all children in these groups. In addition, it was not possible
surveillance of ASD conducted by the ADDM Network. to distinguish Hispanic children living in households in
By screening existing records then applying a consistent which the primary language was English from those with a
methodology by trained and research-reliable clinician different primary language.
reviewers to determine case status, the ADDM Network
is able to conduct population-based surveillance of ASD
in a large and diverse population. This methodology was Future Study Directions
validated, compared with direct examination of children, In 2013, revised diagnostic criteria for ASD were published
and the methods were found to result in a prevalence by the American Psychiatric Association in the DSM-5 (5).
estimate that is likely conservative (41). Beginning with the 2014 surveillance year, the ADDM
Network will be able to estimate ASD case status on the basis
of both DSM-5 and DSM-IV-TR. This evaluation is possible
Limitations because of the data collection methods employed since the
The findings in this report are subject to at least seven inception of the ADDM Network, including the abstraction
limitations. First, data were limited to the information of specific behaviors documented in childrens records. This
available in the source records. The amount and quality of unique component of ADDM Network ASD surveillance
the data define the potential to determine whether a child will enable the ADDM Network investigators to evaluate the
meets the ASD surveillance case definition and the extent change in estimated ASD prevalence that might arise from the
to which the characteristics of the identified population change in diagnostic criteria. Previous analyses have suggested
can be described. In particular, data on intellectual ability that fewer children will meet the behavioral criteria of DSM-5
were not available for all children, and the distribution of compared with DSM-IV-TR (43). However, DSM-5 criteria
intellectual ability among the children with these data might include a provision that children with a well-established
not be generalizable to all children with ASD. Second, the diagnosis of one of the three autism spectrum disorder subtypes
types of source records varied across sites, and the inability under DSM-IV-TR criteria are considered to have ASD under
to review education records at some sites might have led to DSM-5 criteria. Therefore, at least for the initial years following
an underestimate of ASD prevalence in those sites. Third, the publication of DSM-5, ASD prevalence estimates that are
education records generally were not available for children based on DSM-5 criteria should include the children with a
attending private school or being home-schooled. Fourth, the DSM-IV-TR-based diagnosis in order to accurately represent
surveillance areas were selected through a competitive process the number of children who are being treated and served
and were not selected to be representative of children aged for ASD by community providers. Because the surveillance
8 years in the United States or the state where the surveillance methodology of the ADDM Network also includes collection
site was located. Fifth, county-level population counts for of information on ASD diagnoses by community providers,
children by sex and race/ethnicity are not available by single future estimates of the prevalence of ASD under DSM-5 will
year of age in nondecennial census years. Population estimates be able to include children who meet DSM-5 criteria by virtue
published by the National Center for Health Statistics are of a past DSM-IV-TR diagnosis as well as those meeting the
used instead. There is evidence that the error in population DSM-5 behavioral criteria.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 11
 Surveillance Summaries

