Prevalence and Treatment Patterns of Autism Spectrum Disorder in the United States, 2016

Key Points

Question  What are the current prevalence and treatment patterns of autism spectrum disorder (ASD) among US children?

Findings  According to data from the 2016 National Survey of Children’s Health, among 43 032 US children aged 3 to 17 years, 2.79% were ever diagnosed as having ASD. Among 1115 children with current ASD, 43.3% were treated with behavioral treatment only, 6.9% with medication treatment only, 20.3% with both behavioral and medication treatments, and 29.5% with neither treatment.

Meaning  The prevalence of ASD is relatively high in the United States, and about one-third of US children with ASD did not receive behavioral or medication treatment.

Importance  Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder. Previous surveys have reported a steady increase in ASD prevalence in US children over the past decades. Several behavioral therapies and medications have been developed to treat the symptoms of ASD; however, little is known about the current status of treatment usage for children diagnosed as having ASD.

Objective  To estimate the prevalence and treatment patterns of ASD among US children using nationally representative data.

Design, Setting, and Participants  This study used data from the 2016 National Survey of Children’s Health, a nationwide, population-based, cross-sectional survey. We included 43 032 children aged 3 to 17 years. Data were collected through questionnaires completed by a parent or guardian. Data were analyzed from February 2018 to March 2018.

Main Outcomes and Measures  Outcome variables included ASD diagnosed by a physician or health professional and the use of behavioral treatment or medication treatment among children with ASD.

Results  Of the 43 032 included participants, 22 072 (51.3%) were male, and the mean (SD) age was 10.7 (4.4) years. The weighted prevalence of ever-diagnosed ASD and current ASD were 2.79% (95% CI, 2.46-3.12) and 2.50% (95% CI, 2.21-2.79), respectively. The state-level prevalence of ever-diagnosed ASD varied from 1.54% (95% CI, 0.60-2.48) in Texas to 4.88% (95% CI, 2.72-7.05) in Florida. Nationally, about 70% of children with current ASD (70.5%; 95% CI, 65.1-75.8) were treated; 43.3% (95% CI, 37.4-49.2) received behavioral treatment only, 6.9% (95% CI, 3.7-10.1) received medication treatment only, and 20.3% (95% CI, 16.5-24.1) received both behavioral and medication treatments. The remaining 29.5% (95% CI, 24.2-34.9) of children with current ASD did not receive either behavioral or medication treatment.

Conclusions and Relevance  This study showed that the prevalence of ASD in the United States was relatively high, and it varied substantially across US states. Almost 30% of US children with ASD did not receive behavioral or medication treatment, which calls for a critical need to understand and address the barriers for those children to receive appropriate treatments.

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by repetitive behaviors and restricted interests, social impairments, and communication difficulties.¹ The etiology of ASD remains to be understood, but it likely involves a complex interplay of genetic and environmental factors.^2,3 So far, there is no cure for ASD, and children with ASD need long-term and ongoing treatment and care. Autism spectrum disorder not only results in significant caregiving burden for individuals, families, and society but also causes substantial economic costs.⁴ In the United States, estimates for ASD-related individual lifetime costs were on average $1.4 million among individuals with an ASD without intellectual disability and $2.4 million among individuals with an ASD with intellectual disability.⁵

An upward trend in ASD prevalence has been reported in many developed and developing countries around the world,⁴ including the United States.^6-9 For example, the Autism and Developmental Disabilities Monitoring Network (ADDM) reported that the estimated prevalence of ASD increased from 0.67% in 2000 to 1.46% in 2012.⁶ A 2018 study⁹ using data from the National Health Interview Survey found that the prevalence of ASD in US children aged 3 to 17 years was 2.24% in 2014, 2.41% in 2015, and 2.76% in 2016.

