Cortica’s Dr. Suzanne Goh and 360 Behavioral Health’s Dr. Dennis Dixon arrive at the same destination from different starting points: stop arguing about ABA dosage and clinic-versus-home, and start measuring outcomes. The 40-hour standard is a fee-for-service artifact, the home-versus-center fight is a process distraction, and the field’s exit ramp runs through value-based contracting and rigorous outcomes data.
Key Takeaways
- The 40-hour ABA standard and the home-versus-center debate are both fee-for-service artifacts: Goh argues directly that paying for hours produced the 40-hour default, and that the dosage debate is downstream of how the field gets paid. Dixon, whose 2017 study has been used to justify center-based models, says he would like the location debate to “fade away” in favor of outcomes-driven decisions.
- Dixon’s 2017 home-versus-center study has been read more rigidly than its author intended: The study found higher mastery rates per hour in clinic-based treatment, which has been cited in defense of 100% center-based service models. Dixon himself says 100% center-based is a red flag, because skills mastered in a clinic must generalize to the home, school, and community settings where the behaviors actually occur.
- Outcomes measurement is the exit ramp from process debates: Both leaders are converging on the same conclusion: providers competing on dosage benchmarks or location are competing on process, while providers competing on outcomes are competing on results. The OIG audits and tightening payer requirements are accelerating that shift.
- Value-based contracting requires infrastructure most ABA providers have not yet built: Cortica’s value-based partnerships with Aetna and fifteen other health plans work because the payer can ingest integrated clinical data and the provider can produce it. Goh notes that many health plans remain siloed between behavioral and physical health, which limits broader VBC adoption across the field.
Two of the most prominent clinical voices in autism care arrived at the Council of Autism Service Providers’ 2026 conference from very different starting points. Dr. Suzanne Goh is a pediatric neurologist and behavior analyst who co-founded Cortica, a multi-state provider that built a whole-child medical and behavioral model and now leads the industry in value-based partnerships. Dr. Dennis Dixon is the Chief Clinical Officer of 360 Behavioral Health, a large traditional ABA provider, and the lead author of a widely cited 2017 study that found higher mastery rates per hour in clinic-based treatment than in home-based treatment.
On paper, the two leaders sit in different camps. Goh’s model treats ABA as one component inside a broader clinical service line that includes occupational and speech therapy, mental health counseling, and the diagnosis and treatment of co-occurring medical conditions. Dixon’s organization is a large pure-play ABA provider whose research has been used to justify center-based service delivery for years.
What they actually agree on is more interesting than what divides them. Both argue that the location debate is the wrong axis for the field, and both argue that the exit ramp runs through outcomes measurement rather than through deeper trench warfare over hours and clinic-versus-home. Where they part ways is on what is driving the 40-hour default: Goh attributes it directly to fee-for-service incentives, while Dixon points to a subset of providers whose business model, rather than individualized clinical assessment, drives dosage decisions.
Goh’s Hammer Problem: Why the 40-Hour Standard Comes From Fee-for-Service
Goh’s framing of the dosage debate is precise. The 40-hour default exists, in her telling, because the field’s reimbursement system rewards it. “What’s happened is this 40 hour per week standard,” she said. “It’s come about because of the fee for service reimbursement system. So that has to change. We have to stop paying for hours. We have to pay for outcomes, and that will change provider behavior.”
ABA providers operating in fee-for-service environments have one revenue lever: hours billed. A provider operating only an ABA service line, without diagnostic capacity or co-occurring care, cannot earn revenue by addressing a child’s sleep disorder or treating an underlying gastrointestinal issue. The provider can only adjust the intensity of the service it sells. “If all you have is a hammer,” Goh said, “you’ll use it for everything, even when it’s not the right thing. Sometimes you need a scalpel to be much more precise.”
Because Cortica’s model integrates behavioral, developmental, occupational, speech, and family-support services with medical diagnostics and treatment of co-occurring conditions, it can address sleep, GI, metabolic, neurological, and mental-health issues that ABA-only providers route to outside specialists or do not address at all. Children whose underlying medical conditions are treated, Goh argues, often improve faster and require fewer ABA hours. The evidence behind the dosage debate supports the broader principle that hours are not an end in themselves; what matters is the trajectory of the child.
Dixon’s Misread Study: What 100 Percent More Mastery Per Hour Actually Means
Dixon’s 2017 study, published in Behavior Analysis in Practice, examined the records of 313 children with autism who received ABA in either home-based or clinic-based settings, with a secondary paired-sample analysis of children who received services in both. The between-group analysis found significantly higher mastery of exemplars per hour in clinic-based care. The paired-sample analysis was sharper still: children who received both home and clinic services mastered exactly 100 percent more learning objectives per hour while at the clinic than at home. The finding has been cited frequently in defense of center-based models, and it is real. What Dixon emphasized at CASP, and elsewhere, is that the study has been read more rigidly than its conclusions support.
Higher mastery per hour in a controlled clinical environment is not the same as better long-term outcomes for the child. Skills mastered in clinics still need to generalize to the home, school, and community settings where the behaviors actually occur. “It doesn’t matter if you can master all these things in an artificial environment if you’re not generalizing it back,” Dixon said. A child who masters object identification in a clinic but cannot deploy the skill at the family dinner table has not received care that delivered the gain the parents are looking for.