Conclusion Rebecca Harrington, PhD, Johns Hopkins University, Baltimore,

Maryland; Rob Fitzgerald, PhD, Washington University, St. Louis,
Approximately one in 68 children aged 8 years living in Missouri; Josephine Shenouda, MS, Rutgers UniversityNew Jersey
sites participating in the ADDM Network surveillance areas Medical School, Newark; Paula Bell, University of North Carolina,
met the ASD case criteria for the 2012 surveillance year. Chapel Hill; Andrea Boan, PhD, Walter Jenner, MS, Medical
Although the overall prevalence estimate is unchanged from University of South Carolina; Colin Kingsbury, MS, Amanda
surveillance year 2010, prevalence ranged widely across the Bakian, PhD, Amy Henderson, University of Utah, Salt Lake City;
ADDM Network and prevalence increases were reported at Pamela Imm, MS, University of WisconsinMadison. Additional
two sites, suggesting that it is premature to conclude that the assistance was provided by Russell Kirby, PhD, University of South
rising prevalence of ASD observed during the first decade Florida, Tampa; Heather Clayton, PhD, Alyson Goodman, MD,
Lisa Wiggins, PhD, Division of Congenital and Developmental
of the 21st century might be slowing. Ongoing surveillance
Disorders, National Center on Birth Defects and Developmental
of ASD prevalence through the ADDM Network is likely
Disabilities, CDC; project staff including data abstractors, clinician
to provide the most accurate means to monitor trends in reviewers, epidemiologists, and data management/programming
ASD prevalence over time, including those that are related to support. Ongoing ADDM Network support was provided by Victoria
changes in the diagnostic criteria for ASD. ASD surveillance Wright, Tineka Yowe-Conley, National Center on Birth Defects and
informs providers, particularly public schools, of upcoming Developmental Disabilities, CDC.
service needs, and provides feedback on progress made
toward early identification goals. The ADDM Network will References
continue to track age at first comprehensive evaluation to 1. American Psychiatric Association. Diagnostic and statistical manual of
mental disorders. 4th ed. Text revision. Washington, DC: American
monitor progress toward the Healthy People 2020 goal of Psychiatric Association; 2000.
increasing the percentage of children with ASD who receive a 2. Gillberg C, Wing L. Autism: not an extremely rare disorder. Acta Psychiatr
first evaluation by age 36 months. Estimated ASD prevalence Scand 1999;99:399406. http://dx.doi.org/10.1111/j.1600-0447.1999.
was substantially lower among Hispanic and non-Hispanic tb00984.x
3. American Psychiatric Association. Diagnostic and statistical manual
black children compared with non-Hispanic white children. of mental disorders. 3rd ed. Washington, DC: American Psychiatric
In addition, non-Hispanic black and Hispanic children were Association; 1980.
less likely to have a first evaluation by age 36 months and 4. American Psychiatric Association. Diagnostic and statistical manual
of mental disorders. 4th ed. Washington, DC: American Psychiatric
Hispanic children were less likely to have a previous ASD Association; 1994.
diagnosis or classification. This finding suggests that a number 5. American Psychiatric Association. Diagnostic and statistical manual
of nonwhite children with ASD are not being identified and of mental disorders. 5th ed. Arlington, VA: American Psychiatric
Association; 2013.
evaluated, and for those children who are evaluated, a later age 6. Burd L, Fisher W, Kerbeshian J. A prevalence study of pervasive developmental
at the first comprehensive evaluation likely delays initiation of disorders in North Dakota. J Am Acad Child Adolesc Psychiatry 1987;26:700
services for these children. No intervention has been shown to 3. http://dx.doi.org/10.1097/00004583-198709000-00014
reduce the prevalence of ASD; however, early treatment might 7. Ritvo ER, Freeman BJ, Pingree C, et al. The UCLA-University of
Utah epidemiologic survey of autism: prevalence. Am J Psychiatry
maximize the ability of children to function and participate in 1989;146:1949. http://dx.doi.org/10.1176/ajp.146.2.194
their community. Initiation of school-based services prior to 8. Croen LA, Grether JK, Hoogstrate J, Selvin S. The changing prevalence
formal school entry might help to facilitate optimal educational of autism in California. J Autism Dev Disord 2002;32:20715. http://
dx.doi.org/10.1023/A:1015453830880
progress. Continued efforts should be made to promote early 9. Newschaffer CJ, Falb MD, Gurney JG. National autism prevalence trends
identification of all children with ASD so that interventions from United States special education data. Pediatrics 2005;115:e27782.
can be initiated at the youngest age possible. http://dx.doi.org/10.1542/peds.2004-1958
10. California Department of Developmental Services. Autistic spectrum
disorders: changes in the California caseload, an update: 1999 through
Acknowledgments 2002. Sacramento, CA: California Health and Human Services Agency,
Data collection was coordinated at each site by ADDM Department of Developmental Services; 2003.
Network project coordinators: Eric Lott, University of Alabama 11. Blumberg SJ, Bramlett MD, Kogan MD, Schieve LA, Jones JR, Lu MC.
Changes in prevalence of parent-reported autism spectrum disorder in
at Birmingham; Kristen Clancy Mancilla, University of Arizona, school-aged U.S. children: 2007 to 20112012. Natl Health Stat Report
Tucson; Allison Hudson, University of Arkansas for Medical Sciences, 2013;65:111.
Little Rock; Kelly Kast, MSPH, Colorado Department of Public 12. Boyle CA, Boulet S, Schieve LA, et al. Trends in the prevalence of
Health and Environment, Denver; Kwinettaion Jolly, MS, Research developmental disabilities in US children, 19972008. Pediatrics
Triangle Institute, Atlanta, Georgia; Anita Washington, MPH, Ann 2011;127:103442. http://dx.doi.org/10.1542/peds.2010-2989
13. Kogan MD, Blumberg SJ, Schieve LA, et al. Prevalence of parent-
Ussery-Hall, MPH, Lisa Barritt, Gal Frenkel, MPH, Division of reported diagnosis of autism spectrum disorder among children in the
Birth Defects and Developmental Disabilities, National Center on US, 2007. Pediatrics 2009;124:1395403. http://dx.doi.org/10.1542/
Birth Defects and Developmental Disabilities, CDC; Ann Chang, peds.2009-1522

