As the diagnostic criteria for autism spectrum disorders (ASD) have shifted from a rigid definition to a broad spectrum, the rate of diagnosis has increased more than 4x, from 1 in 150 children in 2000 to 1 in 31 children in 2022.1 While expanded awareness and coverage can improve access to beneficial interventions, there is also the potential for overuse.
Documented rates of ASD have increased, with a prevalence rate that is 3x higher in boys than girls.2 As of 2022, CDC estimates that approximately 1 in 31 eight-year-old children (3.2%) have been diagnosed with ASD, though prevalence varied substantially across CDC assessment sites, ranging from 9.7 to 53.1 per 1,000 children.3
Additionally, CMS has reported a prevalence rate of 5% among Medicaid beneficiaries, compared to 2% for commercially insured and uninsured populations.4 The observed variation indicates that non-clinical factors, including service availability and inconsistent provider practice, likely contribute to observed ASD prevalence rates.
In 2014, CMS issued guidance regarding Medicaid coverage for autism-related therapies, stemming from comprehensive coverage directives in the Affordable Care Act.5 This guidance clarified that states are required to cover medically necessary behavioral health services for children with autism, with the Federal government providing matching funds for these covered services. By 2022, all state Medicaid programs had extended coverage to include Applied Behavior Analysis (ABA) therapy.
The ABA workforce consists of professionals with varying levels of credentialing from the Behavior Analyst Certification Board, ranging from registered behavior technicians (RBTs) that have a high school degree and complete a 40-hour training program, to board certified assistant behavior analysts (BCaBAs) with a bachelor's degree, to board certified behavior analysts (BCBAs), which requires a master's degree and supervised experience.6 While the Bureau of Labor Statistics does not track BCBAs and RBTs specifically, projected employment growth for two related occupational categories - psychiatric technicians and aides and substance abuse, behavioral disorder and mental health counselors - is 16% and 17%, respectively, from 2024 to 2034, compared to 3% overall employment growth during the same period.7,8
By its nature, fee-for-service reimbursement creates financial incentives without corresponding required outcomes reporting. Recent Federal audits in Indiana and Wisconsin found that nearly all Medicaid payments for ASD services were considered improper or potentially improper, totaling $56M and $18.5M, respectively.9,10 Common issues included inappropriately credentialed staff, treatment that was not preceded by diagnostic evaluation and billing for services not rendered. In Minnesota, an investigation into Medicaid providers allegedly paying kickbacks to parents of children receiving ABA therapy is ongoing.11 High-profile fraud cases like these highlight systemic vulnerabilities in authorization, delivery and monitoring of autism services.
Understanding trends in utilization is crucial, given that service capacity is finite and the supply of qualified providers and appointment availability is limited, making it essential that resources reach high-need patients and ensure overdiagnosis is avoided.
National all-payer claims data and Trilliant Health’s Provider Directory were used to examine ABA service utilization from 2019 to 2024, measuring both visit volume and provider supply across service types, payer groups and states. This analysis examined ABA services across four domains: assessment, individual treatment, group treatment and family/caregiver training. Assessment CPT codes include CPT 97151 (behavior identification assessment by physician or qualified healthcare professional), CPT 97152 (behavior identification supporting assessment by technician) and CPT 0362T (behavior identification supporting assessment for patients with destructive behavior in customized environments). Individual treatment CPT codes include CPT 97153 (adaptive behavior treatment by protocol administered by technician), CPT 97155 (adaptive behavior treatment with protocol modification by physician or qualified healthcare professional) and CPT 0373T (adaptive behavior treatment with protocol modification for patients with destructive behavior in customized environments). Group treatment CPT codes were defined as CPT 97154 (group adaptive behavior treatment by protocol administered by technician) and CPT 97158 (group adaptive behavior treatment with protocol modification by physician or qualified healthcare professional). Family and caregiver training CPT codes include CPT 97156 (family adaptive behavior treatment guidance) and CPT 97157 (multiple-family group adaptive behavior treatment guidance). CPT codes 97151-97158 were introduced effective January 1, 2019.
Utilization patterns were segmented by payer type – Medicaid, commercial insurance and all payer. The supply of behavioral analysts and technicians was measured based on the number of unique National Provider Identifiers (NPIs) billing ABA codes.
From 2019 to 2024, ABA therapy service volume increased by 266.9%. Individual therapy was the dominant service category throughout the period analyzed, increasing by 256.0% and representing 85.0% of total volume during the study period (Figure 1). Within that category, CPT 97153 was the most common service, accounting for over 60% of visits across the study period. Assessment services increased both in share (3.6% to 4.8%) and volume (397.9%). Family and caregiver adaptive therapy grew by 345.6% in visit volume, while share increased by 1.3 percentage points. Group therapy services were least common, with its 2.8% share remaining constant and visit volume growing by 273.2% over the period.
