Aetna to Build a Preferred Network for Autism Providers

What Aetna Has Said

Aetna is building a preferred network for autism-therapy providers, a move that shifts competition in applied behavior analysis away from the rate a plan pays and toward a harder question: which providers get into the network at all. For an industry that spent a decade competing on capacity and billable hours, that is a bigger change than any single rate cut.

CVS Health’s insurance arm first signaled the plan in June, saying it would steer patients to providers that “get better results and bill in a manner it considers appropriate.” A CVS Health spokesperson later confirmed that Aetna is developing the network and that it may debut in 2026, and described the mechanism directly.

“We would like to work with academic groups to establish industry standards of best practices around providing autism care, and we intend to curate a network of providers whose practices align with those standards.” – CVS Health spokesperson (2026)

The rollout may be gradual. The spokesperson said it “may need to be a phased-in approach over a few years, because it’s possible that there could be certain geographies where too few providers currently meet whatever standards are ultimately established.” In some markets, there may not yet be enough qualifying providers to build a network from.

Aetna’s Existing Quality Tier

Aetna is not starting from scratch. It already runs an Institute of Quality (IOQ) behavioral health network, a designation it awards to ABA provider groups that meet its strict criteria. A 2025 Aetna listing spells out those criteria: timely access to care, meeting medical-necessity standards, measuring treatment progress and outcomes with evidence-based tools, holding accreditation from a recognized agency, involving caregivers in treatment decisions, and coordinating care with medical and behavioral health practitioners.

How many providers hold that designation is not clear from Aetna’s own materials. The listing names ACES, an ABA provider founded in 1996 and operating in nine states, as the first group to be named an ABA Institute of Quality, under a national agreement that took effect in August 2021. ACES describes itself as the only ABA provider recognized by Aetna’s Institutes of Quality. Yet the same 2025 listing includes Cortica locations under its state Institutes of Quality tables, without saying whether Cortica holds the formal designation or simply appears in the program’s provider directory. Aetna has not clarified the distinction. Either way, the tier has remained small, underscoring how narrow the qualifying circle has been.

What CVS Backed in Cortica

The clearest signal of the provider profile Aetna favors is where its parent has put money. CVS Health Ventures, the corporate venture fund of Aetna’s parent, invested in Cortica in 2023, leading a $40 million extension of the company’s Series D, bringing the round to $115 million. Cortica runs a physician-led, whole-child, value-based model that combines medical, behavioral, and developmental therapy.

The stake is no longer only financial. In November 2025, Aetna launched a neurodiversity navigation program for commercial members with autism, ADHD, and other conditions, and named Cortica among the providers those members would be routed to starting January 2026. These are separate moves rather than a single announced strategy, but they point one way. The profile of Aetna’s quality-tier rewards, outcomes measurement, accreditation, and care coordination is the profile CVS backed with capital.

Why Payers Are Moving Now

The timing is driven by cost and by the erosion of the assumption that more hours mean better care. ABA visit volume rose about 267% from 2019 to 2024, with Medicaid-covered visits up 298%. State Medicaid spending on autism services climbed from roughly $660 million in 2019 to $2.2 billion in 2023, making ABA the fastest-growing Medicaid service over that span.

The workforce expanded alongside the spending. The number of active registered behavior technicians (RBTs), the front-line staff who deliver most hands-on therapy, nearly doubled in three years, from about 130,000 at the end of 2022 to roughly 246,000 at the end of 2025. Aetna has tied that growth to billing, which it considers a problem.

Who Writes the Preferred Standard

The payers are drafting the standard that will define inclusion. The Association for Behavioral Health and Wellness (ABHW), a health-plan trade group whose members include Aetna, Optum, Centene, Carelon Behavioral Health and Health Care Service Corp., has asked the federal government to set four benchmarks for ABA: a national credentialing standard for providers, standardized supervision ratios for RBTs, closer scrutiny of companies that both diagnose and treat autism, and evidence-based clinical and quality measures developed with the Centers for Medicare and Medicaid Services. A CMS contractor is separately building a toolkit to police program integrity in autism benefits.

ABHW’s chief executive, Debbie Witchey, casts the push as quality control rather than cost control, arguing that plans spent years covering ABA without the data to judge what worked.