12 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

14. Schieve LA, Rice C, Yeargin-Allsopp M, et al. Parent-reported prevalence 29. Andersson GW, Gillberg C, Miniscalco C. Pre-school children with
of autism spectrum disorders in US-born children: an assessment of suspected autism spectrum disorders: do girls and boys have the same
changes within birth cohorts from the 2003 to the 2007 National Survey profiles? Res Dev Disabil 2013;34:41322. http://dx.doi.org/10.1016/j.
of Childrens Health. Matern Child Health J 2012;16(Suppl 1):S1517. ridd.2012.08.025
http://dx.doi.org/10.1007/s10995-012-1004-0 30. Jarquin VG, Wiggins LD, Schieve LA, Van Naarden-Braun K. Racial
15. Yeargin-Allsopp M, Rice C, Karapurkar T, Doernberg N, Boyle C, disparities in community identification of autism spectrum disorders over
Murphy C. Prevalence of autism in a US metropolitan area. JAMA time; Metropolitan Atlanta, Georgia, 20002006. J Dev Behav Pediatr
2003;289:4955. http://dx.doi.org/10.1001/jama.289.1.49 2011;32:17987. http://dx.doi.org/10.1097/DBP.0b013e31820b4260
16. Autism and Developmental Disabilities Monitoring Network Surveillance 31. Flores G, Abreu M, Tomany-Korman SC. Why are Latinos the most
Year 2010 Principal Investigators. Prevalence of autism spectrum disorder uninsured racial/ethnic group of US children? A community-based
among children aged 8 yearsAutism and Developmental Disabilities study of risk factors for and consequences of being an uninsured
Monitoring Network, 11 sites, United States, 2010. MMWR Surveill Latino child. Pediatrics 2006;118:e73040. http://dx.doi.org/10.1542/
Summ 2014;63(No. SS-2). peds.2005-2599
17. Autism and Developmental Disabilities Monitoring Network 32. Jo H, Schieve LA, Rice CE, et al. Age at autism spectrum disorder (ASD)
Surveillance Year 2008 Principal Investigators. Prevalence of autism diagnosis by race, ethnicity, and primary household language among
spectrum disordersAutism and Developmental Disabilities Monitoring children with special health care needs, United States, 20092010.
Network, 14 sites, United States, 2008. MMWR Surveill Summ Matern Child Health J 2015;19:168797. http://dx.doi.org/10.1007/
2012;61(No. SS-3):119. s10995-015-1683-4
18. Autism and Developmental Disabilities Monitoring Network 33. Zuckerman KE, Mattox K, Donelan K, Batbayar O, Baghaee A,
Surveillance Year 2006 Principal Investigators. Prevalence of autism Bethell C. Pediatrician identification of Latino children at risk for
spectrum disordersAutism and Developmental Disabilities Monitoring autism spectrum disorder. Pediatrics 2013;132:44553. http://dx.doi.
Network, United States, 2006. MMWR Surveill Summ 2009;58(No. org/10.1542/peds.2013-0383
SS-10):120. 34. Zuckerman KE, Sinche B, Cobian M, et al. Conceptualization of autism
19. Autism and Developmental Disabilities Monitoring Network in the Latino community and its relationship with early diagnosis.
Surveillance Year 2002 Principal Investigators. Prevalence of autism J Dev Behav Pediatr 2014;35:52232. http://dx.doi.org/10.1097/
spectrum disordersAutism and Developmental Disabilities Monitoring DBP.0000000000000091
Network, 14 sites, United States, 2002. MMWR Surveill Summ 35. Zuckerman KE, Sinche B, Mejia A, Cobian M, Becker T, Nicolaidis C.
2007;56(No. SS-1):1228. Latino parents perspectives on barriers to autism diagnosis. Acad Pediatr
20. Autism and Developmental Disabilities Monitoring Network 2014;14:3018. http://dx.doi.org/10.1016/j.acap.2013.12.004
Surveillance Year 2000 Principal Investigators. Prevalence of autism 36. Dawson G, Rogers S, Munson J, et al. Randomized, controlled trial of
spectrum disordersAutism and Developmental Disabilities Monitoring an intervention for toddlers with autism: the Early Start Denver Model.
Network, six sites, United States, 2000. MMWR Surveill Summ Pediatrics 2010;125:e1723. http://dx.doi.org/10.1542/peds.2009-0958
2007;56(No. SS-1):111. 37. Eapen V, Crnec R, Walter A. Clinical outcomes of an early intervention
21. Johnson CP, Myers SM; American Academy of Pediatrics Council on program for preschool children with Autism Spectrum Disorder in
Children with Disabilities. Identification and evaluation of children a community group setting. BMC Pediatr 2013;13:3. http://dx.doi.
with autism spectrum disorders. Pediatrics 2007;120:1183215. http:// org/10.1186/1471-2431-13-3
dx.doi.org/10.1542/peds.2007-2361 38. Reichow B, Barton EE, Boyd BA, Hume K. Early intensive behavioral
22. US Department of Health and Human Services. Healthy people 2020. intervention (EIBI) for young children with autism spectrum disorders
Washington, DC: US Department of Health and Human Services; 2010. (ASD). Cochrane Database Syst Rev 2012;10:CD009260.
http://www.healthypeople.gov 39. Rogers SJ, Estes A, Lord C, et al. Effects of a brief Early Start Denver
23. Yeargin-Allsopp M, Murphy CC, Oakley GP, Sikes RK. A multiple- model (ESDM)-based parent intervention on toddlers at risk for autism
source method for studying the prevalence of developmental disabilities spectrum disorders: a randomized controlled trial. J Am Acad Child
in children: the Metropolitan Atlanta Developmental Disabilities Study. Adolesc Psychiatry 2012;51:105265. http://dx.doi.org/10.1016/j.
Pediatrics 1992;89:62430. jaac.2012.08.003
24. US Department of Health and Human Services. Code of Federal 40. Christensen DL, Bilder DA, Zahorodny W, et al. Prevalence and
Regulations. Title 45. Public Welfare CFR 46. Washington, DC: US characteristics of autism spectrum disorder among 4-year-old children
Department of Health and Human Services; 2010. http://www.hhs.gov/ in the Autism and Developmental Disabilities Monitoring Network.
ohrp/humansubjects/guidance/45cfr46.html J Dev Behav Pediatr 2016;37:18. http://dx.doi.org/10.1097/
25. Blumberg SJ, Zablotsky B, Avila RM, Colpe LJ, Pringle BA, Kogan DBP.0000000000000235
MD. Diagnosis lost: differences between children who had and 41. Nonkin Avchen R, Wiggins LD, Devine O, et al. Evaluation of a records-
who currently have an autism spectrum disorder diagnosis. Autism review surveillance system used to determine the prevalence of autism
2015:1362361315607724. spectrum disorders. J Autism Dev Disord 2011;41:22736. http://
26. Zablotsky B, Black LI, Maenner MJ, Schieve LA, Blumberg SJ. Estimated dx.doi.org/10.1007/s10803-010-1050-7
prevalence of autism and other developmental disabilities following 42. Lazarus C, Autry A, Baio J, Avchen RN, Van Naarden Braun K. Impact
questionnaire changes in the 2014 National Health Interview Study. of postcensal versus intercensal population estimates on prevalence of
Natl Health Stat Report 2015;87:120. selected developmental disabilitiesmetropolitan Atlanta, Georgia,
27. Lai MC, Lombardo MV, Suckling J, et al.; MRC AIMS Consortium. 19911996. Am J Ment Retard 2007;112:4626. http://dx.doi.
Biological sex affects the neurobiology of autism. Brain 2013;136:2799 org/10.1352/0895-8017(2007)112[462:IOPVIP]2.0.CO;2
815. http://dx.doi.org/10.1093/brain/awt216 43. Maenner MJ, Rice CE, Arneson CL, et al. Potential impact of
28. Werling DM, Geschwind DH. Understanding sex bias in autism DSM-5 criteria on autism spectrum disorder prevalence estimates.
spectrum disorder. Proc Natl Acad Sci U S A 2013;110:48689. http:// JAMA Psychiatry 2014;71:292300. http://dx.doi.org/10.1001/
dx.doi.org/10.1073/pnas.1301602110 jamapsychiatry.2013.3893

