Brief Update: Prevalence of Autism Spectrum Disorders (ASDs) --- Autism and Developmental Disabilities Monitoring (ADDM) Network, United States, 2004
CDC and seven site project teams from the Autism and Developmental Disabilities Monitoring (ADDM) Network (1,2) (Table 1) collaborated in monitoring the prevalence of autism spectrum disorders (ASDs) in 2004 in selected areas of eight states: Alabama (three counties in central Alabama), Arizona (districts in one county, including metropolitan Phoenix), Georgia (the CDC site in five counties in metropolitan Atlanta), Maryland (five counties in suburban Baltimore), Missouri (five counties in metropolitan St. Louis), North Carolina (eight central counties), South Carolina (23 counties in the Coastal and PeeDee regions), and Wisconsin (three counties in south-central Wisconsin).
The 2004 surveillance year represented a smaller-scale effort than other ADDM Network surveillance years conducted previously (3,4) and in 2006. Participation in the 2004 surveillance year was undertaken by sites that had the available time and resources to collect additional data and it included a smaller catchment area than in 2006 for five of the eight sites (Alabama, Arizona, Maryland, North Carolina, and Wisconsin). Because the smaller population monitored in 2004 might not be comparable to the larger area surveyed in other years, caution is needed when comparing 2004 results with those from other surveillance years. However, these additional data provide information on the status of identified ASD prevalence in the sites included in this summary.
Prevalence of ASDs in U.S. children was estimated through a systematic retrospective review of evaluation records in multiple sites participating in the ADDM Network. Data for 2004 were collected retrospectively from existing records from eight ADDM Network sites. Children aged 8 years (i.e., those born in 1996) with a notation of an ASD or descriptions consistent with an ASD were identified through screening and abstraction of existing health and education records containing reported professional assessments of the child's developmental progress at health-care or education facilities. Children aged 8 years whose parent(s) or legal guardian(s) resided in the respective areas in 2006 met the case definition for an ASD if their records documented behaviors consistent with the Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision (DSM-IV-TR) criteria for autistic disorder, pervasive developmental disorder, not otherwise specified (PDD NOS), or Asperger disorder. Presence of an identified ASD was determined through a review of data abstracted from developmental evaluation records by trained clinician reviewers.
For 2004, a limited (or streamlined) abstraction and review of evaluation records was conducted by six of the eight sites (all except North Carolina and South Carolina) for children with a previously documented classification of an ASD (i.e., a diagnosis of an ASD or a special education eligibility classification of autism documented in the records). This limited review was based on data obtained from MADDSP ASD surveillance and indicated that 98.0% of children aged 3--10 years with a previous ASD diagnosis and 99.0% of children with a previous autism eligibility for special education services satisfied the surveillance criteria for having an ASD (5). This addition was implemented to improve efficiency without altering the basic case classification standards. The limited abstraction included all information as noted above except the verbatim behavior descriptions. CDC's analysis of the resource savings indicated that they were not worth the loss of the variables summarizing the behavioral descriptions collected. Therefore, for the 2006 surveillance year, the standard abstraction and review methodology was used instead of the limited review used in 2004.
For 2004 surveillance year records that underwent limited abstraction, children were considered to meet the ASD case definition on the basis of the previously documented ASD diagnosis unless 1) conflicting information was noted in the record, 2) the reviewer needed additional information, or 3) the record indicated that an ASD had been ruled out as a diagnosis. In those circumstances, a full abstraction was performed, and the case was reviewed again by the clinician reviewer. For any record, if a child met the ASD case definition, and the clinician reviewer had cause to question the case status, the reason was noted, and a blinded secondary review was undertaken. Final ASD case status was established on the basis of a consensus review. For all sites, the range of agreement for final case definition was acceptable (76.0%--94.0%; Kappa range 0.5--0.9) for 2004. Methods for 2004 were otherwise comparable to those used for the 2006 surveillance year. Prevalence results are reported per 1,000 children aged 8 years. Chi-square tests were used to compare prevalence estimates within and across sites.
On the basis of postcensal estimates, the number of children aged 8 years in the eight surveillance sites ranged from 11,312 in Wisconsin to 45,190 in Georgia (Table 1) (6). Distribution according to race or ethnicity among children aged 8 years varied across surveillance sites. The percentage of non-Hispanic white children residing in each surveillance area ranged from 40.4% in Georgia to 84.8% in Wisconsin; the percentage of non-Hispanic black children ranged from 4.9% in Wisconsin to 42.4% in Georgia; the percentage of Hispanic children ranged from 2.7% in Missouri to 40.9% in Arizona; the percentage of Asian/Pacific Islander children ranged from 1.1% in South Carolina to 5.4% in Georgia; and the percentage of American Indian/Alaska native children ranged from ≤0.5% in all sites except Arizona, with 2.5%. Although most sites had a similar distribution of the population by race/ethnicity, compared with the larger surveillance areas for 2006, Alabama's smaller 2004 surveillance area included a greater proportion of non-Hispanic black children, and Wisconsin's 2004 area included more non-Hispanic white children (Table 1). The breakdown by sex was similar across sites, with approximately equal distribution of male and female children (Table 1).