While treatments targeting ASD symptoms abound, behavioral and medication treatments have been subject to rigorous studies of effectiveness. Targeted and comprehensive behavioral treatment models, based on the principles of applied behavior analysis, aim to reinforce socially acceptable behaviors and decrease or eliminate undesirable behaviors and are regularly identified as well-established treatments for ASD.^10-12 Early intensive behavioral intervention for children with ASD has the potential to greatly affect developmental outcomes, particularly with respect to behavior, adaptive skills, and communication.^12-16 Because developmental levels and symptoms vary substantially in children with ASD, there is so far no one universally effective treatment method for all individuals with ASD. In addition to behavioral interventions, psychotropic medications are often used in children with ASD to treat behavioral symptoms, including hyperactivity, inattention, impulsivity, aggression, outbursts, self-injury, obsessive-compulsive behaviors, rigidity, repetitive behaviors, anxiety, depression, and sleep disorders. Previous studies suggest that 46% to 56% of children with ASD had taken at least 1 psychotropic medication in the past year.^17-19 Presently, little is known about the status of behavioral and pharmacological treatment patterns of ASD in US children.

In this study, we analyzed nationally representative data from the National Survey of Children’s Health (NSCH) to estimate the most recent prevalence of ASD among US children at the national and state level in 2016. Moreover, we estimated the proportions of behavioral and medication treatments used among those diagnosed as having ASD.

The NSCH is a leading national survey on children’s health and well-being in the United States. It is sponsored by the US Department of Health and Human Services, Health Resources and Services Administration, Maternal and Child Health Bureau. The NSCH provides rich data on multiple, intersecting aspects of children’s lives, including physical and mental health, access to quality health care, and the child’s family, neighborhood, school, and social context.²⁰ The National Center for Health Statistics Research Ethics Review Board approved all data collection procedures for the survey. The University of Iowa institutional review board determined that the present study was exempt based on the use of deidentified data.

The NSCH was conducted periodically in 2003, 2007, and 2011/2012 by the National Center for Health Statistics and then annually since 2016 by the US Census Bureau. In the present study, we analyzed data from the 2016 NSCH because information on ASD treatment was ascertained only in this cycle.

The NSCH collects data on demographic characteristics and a broad range of health topics through questionnaires. The respondents to the NSCH survey are parents or guardians who know the child’s health best. For the 2016 NSCH survey, households received a mailed invitation asking an adult who is familiar with their child’s health and health care in the household (usually the parent) to go online to complete a short screener questionnaire. The screener questionnaire asked participants to identify all children aged 0 to 17 years living in the household. If a child lived in the household, the participants were then immediately directed to a more detailed, age-specific topical questionnaire. Only 1 child per household was randomly selected to be the subject of the detailed topical questionnaire. All nonresponding households received a reminder in the mail. A mailed paper-and-pencil screener was provided if the household did not respond to the first 2 web survey invitations. A total of 50 212 interviews were completed nationally by parents and caregivers of children. The number of respondents averaged approximately 985 per US state and the District of Columbia and ranged from 638 in Mississippi to 1351 in Minnesota.

In the 2016 NSCH, respondents were asked, “Has a doctor or other health care provider ever told you that this child has Autism or Autism Spectrum Disorder (ASD)? Include diagnoses of Asperger’s Disorder or Pervasive Developmental Disorder (PDD).” This question was responded to by 99.4% of participants. Respondents who reported a history of ASD were further asked whether the child currently had ASD, at what age the diagnosis was made, and whether the child had received behavioral or medication treatment for ASD. Parent report of behavioral treatment was combined with medication treatment to estimate overall ASD treatment prevalence among children with current ASD.

Information on age, sex, race/ethnicity, family income, and insurance type was collected using a standardized questionnaire. Age was grouped into 3 categories: aged 3 to 5 years, 6 to 11 years, and 12 to 17 years. Self-reported race and Hispanic origin were categorized into Hispanic, non-Hispanic white, non-Hispanic black, non-Hispanic Asian, and other, according to 1997 US Office of Management and Budget standards. Family income levels were classified based on the ratio of family income to poverty guidelines specific to the survey year and were categorized as less than 1.0, 1.0 to 1.9, 2.0 to 3.9, and 4.0 or greater.