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 13
 Surveillance Summaries

TABLE 1. Number* and percentage of children aged 8 years, by race/ethnicity and site Autism and Developmental Disabilities Monitoring
Network, 11 sites, United States, 2012
White, Black, API, AI/AN,
Total non-Hispanic non-Hispanic Hispanic non-Hispanic non-Hispanic
Site Site institution Surveillance area No. No. (%) No. (%) No. (%) No. (%) No. (%)
Arizona University of Arizona Part of 1 county in 32,615 15,525 (47.6) 1,856 (5.7) 13,180 (40.4) 1,276 (3.9) 778 (2.4)
metropolitan
Phoenix
Arkansas University of Arkansas 16 counties in 14,153 9,083 (64.2) 3,739 (26.4) 1,025 (7.2) 226 (1.6) 80 (0.6)
for Medical Sciences Arkansas
Colorado Colorado Department 7 counties including 40,538 22,370 (55.2) 2,469 (6.1) 13,448 (33.2) 2,029 (5.0) 222 (0.5)
of Public Health and metropolitan
Environment Denver
Georgia CDC 5 counties including 49,720 16,451 (33.1) 20,556 (41.3) 9,019 (18.1) 3,588 (7.2) 106 (0.2)
metropolitan
Atlanta
Maryland Johns Hopkins University 1 county in 9,577 5,019 (52.4) 3,171 (33.1) 656 (6.9) 696 (7.3) 35 (0.4)
suburban
Baltimore
Maryland Johns Hopkins University 5 counties in 18,154 12,293 (67.7) 3,042 (16.8) 1,384 (7.6) 1,383 (7.6) 52 (0.3)
suburban
Baltimore
Missouri Washington University 5 counties including 25,870 17,211 (66.5) 6,516 (25.2) 1,109 (4.3) 970 (3.7) 64 (0.2)
St. Louis metropolitan
St. Louis
New Jersey Rutgers UniversityNew 4 counties including 32,581 13,829 (42.4) 7,100 (21.8) 9,787 (30.0) 1,781 (5.5) 84 (0.3)
Jersey Medical School metropolitan
Newark
North Carolina University of North 11 counties in 38,913 20,789 (53.4) 9,544 (24.5) 6,517 (16.7) 1,906 (4.9) 157 (0.4)
CarolinaChapel Hill central North
Carolina
South Carolina Medical University of 23 counties in coastal 24,356 12,485 (51.3) 9,404 (38.6) 1,964 (8.1) 387 (1.6) 116 (0.5)
South Carolina and Pee Dee regions
Utah University of Utah 3 counties in 24,945 18,217 (73.0) 568 (2.3) 4,851 (19.4) 1,151 (4.6) 158 (0.6)
northern Utah
Wisconsin University of 10 counties in 35,556 21,758 (61.2) 6,342 (17.8) 5,915 (16.6) 1,392 (3.9) 149 (0.4)
WisconsinMadison southeastern
Wisconsin

Total 346,978 185,030 (53.3) 74,307 (21.4) 68,885 (19.9) 16,785 (4.8) 2,001 (0.6)
Abbreviations: AI/AN = American Indian/Alaska Native; API = Asian/Pacific Islander.
* Total numbers of children aged 8 years in each surveillance area were obtained from CDCs July 1, 2012 bridged-race population estimates.
Education and health records review area.
Health records only review area.