ABA visits increased substantially across all payer types from 2019 to 2024. However, while there were similar growth rates across payers between 2019 to 2020 (Medicaid 56.8%, commercial 56.5%, all payer 57.1%), Medicaid utilization accelerated after 2020, reaching a growth rate of 341.2% in 2023 (Figure 2). Medicaid growth outpaced commercial insurance by 38.7 percentage points in 2021, 152.7 percentage points in 2022 and 171.9 percentage points in 2023. By 2024, the gap narrowed as Medicaid growth decelerated to 298.2% of 2019 volumes, while commercial growth increased to 249.1%..
The number of unique behavioral analysts and technicians billing for ABA increased by 135.3% from 2019 to 2024, ranging from a 32.0% decline in Wyoming to 761.3% increase in Nebraska (Figure 3). The number of analysts and technicians increased in every state except Wyoming, with 41 states experiencing growth exceeding 50%. Five states, plus Washington, D.C., saw growth in excess of 200%. The supply of providers in the largest states increased at a rate that was close to or above the national average, including California (149.4%), Texas (135.6%), Florida (184.2%), New York (124.3%) and Pennsylvania (175.0%).
📌 The graph below in interactive. Hover over the dot(s) for more information.
This analysis reveals outsized growth in ABA therapy in recent years, along with substantial variation in utilization by payer type and provider supply across states. While any new procedure code is likely to increase in volume, the 2.5x difference in reported autism rates between publicly and privately insured children, combined with 341% peak growth in Medicaid service utilization and the 761% maximum state-level growth in provider supply, suggests that systemic factors beyond clinical need are likely contributing to the increase in use of autism behavioral therapy.
Some of this variation may reflect legitimate differences in access, provider availability and population characteristics. Increased awareness and improved identification of autism, particularly among historically underdiagnosed populations, represent important progress. Additionally, the rapid workforce expansion enabled by tiered credentialing requirements, particularly at the technician level where entry barriers are relatively low, has likely meaningfully increased service capacity. However, the data raises concerns about whether diagnostic standards are being applied consistently, whether financial incentives are driving inappropriate utilization and whether finite, yet growing, specialized provider capacity is reaching the children who need services most.
The documented correlation between increasing provider supply and utilization suggests that some amount of supply-induced demand, a well-established phenomenon in healthcare economics where greater provider availability leads to higher service utilization independent of underlying need, is likely. In Minnesota, where provider supply increased by nearly 700%, Federal indictments have since revealed systematic fraud involving kickbacks to parents and fraudulent billing for services that were not actually rendered. While Minnesota's case represents an extreme example, perverse incentive structures may be affecting resource allocation more broadly – because of the Federal Medical Assistance Percentage (FMAP), every dollar that states spend on covered services like ABA generates additional Federal matching funds.
For state Medicaid programs, these patterns present significant challenges in managing program integrity while ensuring access to necessary services. The current payment structure, which reimburses based on service volume rather than outcomes, creates misaligned incentives. Providers benefit financially from maximizing billable hours regardless of whether treatment produces meaningful improvements. Without systematic tracking of diagnostic processes or treatment outcomes, states lack the data necessary to determine whether spending increases reflect appropriate expansion of services or systemic problems with overdiagnosis and unnecessary treatment. The clinical complexity of diagnosis reversal, which requires formal re-assessment by specialists, combined with flexible documentation requirements for educational accommodations, may contribute to diagnostic persistence even when clinical presentation changes.
The consequences extend beyond fiscal concerns. When children who do not meet diagnostic criteria for autism occupy limited treatment slots, children with unmet therapeutic needs face longer wait times or inability to access care. Enhanced oversight mechanisms represent an important step toward addressing these challenges. Minnesota's decision to reclassify autism services as high-risk, requiring enhanced background checks, screening visits and permitting unannounced site visits, demonstrates one approach to strengthening program integrity. However, implementation requires adequate resources, including staff to conduct reviews, systems to track outcomes and capacity for case-level investigation. While Federal audits in Indiana and Wisconsin identified widespread improper payments, these audits are periodic snapshots rather than ongoing and robust monitoring across states.
Beyond enforcement, structural reforms merit consideration. Payment models that incorporate outcomes-based incentives could better align provider behavior with patient needs. Requirements for standardized diagnostic evaluations, conducted by qualified professionals independent of treatment providers, could improve diagnostic consistency. Mandatory reporting of treatment intensity and duration, compared against evidence-based guidelines, could identify outlier practices warranting review. Data systems that track not just utilization but also functional outcomes could enable assessment of whether services are producing meaningful benefits.
As state and Federal policymakers consider reforms to autism services programs, the evidence-based patterns documented in this analysis should inform evidence-based approaches that protect both program integrity and access to care. The goal must be ensuring that every child who meets diagnostic criteria for autism can access appropriate, high-quality services delivered by qualified providers, while preventing the diversion of finite resources to children who would benefit more from alternative interventions or to providers prioritizing billing optimization over therapeutic outcomes.