In 2004, the overall identified ASD prevalence per 1,000 children aged 8 years varied across ADDM sites (range: 4.6 [Alabama] -- 9.8 [Arizona]) (Table 2). The average across all sites was 8.0 (CI = 7.6--8.4) per 1,000 children. Among the eight 2004 sites, six were clustered in a tighter range (7.8--9.8 per 1,000 children), and these rates did not differ from each other significantly. However, prevalence in Alabama (4.6) and South Carolina (5.3) was significantly (p<0.01) lower than in the other six sites. In general, identified ASD prevalence estimates in 2004 were lower in sites that relied solely on health sources to identify cases (mean: 7.3; CI = 6.5--8.1) compared to sites that also had access to education sources (mean: 8.3; CI =7.7--8.8) (p<0.05).
A consistent finding in all sites was a significantly higher (p<0.001) prevalence of ASDs among boys than among girls (Table 2). Identified ASD prevalence among males ranged from 6.8 in Alabama to 15.8 in Arizona with an average of 12.9 (CI = 12.2--13.7) per 1,000 children aged 8 years. Female prevalence ranged from 1.5 in South Carolina to 3.7 in Wisconsin with an average of 2.9 (CI = 2.6--3.3) per 1,000 children aged 8 years. When male-to-female prevalence was compared, observed sex ratios ranged from 3.0:1.0 in Alabama to 6.1:1.0 in both North and South Carolina, with an overall ratio for all sites of 4.5:1.0.
ASD prevalence varied by race/ethnicity (Table 2). In 2004, the average prevalence among non-Hispanic white children (9.7; CI = 9.1--10.4) was greater than for non-Hispanic black (6.9; CI = 6.2--7.6) (p<.001) and Hispanic (6.2; CI = 5.0--7.5) (p<0.001) children. Although several sites trended toward higher ASD prevalence among non-Hispanic white than among non-Hispanic black children, the difference was significant in only one site (Missouri) and another site (Maryland) showed the opposite with higher prevalence among non-Hispanic black children. Prevalence was lower for Hispanic than for non-Hispanic white children in all sites except Maryland, but was significantly lower only in Arizona and Georgia. For Asian/Pacific Islander children, the prevalence ranged between 4.3 and 12.3 per 1,000 children, but most sites had small numbers of cases, if any, identified in this group. Small populations and case numbers prohibited reliable prevalence estimates for American Indian/Alaska Native children.
Data regarding cognitive functioning are reported for sites for which at least 70% of children had test results. For 2004, the proportion of children with ASDs who also had cognitive impairment ranged from 37.9% in Arizona to 63.0% in Alabama (average: 43.8%). Overall, females were more likely than males (56.9% and 44.5%, respectively) to have cognitive impairment.
More than half of children meeting the ASD case definition for ASDs had documented developmental concerns noted to occur before the age of 3 years. The most commonly documented early developmental concern was for language skills. General concerns about development were documented to have occurred prior to a median age of 24 months in all sites, except South Carolina, where the median age was 36 months. The median age of earliest reported ASD diagnosis documented in the children's records ranged from 48 months in North Carolina to 67 months in Arizona (average of 57 months).
2004 Identified ASD Prevalence Summary
ASD prevalence estimates for children aged 8 years ranged from 4.6 to 9.8 among 8 sites with an average of 8.0 per 1,000 children in 2004. On average, 4.5 males were identified as having an ASD compared with every one female, and 44% of children identified also had cognitive impairment. Variability existed by race and ethnicity, with some sites showing higher ASD prevalence estimates among non-Hispanic white children than non-Hispanic black children and Hispanic children.
Considering data from the 2004 surveillance year illustrates the incremental increases in identified ASD prevalence over the short time period 2002--2006 based on the ADDM Network methods. For all sites except South Carolina, where ASD prevalence decreased from 2002 to 2004 and then increased considerably from 2004 to 2006, the data indicate the incremental and steady increase in identified ASD prevalence over the period reported.
- Rice CE, Baio J, Van Naarden Braun K, Doernberg N, Meaney F J, Kirby RS, for the ADDM Network. A public health collaboration for the surveillance of autism spectrum disorders. Paediatr Perinat Epidemiol 2007;21:179--90.
- CDC. Evaluation of a methodology for a collaborative multiple source surveillance network for autism spectrum disorders---Autism and Developmental Disabilities Monitoring Network, 14 sites, United States, 2002. In: Surveillance Summaries, February 9, 2007. MMWR 2007;56(No. SS-1):29--40.
- CDC. Prevalence of autism spectrum disorders---Autism and Developmental Disabilities Monitoring Network, six sites, United States, 2000. In: Surveillance Summaries, February 9, 2007. MMWR 2007;56(No. SS-1):1--11.
- CDC. Prevalence of autism spectrum disorders---Autism and Developmental Disabilities Monitoring Network, 14 sites, United States, 2002. In: Surveillance Summaries, February 9, 2007. MMWR 2007;56(No. SS-1):12--28.
- Yeargin-Allsopp M, Rice C, Karapurkar T, Doernberg N, Boyle C, Murphy C. Prevalence of autism in a US metropolitan area. JAMA 2003;289:49--55.
- CDC, National Center for Health Statistics. Estimates of the July 1, 2000--July 1, 2007, United States resident population from the vintage 2007 postcensal series by year, county, age, sex, race, and Hispanic origin, prepared under a collaborative arrangement with the U.S. Census Bureau. Bethesda, MD: U.S. Department of Health and Human Services, CDC, National Center for Health Statistics; 2007. Available at: . Accessed December 7, 2009.