14 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Please note: An erratum has been published for this issue. To view the erratum, please click here.
Surveillance Summaries

TABLE 2. Estimated prevalence* of autism spectrum disorder among 1,000 children aged 8 years, by sex Autism and Developmental
Disabilities Monitoring Network, 11 sites, United States, 2012
Sex
Total Male Female
Total no. Male-to-female
Site Total with ASD Prevalence (95% CI) Prevalence (95% CI) Prevalence (95% CI) prevalence ratio
Arizona 32,615 494 15.2 (13.916.5) 24.2 (22.026.7) 5.7 (4.77.0) 4.2 (3.45.3)
Arkansas 14,153 170 12.0 (10.314.0) 19.2 (16.322.7) 4.6 (3.26.5) 4.2 (2.96.2)
Colorado 40,538 436 10.8 (9.811.8) 17.1 (15.419.0) 4.2 (3.45.2) 4.1 (3.25.2)
Georgia 49,720 771 15.5 (14.416.6) 25.6 (23.727.6) 5.2 (4.36.1) 4.9 (4.16.0)
Maryland 9,577 174 18.2 (15.721.1) 29.4 (25.034.6) 6.2 (4.39.0) 4.7 (3.27.0)
Maryland 18,154 148 8.2 (6.99.6) 13.9 (11.716.5) 2.2 (1.43.5) 6.3 (3.910.0)
Missouri 25,870 297 11.5 (10.212.9) 18.9 (16.721.4) 3.8 (2.85.0) 5.0 (3.76.8)
New Jersey 32,581 800 24.6 (22.926.3) 39.1 (36.242.2) 9.3 (7.910.9) 4.2 (3.55.0)
North Carolina 38,913 656 16.9 (15.6, 18.2) 27.5 (25.3, 29.9) 6.0 (5.0, 7.2) 4.6 (3.85.6)
South Carolina 24,356 302 12.4 (11.113.9) 19.9 (17.622.5) 4.6 (3.56.0) 4.3 (3.25.8)
Utah 24,945 431 17.3 (15.719.0) 27.7 (24.930.7) 6.4 (5.18.0) 4.3 (3.45.5)
Wisconsin 35,556 384 10.8 (9.811.9) 17.2 (15.419.2) 4.1 (3.25.2) 4.2 (3.25.4)
Total 346,978 5,063 14.6 (14.215.0) 23.6 (22.924.3) 5.3 (4.9, 5.6) 4.5 (4.24.8)
Abbreviations: ASD = autism spectrum disorder; CI = confidence interval; E+H = education plus health.
* Per 1,000 children aged 8 years.
All sites identified significantly higher prevalence among males compared with females (Pearson chi-square, p<0.01).
Education and health records review area.
Health records only review area.

TABLE 3. Estimated prevalence* of autism spectrum disorder among 1,000 children aged 8 years, by race/ethnicity Autism and Developmental
Disabilities Monitoring Network, 11 sites, United States, 2012
Race/Ethnicity Prevalence ratio
White, non-Hispanic Black, non-Hispanic Hispanic API, non-Hispanic
White-to- White-to- Black-to-
Site Prevalence (95% CI) Prevalence (95% CI) Prevalence (95% CI) Prevalence (95% CI) black Hispanic Hispanic
Arizona 16.9 (15.019.1) 19.4 (14.026.9) 11.3 (9.613.3) 13.3 (8.321.4) 0.9 1.5 1.7
Arkansas 12.8 (10.415.3) 9.9 (7.213.7) 1.3
Colorado 12.2 (10.913.8) 10.5 (7.215.1) 6.5 (5.28.0) 8.4 (5.213.5) 1.2 1.9 1.6
Georgia 17.6 (15.619.7) 13.4 (11.915.1) 11.5 (9.514.0) 13.4 (10.117.8) 1.3 1.5 1.2
Maryland 18.5 (15.122.7) 18.6 (14.424.0) 12.2 (6.124.4) 10.1 (4.8, 21.1) 1.0 1.5 1.5
Maryland** 8.6 (7.110.4) 6.9 (4.510.6) 5.8 (2.911.6) 1.2 1.5 1.2
Missouri 12.0 (10.513.8) 9.0 (7.011.7) 8.1 (4.215.6) 1.3 1.5 1.1
New Jersey 26.6 (24.029.5) 23.7 (20.327.5) 17.6 (15.120.4) 21.9 (16.030.0) 1.1 1.5 1.3
North Carolina 18.9 (17.1, 20.8) 15.5 (13.2, 18.2) 9.1 ((7.0, 11.7) 18.4 (13.2, 25.6) 1.2 2.1 1.7
South Carolina 12.7 (10.814.8) 10.6 (8.712.9) 6.6 (3.811.4) 1.2 1.9 1.6
Utah 17.7 (15.819.7) 12.3 (5.925.8) 13.2 (10.316.9) 5.2 (2.311.6) 1.4 1.3 0.9
Wisconsin 12.0 (10.613.5) 5.8 (4.28.0) 7.4 (5.510.0) 3.6 (1.58.6) 2.1 1.6 0.8
Total 15.5 (14.916.1) 13.2 (12.414.0) 10.1 (9.410.9) 11.3 (9.8, 13.0) 1.2 1.5 1.3
Abbreviations: API = Asian/Pacific Islander; ASD = autism spectrum disorder; CI = confidence interval; E+H = education plus health.
* Per 1,000 children aged 8 years.
Prevalence ratio significant at p<0.05.
Prevalence not calculated when n<5.
Education and health records review area.
** Health records only review area.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 15
 Surveillance Summaries