“As soon as [members] try to implement some changes, it’s automatically ‘they don’t want to pay for these things’ not ‘they don’t want to pay for ineffective services.’” – Debbie Witchey, President and CEO, ABHW (2026)

The framing is contestable. The same data that lets a plan flag ineffective therapy also lets it cut spending, and the cost pressure is not subtle: autism claims have tripled in five years. Whether “outcomes” works here primarily as a quality standard or as a cost-containment tool is a question that the payers’ own materials leave open.

Aetna is not alone in building that judgment into contracts. Magellan Healthcare has run a value-based collaboration with Kyo Autism Therapy since 2022, using adaptive-behavior (Vineland) scores to track progress; in its own autism program, 75% of members made substantial developmental gains at a cost of about $29,000 per year. A senior Magellan executive, Tim Daley, has described paying providers for measurable outcomes as “a core pillar of Magellan’s network strategy.” Centene and Cigna each contracted with the Behavioral Health Center of Excellence (BHCOE) in 2022 to develop autism quality measures. The preferred-network bar and the value-based bar converge on the same requirement: outcomes data.

What none of them has defined is “outcomes” itself: which instruments, how validated, and which differences in patient complexity to adjust for. The gap is not academic. A network that pays on measured results creates an incentive to improve the result, and the cheapest way to do that is to enroll less complex, better-resourced clients rather than deliver better care. Aetna has not said how it would keep an outcomes standard from rewarding provider selection over provider quality.

The Network-Adequacy Problem

A curated network runs into the law that governs Medicaid managed care. Federal rules require plans to keep networks adequate to meet demand, and the Mental Health Parity and Addiction Equity Act bars plans from covering behavioral care more restrictively than medical care. Both are already being tested in Arizona, and Aetna is closer to that fight than its preferred-network plans suggest.

Between late 2025 and March 2026, three Arizona Medicaid managed care organizations took action against the state’s two largest ABA providers following disputes over reimbursement rates. Mercy Care, which Aetna has administered under a plan-management agreement since 2007, terminated its contracts with both Action Behavior Centers and Centria; Arizona Complete Health and UnitedHealthcare Community Plan each dropped Action Behavior Centers only. Advocates put the number of children at risk of losing care close to 1,000. Two lawsuits followed in Maricopa County Superior Court, including a class action filed in February on behalf of 11 affected children that alleges the plans maintained “ghost networks,” directories listing providers who had no openings, did not take Medicaid, or had closed.

The distinction matters for any provider reading the plan as a threat to small independents alone. In Arizona, the companies cut from the networks were the largest, private-equity-backed operators, dropping over rate rather than quality. A preferred network sorts on outcomes; a rate dispute sorts on price. A provider can clear the first bar and still lose on the second.

For families, the mechanism matters less than the result. A narrower network, drawn on rates or outcomes, means fewer in-network providers to choose from, longer waits when qualifying providers are scarce, and the risk of losing a therapist mid-treatment when a contract ends. Aetna’s own spokesperson has conceded that some regions may not have enough qualifying providers to begin with.

What Qualifying Will Take

For providers, the criteria are no longer hypothetical because Aetna has already published the criteria it uses for its quality tiers: measuring outcomes with evidence-based tools, holding accreditation from a recognized agency, coordinating care, and documenting caregiver involvement. Those map closely onto the credentialing, supervision, and quality standards ABHW wants CMS to adopt. An operator that cannot show longitudinal outcome data or hold accreditation has little to submit when a preferred network opens. Platforms with institutional capital and value-based operators such as Cortica and Kyo already have that infrastructure.

Aetna has not published the criteria for the preferred network, named its academic partners, or set a firm launch date beyond a possible 2026 debut. ABHW says it will keep refining its recommendations before CMS acts. And the Arizona class action is still awaiting a ruling on whether the state’s approval of the terminations violated network-adequacy law. This question will shape how far any insurer, Aetna included, can narrow an autism network before a court steps in. The standard Aetna writes for “preferred” will decide which providers get the volume. Whether it can write one that also satisfies the adequacy law is the harder test, and that one is already before a judge.

AT A GLANCE

SOURCES & REFERENCES