In the present study, we included all children aged 3 to 17 years with available information on ever having a diagnosis of ASD. The same age range was used in previous reports about ASD prevalence in national surveys.^7,9,21,22

Because the NSCH data were collected in a manner that allows comparisons among states as well as nationally, we estimated the prevalence of ASD both nationwide and by state. We also estimated the national prevalence of ASD according to population characteristics, including age, sex, race/ethnicity, family income to poverty ratio, and insurance type. P values for overall differences in prevalence or percentages across strata were calculated by the χ² test. Logistic regression models were used to assess the associations of age, sex, race/ethnicity, family income to poverty ratio, and insurance type with national prevalence and with dichotomized treatment types of ASD. In addition, multinomial logistic regression models were used for mutually exclusive combinations of treatment types (ie, neither behavioral nor medication treatment, behavioral treatment only, medication treatment only, and both behavioral and medication treatments), and the neither behavioral nor medication treatment category was used as the reference group for the calculations of the relative risk ratios.²³

We used survey weights, strata, and primary sampling units, which were provided along with the NSCH data, in all the analyses so that the results are nationally representative to the noninstitutionalized children in the United States. All analyses were performed using survey procedures in SAS version 9.4 (SAS Institute). All P values were 2-sided, and a P value less than .05 was considered statistically significant.

Of the 43 032 eligible children aged 3 to 17 years in the 2016 NSCH, 1229 (2.9%) were reported to have ever been diagnosed as having ASD. Of those with data on age at diagnosis, 738 of 1121 of those with ASD (65.8%) were diagnosed at age 5 years or younger. The weighted prevalence of ever-diagnosed ASD was 2.79% (95% CI, 2.46-3.12). Consistent with previous findings,^6-9 the prevalence of ASD was higher in boys (4.35%; 95% CI, 3.76-4.93) than in girls (1.16%; 95% CI, 0.88-1.45) (Table 1). The state-level prevalence varied from 1.54% (95% CI, 0.60-2.48) in Texas to 4.88% (95% CI, 2.72-7.05) in Florida (Figure). Among those who had ever been diagnosed as having ASD, 92.79% (95% CI, 90.19-95.39) reported currently having ASD. The weighted prevalence of current ASD was 2.50% (95% CI, 2.21-2.79) (Table 2). The patterns of current ASD prevalence by population characteristics were similar to the patterns of ever-diagnosed ASD. In multivariable logistic regression models, we found that female sex and non-Hispanic Asian race were significantly associated with lower odds of both ever-diagnosed ASD and current ASD, whereas public insurance was associated with higher odds of both ever-diagnosed ASD and current ASD (Table 2).

Most children with current ASD (70.5%; 95% CI, 65.1-75.8) were reported to be receiving treatment; 43.3% (95% CI, 37.4-49.2) received behavioral treatment only, 6.9% (95% CI, 3.7-10.1) received medication treatment only, and 20.3% (95% CI, 16.5-24.1) received both behavioral and medication treatments. The remaining 29.5% (95% CI, 24.2-34.9) of children with current ASD did not receive either behavioral or medication treatment. Younger children were more likely to receive behavioral therapies, while older children were more likely to receive medication therapies (Table 3). In addition, male and non-Hispanic black children with ASD were more likely to receive medication therapies. Insured children were more likely to receive behavioral treatment and medication treatment. There was no significant difference across age, sex, race/ethnicity, or family income levels for no treatment among children with ASD (Table 3). Binary logistic regression analysis showed that age was significantly associated with the use of behavioral treatment and medication treatment among US children with current ASD, while sex, race/ethnicity, and family income were not significantly associated with treatment use (Table 4). Insurance type was a predictor associated with no treatment among children with ASD. In the multinomial logistic regression, older age was associated with a higher likelihood of using behavioral treatment only, female sex was associated with a higher likelihood of using medication treatment only, and no insurance was associated with a higher likelihood of using behavioral treatment only as well as both behavioral and medication treatments (eTable in the Supplement).