TABLE 4. Number and percentage of children aged 8 years* identified with autism spectrum disorder who received a comprehensive evaluation
by a qualified professional at age 36 months, 3748 months, or >48 months, and those with a mention of a developmental concern by age
36 months Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2012
Earliest age when child received a comprehensive evaluation
Mention of a developmental
36 mos 3748 mos >48 mos concern by age 36 months
Site No. (%) No. (%) No. (%) No. (%)
Arizona 149 (39.2) 70 (18.4) 161 (42.4) 341 (89.7)
Arkansas 33 (24.2) 38 (27.9) 65 (47.8) 119 (87.5)
Colorado 131 (40.6) 60 (18.6) 132 (40.9) 278 (86.1)
Georgia 222 (41.1) 113 (20.9) 205 (38.0) 473 (87.6)
Maryland 83 (55.0) 27 (17.9) 41 (27.2) 143 (94.7)
Maryland 34 (31.2) 22 (20.2) 53 (48.6) 101 (92.7)
Missouri 103 (40.6) 34 (13.4) 117 (46.1) 210 (82.7)
New Jersey 277 (42.9) 137 (21.1) 233 (36.0) 527 (81.5)
North Carolina 288 (59.8) 71 (14.7) 123 (25.5) 444 (92.1)
South Carolina 79 (38.5) 52 (25.4) 74 (36.1) 189 (92.2)
Utah 119 (37.5) 66 (20.8) 132 (41.6) 258 (81.4)
Wisconsin 133 (41.8) 63 (19.8) 122 (38.4) 286 (89.9)
Total 1,662 (42.8) 756 (19.5) 1,463 (37.7) 3,386 (87.2)
* Includes 3,881 children identified with autism spectrum disorder who were linked to an in-state birth certificate.
Education and health records review area.
Health records only review area.

TABLE 5. Median age of earliest known autism spectrum disorder diagnosis and number and proportion within each diagnostic subtype
Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2012
ASD subtype
Autistic disorder ASD-NOS/PDD-NOS Asperger disorder Any ASD subtype
Median age Median age Median age Median age
Site (mos) No. (%) (mos) No. (%) (mos) No. (%) (mos) No. (%)
Arizona 50.0 254 (74.5) 64.0 72 (21.1) 77.0 15 (4.4) 55.0 341 (69.0)
Arkansas 53.0 97 (72.9) 60.0 20 (15.0) 77.5 16 (12.0) 60.0 133 (78.2)
Colorado 48.0 184 (66.2) 59.0 55 (19.8) 80.0 39 (14.0) 55.0 278 (63.8)
Georgia 47.5 262 (48.0) 51.0 231 (42.3) 71.0 53 (9.7) 51.0 546 (70.8)
Maryland* 41.0 56 (40.6) 48.0 79 (57.2) 44.0 3 (2.2) 45.5 138 (79.3)
Maryland 46.5 44 (33.5) 44.5 72 (58.1) 44.0 8 (6.4) 48.0 124 (83.8)
Missouri 50.0 67 (26.1) 51.0 145 (56.4) 78.0 45 (17.5) 58.0 257 (86.5)
New Jersey 44.5 192 (29.7) 43.0 378 (58.4) 74.0 77 (11.9) 47.0 647 (80.9)
North Carolina 37.0 207 (53.6) 55.5 156 (40.4) 72.0 23 (6.0) 48.0 386 (58.8)
South Carolina 45.0 143 (65.0) 58.0 70 (31.8) 74.0 7 (3.2) 48.0 220 (72.8)
Utah 45.0 114 (31.7) 48.0 178 (49.4) 63.5 68 (18.9) 50.0 360 (83.5)
Wisconsin 45.5 106 (34.8) 49.0 173 (56.7) 74.0 26 (8.5) 50.0 305 (79.4)
Total 46.0 1,726 (46.2) 49.0 1,629 (43.6) 74.0 380 (10.2) 50.0 3,735 (73.8)
Abbreviations: ASD-NOS = autism spectrum disordernot otherwise specified; PDD-NOS = pervasive developmental disordernot otherwise specified.
* Education and health records review area.
Health records only review area.

16 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

TABLE 6. Number and percentage of children aged 8 years identified with autism spectrum disorder with available special education records, by
primary special education eligibility category* Autism and Developmental Disabilities Monitoring Network, seven sites, United States, 2012
Primary special education eligibility
category Arizona Georgia Maryland New Jersey North Carolina South Carolina Utah
Autism (%) 61.4 58.1 70.2 56.2 69.0 61.0 53.4
Emotional disturbance (%) 4.6 1.1 1.6 0.7 1.4 0 1.7
Specific learning disability (%) 6.8 3.1 8.1 4.6 7.9 4.9 8.5
Speech or language impairment (%) 6.8 1.6 0 10.3 2.8 2.7 18.6
Hearing or visual impairment (%) 0 0.6 0 0.3 0 0 0
Health or physical disability (%) 4.4 3.7 12.1 19.1 10.4 5.4 10.6
Multiple disabilities (%) 2.0 0 4.0 6.4 0.8 0 0
Intellectual disability (%) 8.2 2.6 3.2 1.0 3.8 3.6 4.2
Developmental delay/preschool (%) 5.7 29.2 0.8 0.4 3.8 20.2 3.0
Other (%) 0 0 0 0.9 0.2 2.2 0
Total no. of ASD cases 494 771 174 800 656 302 431

Total no. (%) of ASD cases with special 454 (91.9) 621 (80.5) 124 (71.3) 698 (87.3) 507 (77.3) 223 (73.8) 236 (54.8)
education records
Abbreviation: ASD = autism spectrum disorder.
* Some state-specific categories were recoded or combined to match current US Department of Education categories.
Education and health records review area.

