Reducing disparities in early intervention access will require more data collection

Sharing federal data about infants and toddlers who are referred, evaluated and found eligible for early intervention services under the Individuals with Disabilities Education Act would assist states in identifying and solving disparities in services, according to a report released Oct. 5 by the U.S. Government Accountability Office.

The IDEA Part C program provides grants to states for early intervention services, such as speech and physical therapies, to children under 3 years old who have developmental delays or who are at risk for delays and disabilities. In 2021, more than 770,000 infants and toddlers were served under Part C.

The U.S. Department of Education does not collect, or require states to collect, demographic data on children prior to enrollment in Part C. The GAO found 28 states do collect demographic data for pre-enrolled babies, but of those, the percentage of infants and toddlers who advanced from being referred to being evaluated “differed widely” by race.

For example, the percentage of children referred who received an evaluation ranged from 59 percent of American Indian or Alaska Native children to 86 percent of Asian children (a 27-percentage point difference).

However, the percentage of those who made it from the eligible stage to enrollment was more balanced across all races, the survey showed. The percentage of children deemed eligible who then enrolled ranged from 91 percent of American Indian or Alaska Native children to 95 percent of Asian and white children (just a 4-point difference).

California findings

In California, children with disabilities or developmental delays receive early intervention services through the Early Start program, which aims to provide a coordinated, family-centered system of services that are available statewide.

According to the GAO’s findings, the program appears to be streamlining the referral and evaluation process.

“The number of children who move through each step of the enrollment process as a percent of the prior stage also varies considerably from one state to another. For example, in California and Florida, 100 percent of children who were referred to Part C were evaluated, compared to 43 percent of referred children in Texas and 37 percent of referred children in the Northern Mariana Islands,” researchers stated. “A variety of factors, such as varying state policies, may contribute to these differences. For example, California indicated in our survey a programmatic policy to evaluate every child that is referred for Part C services. Our 2019 report on Child Find also found that sometimes parents do not provide consent for their child’s evaluation — either because they believe their child’s problem may be resolved over time or because of a lack of awareness about early intervention services or the early intervention process.”

In California, 97 percent of those evaluated are deemed eligible for services, far outpacing Arizona — the state with the next highest rate — at 63 percent.

“Similarly, the percentage of children who are determined eligible for services after they are evaluated varies from less than 50 percent of evaluated children in some states to 100 percent in other states,” according to the report. “Higher percentages of children who are evaluated and determined to be not eligible for services may indicate that primary referral sources, such as physicians, do not understand the eligibility criteria for their state’s Part C program.”

Actually serving children once identified is often a challenge, according to the survey, which found 46 out of 50 states that responded to the question about challenges said there was a lack of qualified service providers.

In 2019, the Legislative Analyst’s Office noted that any effort to serve more children in California may be complicated by a shortage of qualified providers. “Regional centers indicate they already have difficulty finding providers to serve current caseloads,” the LAO wrote at the time. “Although the state could address relatively small shortages by increasing provider rates (which have not been increased on a regular basis for many years), addressing larger shortages might require long term strategies such as expanding training programs and attracting more candidates into relevant fields.”


Federal education officials told the GAO that IDEA does not provide them with the authority to collect demographic data on children prior to enrollment in Part C. If the Education Department had statutory authority to collect such data throughout the enrollment process, the GAO said it could focus its assistance on maximizing access to early intervention services for all infants and toddlers who need it — a key goal of IDEA.

“Congress should consider providing authority to [the Education Department] to collect demographic data from states on children throughout the Part C process and require Education to use these data to better assist states to identify and rectify gaps in access to services,” the GAO concluded. “GAO recommends that [the Education Department] encourage states to use existing data to maximize children’s access to Part C services.”

While many states already collect such data, and some use it to identify ways to improve Part C access, the GAO also recommended the Education Department encourage all states to improve their Child Find efforts by using the data they already collect would help them better identify and serve those infants and toddlers who need support.