Dixon’s own preferred operational answer is closer to a balanced clinic-and-home mix than to either extreme, and his standard is louder when it comes to the maximalist reading of his own research. “Anytime I see a clinic that’s reporting out 100 percent center based,” he said, “to me, that’s a red flag that something needs to be reconsidered there.” The 2017 paper measured one dimension of treatment efficiency under controlled conditions. It did not endorse a service model in which generalization is treated as someone else’s problem.
Where Goh and Dixon Converge: Outcomes Are the Exit Ramp
The convergence between the two leaders is sharpest when the conversation moves from process to outcomes. Goh quoted Dr. Warren Jones, the director of research at Marcus Autism Center, from his talk earlier in the conference: “Parents don’t want treatment. They want gains. Treatment is about what we do to get the gains.” Dixon used the same line. “I’ve used that lens to say center based, home based, this many hours, that many hours,” he said. “None of it matters unless it actually impacts the kid, the client, the adult, in their real life, in the actual functionality that they have.”
What outcomes measurement looks like in practice differs across the two organizations. Cortica’s framework spans 10 domains, integrating medical co-occurring conditions (sleep, epilepsy, genetic and metabolic disorders, GI), medication management, the developmental domains (language, communication, social, motor, sensory), and family quality of life. Goh’s view is that no single standardized assessment captures the picture. The Vineland, the most widely used adaptive behavior assessment in the field, captures one component well but misses most of the medical and quality-of-life dimensions her model is built to address.
Dixon’s organization is anchored on the Vineland and on the benchmarking work the field is now able to do against meta-analytic estimates from the Eldevik research. He pointed to CASP’s 2025 white paper on outcomes benchmarks as a meaningful step forward: providers can now judge their own performance against external standards rather than reporting in isolation. He and other clinical officers presented outcomes data on Vineland measures at the APBA conference the prior month, and again at CASP. The two approaches are not in conflict. One is more comprehensive in scope; the other is more comparable across providers. Both are improvements over what the field had a decade ago, which was a process conversation without an outcomes vocabulary. The same Eldevik benchmarks have been the subject of an active debate over what the data actually shows about high-intensity ABA, which is itself a useful illustration of why a field with rigorous outcomes data is better positioned than a field without it.
Value-Based Contracting in ABA: What Cortica’s Partnerships With Health Plans Actually Require
Cortica’s value-based partnerships with Aetna and fifteen other health plans across the country are the clearest live examples in the field of what outcomes-based reimbursement looks like in autism care. The partnerships succeed, in Goh’s telling, because three pieces of infrastructure are in place.
First, the payer can accept integrated clinical data. Many health plans, Goh noted, remain siloed between behavioral health and physical health. A health plan that cannot ingest data on sleep outcomes, medication management, and adaptive behavior in the same record cannot meaningfully reward a provider for improving them in concert. Second, the provider can produce that data with the rigor that a VBC contract requires. Cortica built its own technology platform in-house and now describes itself as a tech-enabled clinical care delivery service. Third, the contract is structured around demonstrable cost savings to the plan, better clinical outcomes for members, and member-experience metrics including access time.
The implication for other providers is that VBC is not a marketing label that can be applied to a fee-for-service contract. It is a different operational model that requires data infrastructure most ABA providers have not yet built. The accountability era arriving for autism care is being driven jointly by payer expectations and federal audit pressure, and outcomes data is what unlocks the contract structures Cortica has been able to negotiate.
The OIG Audits, the 40-Hour Default, and the Maturation Argument
Dixon framed the broader compliance environment in similar terms. “I’m not worried about the audits,” he said. “If you’re using a good software system and you’ve trained your clinicians to be taking notes properly and documenting all the things they’re doing, then the audits are not a problem. I think these are necessary steps forward in our field. It’s been the wild west for a very long time, and some folks maybe are well intended, but they’re doing sloppy work that needs to stop.” The federal audit cascade now reshaping ABA is, in his view, a maturation step. The specific documentation patterns the OIG has been flagging (vague session notes, missing signatures, time-billing inconsistencies) are operational, not clinical, and providers running quality clinical operations are positioned to come out the other side stronger. Providers who lack actual treatment planning and simply default to a standard 40-hour model, or who have weak documentation, are exposed.
Both leaders’ framings rhyme with a structural shift the OIG audits and state Medicaid responses are already enforcing. Indiana’s recent reform bulletin caps comprehensive ABA at 4,000 lifetime hours and adds family guidance and supervision requirements designed to push providers toward outcomes-oriented care. Other states are moving in similar directions. The question, increasingly, is not whether the 40-hour default will fall but how providers position themselves before it does.
The Conversation the Field Should Be Having
Dixon’s preferred ending for the home-versus-center debate is the cleanest articulation of where both leaders want the field to go. “I would just like that to fade away,” he said. “Let’s talk about outcomes. Let’s get to a point as a field where we don’t squabble about process as much as using an analysis of what process got the best outcome. So instead of debating about whether you do it in the clinic or at home, it doesn’t matter if you have a process that works.”
The implications are operationally significant. A provider whose clinical strategy is anchored on dosage benchmarks (40 hours, 30 hours, 20 hours) or on setting (clinic, home, hybrid) is competing on process. A provider whose clinical strategy is anchored on outcomes (Vineland scores, sleep improvement, family quality of life, generalization to natural settings) is competing on results. Goh and Dixon, from very different parts of the field, are describing the same shift: results are the durable terrain, and process is the conversation the field has had because results have been hard to measure consistently. That is changing. The leaders who are ahead of the change are positioning their organizations accordingly.