TABLE 7. Comparison of autism spectrum disorder prevalence among sites with comparable surveillance area in 2010 and 2012, by record
source type, sex, and race/ethnicity, Autism and Developmental Disabilities Monitoring Network, eight sites, United States
2010 2012
2012-to-2010 prevalence ratio
Characteristic Prevalence (95% CI) Prevalence (95% CI) (95% CI)
Record source
E+H areas* 17.5 (16.918.2) 17.6 (17.018.3) 1.01 (0.961.06)
HO areas 10.8 (10.111.4) 11.0 (10.41.6) 1.02 (0.941.11)
E+H-to-HO prevalence ratio 1.6 (1.51.7) 1.6 (1.51.7)
Site
Arizona 15.7 (14.417.1) 15.2 (13.916.5) 0.97 (0.851.10)
Colorado 9.9 (9.010.9) 10.8 (9.819.0) 1.09 (0.951.25)
Georgia 15.5 (14.316.8) 15.5 (14.416.6) 1.00 (0.901.10)
Missouri 14.2 (12.815.7) 11.5 (10.212.9) 0.81 (0.700.95)
New Jersey 21.9 (20.423.6) 24.6 (22.926.3) 1.12 (1.011.24)
North Carolina 17.3 (16.118.7) 16.9 (15.618.2) 0.97 (0.901.08)
Utah 18.6 (16.920.4) 17.3 (15.719.0) 0.93 (0.811.06)
Wisconsin 9.3 (8.310.3) 10.8 (9.811.9) 1.16 (1.011.35)
Sex
Male 24.4 (23.625.2) 24.5 (23.725.4) 1.02 (0.961.05)
Female 5.4 (5.05.8) 5.5 (5.15.9) 1.02 (0.921.13)
Male-to-female prevalence ratio 4.5 (4.25.0) 4.4 (4.14.8)
Race/Ethnicity
White, non-Hispanic 16.2 (15.516.8) 16.3 (15.717.0) 1.02 (0.961.07)
Black, non-Hispanic 12.9 (12.013.9) 13.9 (12.914.9) 1.07 (0.971.19)
Hispanic 11.2 (10.412.1) 10.4 (9.6, 11.2) 0.93 (0.831.03)
White-to-black prevalence ratio 1.3 (1.21.4) 1.2 (1.11.3)
White-to-Hispanic prevalence ratio 1.4 (1.31.6) 1.6 (1.41.7)
Black-to-Hispanic prevalence ratio 1.2 (1.01.3) 1.3 (1.21.5)
Total 15.1 (14.615.5) 15.2 (14.715.6) 1.01 (0.971.05)
Abbreviations: CI = confidence interval; E+H = education and health records review; HO = health records only review.
* Sites reviewing education and health records: Arizona, Georgia, New Jersey, North Carolina, and Utah.
Sites reviewing health records only: Colorado, Missouri, and Wisconsin.
Ratios of prevalence ratios were not calculated.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 17
 Surveillance Summaries

TABLE 8. Comparison of autism spectrum disorder prevalence among sites with comparable surveillance areas, by sex, race/ethnicity, and
most recent score on intelligence quotient test, Autism and Developmental Disabilities Monitoring Network, five sites,* United States,
2010 and 2012
2010 2012
Prevalence ratio 2012 to 2010
Characteristic Prevalence (95% CI) Prevalence (95% CI) (95% CI)
Sex
Male 28.5 (27.429.6) 28.5 (27.429.6) 1.00 (0.951.06)
Female 6.2 (5.76.8) 6.3 (5.86.9) 1.02 (0.911.15)
Male-to-female prevalence ratio 4.6 (4.25.0) 4.5 (4.14.9)
Race/Ethnicity
White, non-Hispanic 19.4 (18.520.4) 19.3 (18.420.3) 1.00 (0.931.07)
Black, non-Hispanic 15.2 (14.016.4) 16.2 (15.017.5) 1.06 (0.951.19)
Hispanic 13.5 (12.414.6) 12.1 (11.113.2) 0.90 (0.801.01)
White-to-black prevalence ratio 1.3 (1.21.4) 1.2 (1.11.3)
White-to-Hispanic prevalence ratio 1.4 (1.31.6) 1.6 (1.51.8)
Black-to-Hispanic prevalence ratio 1.1 (1.01.3) 1.4 (1.21.5)
IQ
70 4.6 (4.34.9) 4.3 (4.04.7) 0.94 (0.861.04)
>70 10.6 (10.111.1) 10.0 (9.610.5) 0.95 (0.891.01)
Unknown 2.4 (2.12.6) 3.3 (3.03.5) 1.39 (1.221.57)
>70-to-70 prevalence ratio 2.3 (2.12.5) 2.3 (2.12.5)
Total 17.5 (16.918.2) 17.6 (17.018.3) 1.01 (0.961.06)
Abbreviations: CI = confidence interval; IQ = intelligence quotient.
* Arizona, Georgia, New Jersey, North Carolina, and Utah.
Ratios of prevalence ratios were not calculated.

