Researchers from the Centers for Disease Control and Prevention are offering up a new estimate of the number of American children with intellectual disability.

In a study looking at 8-year-olds in several communities across the nation, CDC researchers found that 1.2% had IQ scores of 70 or below qualifying them for an intellectual disability diagnosis.

The findings published late last year in the Disability Health Journal are based on data collected through the CDC’s Autism and Developmental Disabilities Monitoring Network, which regularly reviews health and educational records for 8-year-olds in selected communities. The CDC uses the same methods to determine its autism prevalence estimates, which are generally updated every two years.

For the study, researchers looked at records for over 215,000 children in nine states in 2014, identifying children as having intellectual disability if they had an IQ score of 70 or less or if there was a written statement from a qualified professional indicating that the child’s intellectual functioning fell within that range.

The vast majority of children identified — 78% — had mild intellectual disability while 12% were classified in the moderate category and 1% were considered severe or profound. Researchers did not have precise IQ scores for the remaining children and could not specify severity.

Intellectual disability was about twice as likely in boys compared to girls and in Black children compared to white kids. Prevalence also varied dramatically by location with Arkansas at the high end seeing double the cases of Minnesota and Tennessee, which reported the lowest rates.

Of the children with intellectual disability, the study found that 39% also had autism.

Even though most cases of intellectual disability can be reliably diagnosed before age 5, the researchers noted that almost a quarter of the children studied did not undergo an IQ test until after age 6.

The researchers said their findings are largely consistent with previous estimates of intellectual disability prevalence, which were based on surveys, administrative records or data from single states, but those prior studies lacked details on subpopulations.

The study authors indicated that given the “substantial disparities” seen across racial, ethnic and socioeconomic groups, the “results could be used to help inform strategies to enhance early access to intervention services to improve quality of life for children with ID.”