Autism Identified Earlier, Then Interrupted by COVID-19: 4-Year-Olds in 2020

CDC's Autism and Developmental Disabilities Monitoring (ADDM) Network tracks autism spectrum disorder (ASD) by reviewing health and education records for children aged 4 and 8. In 2020, 11 sites — in Arizona, Arkansas, California, Georgia, Maryland, Minnesota, Missouri, New Jersey, Tennessee, Utah and Wisconsin — took part. This report looks at 4-year-olds, and compares them with the 8-year-olds of the same year to see how early identification has changed.

Image from CDC's report on early identification of autism among 4-year-olds.

A child counted as having ASD if their records ever showed an ASD diagnosis from a qualified professional, a special education classification of autism, or an ASD diagnostic code. A child had suspected ASD if a qualified professional had documented a suspicion of autism but none of those applied.

How common

• 21.5 per 1,000 4-year-olds — 1 in 46 — had ASD, up 26% from 17.0 per 1,000 in 2018.
• Prevalence ranged from 12.7 per 1,000 in Utah to 46.4 in California.
• Boys: 32.3 per 1,000, 3.1 times the rate for girls (10.4). For the first time in ADDM Network surveillance, prevalence among girls topped 10 per 1,000 — and it rose more than among boys since 2018.
• 80.2% of children with ASD had a recorded diagnosis (from 54.1% to 98.3% across sites).

Compared with White children, prevalence was:

This pattern, first seen in 2018, may reflect efforts to make services and health care equitable — uncovering higher risk once masked by unequal access. Autism is being identified more among lower-income families than before, though they still face barriers such as transportation, childcare and job accommodations.

Intellectual disability and early evaluation

• Among the 58.3% of children with information on intellectual ability, 48.5% had an IQ of 70 or below or an examiner's statement of intellectual disability — from 24.7% in California to 67.3% in Arkansas. The share was similar for boys (49.2%) and girls (46.2%), and higher among Black children (62.7%).
• 78.0% of children with a documented developmental evaluation were evaluated by age 36 months, from 66.7% in Tennessee to 87.8% in Arkansas.

Earlier identification

At every site, 4-year-olds were more likely than 8-year-olds to have been identified with autism by age 48 months: 17.5 per 1,000 compared with 11.2, or 1.6 times as high, with site ratios from 1.3 (New Jersey and Utah) to 2.0 (Tennessee). That is progress in early identification.

California stood out, with higher prevalence at age 4 than at age 8. Local factors likely help, including hundreds of pediatricians in the Get SET Early model and a state-funded regional center system that provides evaluations and service coordination.

Then the pandemic

From 2016 through February 2020, the younger cohort was getting more evaluations and identifications than the older cohort had at the same ages four years earlier. Then the pattern flipped:

Most sites had not recovered to pre-pandemic levels by the end of 2020. Sites that recovered more may have shifted to telehealth evaluations; sustained drops may reflect long periods of remote school in places such as Minnesota, where many children are identified through schools.

Suspected autism

Suspected ASD ranged from 0.5 per 1,000 in California to 10.4 in Arkansas, and rose from 2018 at five sites (Arizona, Arkansas, Maryland, New Jersey and Utah) — possibly because the pandemic made follow-up evaluations harder. It was highest in the three states with the lowest ASD prevalence (Utah, Arizona and Arkansas), which may reflect local diagnostic rules: until 2021, Arkansas required agreement by three professionals in specified categories for an autism diagnosis; the revised code requires two qualified professionals. Children with suspected ASD resembled those with ASD, so following them up could shorten delays in diagnosis and support.

Limitations

• Surveillance areas may not represent their whole states or the country.
• Results depend on what records contain, including how sex, race and ethnicity were recorded.
• Sites and areas can change between years; comparing 4- and 8-year-olds in the same areas limits that problem.
• The surveillance definition of intellectual disability isn't a clinical diagnosis, and young children's IQ scores can be unstable.

What's being done

CDC's Act Early Initiative helps early childhood programs work together so children with autism or other delays are found and referred early; during the pandemic it funded 43 state and territorial Act Early COVID-19 response teams. The "Learn the Signs. Act Early." program offers free developmental monitoring tools in many languages and the Milestone Tracker app for parents, educators and providers. ADDM will keep watching whether the pandemic's disruption to early identification has lasting effects.

Sources

Based on Shaw KA, Bilder DA, McArthur D, et al., "Early Identification of Autism Spectrum Disorder Among Children Aged 4 Years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2020," MMWR Surveillance Summaries 72(1), Centers for Disease Control and Prevention; a work of the United States government in the public domain. Original report.