18 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

FIGURE 1. Estimated prevalence* of autism spectrum disorder among children aged 8 years Autism and Developmental Disabilities Monitoring
Network, 11 sites, United States, 2012
30

Sites accessing data from education and health records

Sites accessing data from health records only
25
Prevalence for all sites combined

20
Prevalence

15

10

0
Maryland Wisconsin Colorado Missouri Arkansas South Arizona Georgia North Utah Maryland New Jersey
(HO) Carolina Carolina (E+H)
ADDM site
Abbreviations: ADDM = Autism and Developmental Disabilities Monitoring Network; E+H = education and health records; HO = health records only.
* Cases per 1,000 children aged 8 years. Bars represent 95% confidence intervals.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 19
 Surveillance Summaries

FIGURE 2. Scores of most recent intelligence quotient tests for children identified with autism spectrum disorder for whom test data were
available Autism and Developmental Disabilities Monitoring Network, nine sites,* United States, 2012

Intellectually disabled range (IQ70) Borderline range (IQ = 7185) Average or above average (IQ>85)
100

90

80

70
Percentage

60

50

40

30

20

10

0
M F M F M F M F M F M F M F M F M F M F Total
Arizona Arkansas Colorado Georgia Maryland New Jersey North South Utah All sites combined
(E+H) Carolina Carolina
ADDM site

Abbreviations: ADDM = Autism and Developmental Disabilities Monitoring Network; ASD = autism spectrum disorder; E+H = education and health records;
IQ = intelligence quotient.
* Includes sites having information on intellectual ability available for 70% of children who met the ASD case definition (N = 3,390).

20 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

FIGURE 3. Estimated prevalence* of autism spectrum disorder among children aged 8 years, by most recent intelligence quotient score and
by site Autism and Developmental Disabilities Monitoring Network, nine sites, United States, 2012

30

IQ>70
IQ unknown
25
IQ70

20
Prevalence

15

10

0
Arizona Arkansas Colorado Georgia Maryland New Jersey North South Utah All sites
(E+H) Carolina Carolina combined
ADDM site
Abbreviations: ADDM = Autism and Developmental Disabilities Monitoring Network; ASD = autism spectrum disorder; E+H = education and health records;
IQ = intelligence quotient.
* Cases per 1,000 children aged 8 years.
Includes sites having information on intellectual ability available for 70% of children who met the ASD case definition (N = 3,390). Maryland source type is education
and health records.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 21
 Surveillance Summaries

FIGURE 4. Estimated prevalence* of autism spectrum disorder among children aged

8 years, by most recent intelligence quotient score, by sex and race/ethnicity Autism
and Developmental Disabilities Monitoring Network, nine sites, United States, 2012

30

IQ>70
25 IQ unknown
IQ70

20
Prevalence

15

10

0
Male Female White Black Hispanic
non-Hispanic non-Hispanic

Sex Race/ethnicity
Abbreviations: ASD = autism spectrum disorder; IQ = intelligence quotient.
* Cases per 1,000 children aged 8 years.
Includes nine sites (Arizona, Arkansas, Colorado, Georgia, Maryland [education and health records],
New Jersey, North Carolina, South Carolina, and Utah) having information on intellectual ability
available for 70% of children who met the ASD case definition (N = 3,390).

22 MMWR/April 1, 2016/Vol. 65/No. 3 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Surveillance Summaries

FIGURE 5. Percentage of children with autism spectrum disorder at age 8 years who had previous autism spectrum disorder classification on
record, suspicion of the disorder noted, or no mention of the disorder, by site Autism and Developmental Disabilities Monitoring Network,
11 sites, United States, 2012

Previous ASD classification on record Suspicion of ASD noted No mention of ASD

100

90

80

70

60
Percentage

50

40

30

20

10

0
Arkansas Arizona Colorado Georgia Maryland Missouri North New Jersey South Utah Wisconsin All sites
Carolina Carolina combined
ADDM site
Abbreviations: ADDM = Autism and Developmental Disabilities Monitoring Network; ASD = autism spectrum disorder.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR/April 1, 2016/Vol. 65/No. 3 23
The Morbidity and Mortality Weekly Report (MMWR) Series is prepared by the Centers for Disease Control and Prevention (CDC) and is available free of
charge in electronic format. To receive an electronic copy each week, visit MMWRs free subscription page at http://www.cdc.gov/mmwr/mmwrsubscribe.html.
Paper copy subscriptions are available through the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402; telephone
202-512-1800.
Readers who have difficulty accessing this PDF file may access the HTML file at http://www.cdc.gov/mmwr/volumes/65/ss/ss6503a1.htm?s_cid=ss6503a1_w.
Address all inquiries about the MMWR Series, including material to be considered for publication, to Executive Editor, MMWR Series, Mailstop E-90,
CDC, 1600 Clifton Rd., N.E., Atlanta, GA 30329-4027 or to mmwrq@cdc.gov.
All material in the MMWR Series is in the public domain and may be used and reprinted without permission; citation as to source, however, is appreciated.
Use of trade names and commercial sources is for identification only and does not imply endorsement by the U.S. Department of Health and Human Services.
References to non-CDC sites on the Internet are provided as a service to MMWR readers and do not constitute or imply endorsement of these organizations
or their programs by CDC or the U.S. Department of Health and Human Services. CDC is not responsible for the content of these sites. URL addresses
listed in MMWR were current as of the date of publication.

ISSN: 1546-0738 (Print)