In a large, nationwide, population-based study, the estimated prevalence of ever-diagnosed ASD was 2.79% among US children in 2016. This number was comparable with the reported prevalence in 2016 (2.76%) using nationally representative data from another national health survey (the National Health Interview Survey).^9,21 The prevalence was higher than estimates in previous years (up to 2012) from the ADDM,⁶ although 2016 estimates from the ADDM are not currently available.

To our knowledge, this is the first study reporting state-level prevalence of ASD in the United States. We found that the state with the highest prevalence (4.88% in Florida) had triple that of the state with the lowest prevalence (1.54% in Texas). Previous studies have reported geographical variation by census regions (ie, Northeast, Midwest, South, and West).^6,9 With a large sample size and the unique study design, the present study was able to show that ASD prevalence varied appreciably by states even in the same census region (eg, Texas and Florida, both in the South region). More investigation is warranted to understand the reasons for the geographic variations in ASD prevalence in the United States.

Previous research suggests that evidence-based interventions, especially if initiated early,²⁴ can alleviate problem behaviors and promote improvement in core symptoms common to children with ASD.^14-16 Although there is currently no cure for ASD, symptoms can decrease over time with appropriate treatment and, in a small minority, be minimized to the extent that they lose their diagnosis of ASD.^25-27 The establishment of appropriate treatment strategies in the early years could minimize or even avoid subsequent behavioral problems.²⁴ In the current study, we found that nearly 30% of children with current ASD did not receive behavioral or medication treatment. The percentage of no treatment among US children with ASD did not vary significantly by age, sex, race/ethnicity, or family income level. Future studies are required to identify the underlying reasons why some children with ASD do not receive treatment.

The major strength of the present study is the use of nationally representative data from a leading national health survey, making our findings representative of US children. In addition, the large sample size and multiracial and multiethnic diverse population allowed us to investigate population disparities and state-level variations in ASD prevalence.

This study has several limitations. First, information on physician diagnosis of ASD was self-reported by parents or guardians, which may be subject to misreporting and recall bias. However, the prevalence of ASD in this study was comparable with the estimates in another nationwide survey in the same year.^9,21 Second, although ASD is diagnosed using standardized tools, it is still possible that ASD may be overdiagnosed or underdiagnosed by certain health care professionals and institutions. Further investigation is needed to identify the influence of overdiagnosis and underdiagnosis patterns on national estimates and state variations in ASD prevalence. Third, we did not know which specific behavioral therapies or medications were used for treatment or the timing and dosage of behavioral and medication therapies. In addition, no information was gathered on other common auxiliary treatments for ASD, such as speech and language therapy, occupational therapy, and complementary and alternative therapies. Despite the lack of high-quality evidence, previous studies have shown that some complementary and alternative therapies, such as vitamin supplements, special diets, and body-based practices (eg, massage therapy and chiropractic), were frequently used by families of children with ASD.^28-31

This study showed a relatively high prevalence of ASD among US children, which varied substantially across US states. Moreover, almost 30% of US children with ASD did not receive behavioral or medication treatment. Continued monitoring of the prevalence and treatment of ASD is warranted. Future research and policy efforts are critically needed to understand and address the barriers for children with ASD to receive appropriate treatments.

Accepted for Publication: August 27, 2018.

Corresponding Author: Wei Bao, MD, PhD, Department of Epidemiology, College of Public Health, University of Iowa, 145 N Riverside Dr, Room S431 CPHB, Iowa City, IA 52242 (wei-bao@uiowa.edu).

Published Online: December 3, 2018. doi:10.1001/jamapediatrics.2018.4208

Author Contributions: Dr Bao had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Strathearn, Bao.

Acquisition, analysis, or interpretation of data: Xu, Liu, O’Brien, Kopelman, Zhu, Snetselaar, Bao.

Drafting of the manuscript: Xu.

Critical revision of the manuscript for important intellectual content: Strathearn, Liu, O’Brien, Kopelman, Zhu, Snetselaar, Bao.

Statistical analysis: Xu.

Administrative, technical, or material support: Liu, Snetselaar, Bao.

Study supervision: Strathearn, Bao.

Conflict of Interest Disclosures: None reported.