Magellan’s Cara Albanese and CentralReach’s Yagnesh Vadgama explained how payer reviewers read ABA requests, and why treatment hours should move both ways.
Key Takeaways
- Denials often trace to documentation: Albanese said three-month approvals usually signal gaps in the record, and urged providers to track denial rationales over time.
- Federal guidance now addresses intensity directly: CMS’s August toolkit says 40 hours a week is not a best practice and that higher intensity should not be authorized by default.
- Starting low does not lock a child in: Both panelists said a well-supported request to raise hours should hold up, and Albanese said a 40-hour default can itself invite scrutiny.
- The fix runs through people, not portals: Albanese urged building a relationship with utilization management, and Vadgama argued shared measures give providers collective weight.
Late in Acuity Media Network’s September webinar, a question arrived that many clinicians have wrestled with. Most ABA programs begin at high intensity and taper. Some early evidence suggests that starting lower and adding hours as needed may work as well or better. But clinicians worry that a modest first request will make it harder to get more later. Is that fear justified?
Both panelists answered the question with an immediate no. Acuity Media Network’s Sara Gershfeld put the question to the panel directly: does an approved authorization set a precedent for the next one? No, said Yagnesh Vadgama. No, said Cara Albanese. Treatment requests, Vadgama explained, are “contingent upon the needs of the individual,” not upon what was asked for last time.
Both speak from the reviewer’s side of the desk. Albanese, a Board Certified Behavior Analyst, is the Director of Clinical Services, Autism, at Magellan Health, where she has conducted utilization management reviews and managed autism strategy for a decade. Vadgama, also a BCBA, spent ten and a half years in managed care, much of it overseeing autism utilization management at Magellan, before joining CentralReach in February as Vice President of Payor Strategy and Network Development. The webinar, “Payor Relations for ABA Providers,” was presented by Acuity and sponsored by CentralReach, and denials made up the second-largest cluster of questions submitted by its more than 100 registrants. A couple of enterprise organizations described rising denials for children requesting treatment for the first time, rather than for reassessments.
How a Payer Reviewer Reads an ABA Medical Necessity Request
Albanese began with a concession. A provider operating in several states, or with several commercial and Medicaid plans in one state, may be working against many medical necessity definitions at once; providers have told her they track as many as 27. Some criteria are vague, she acknowledged. Her practical advice was to find the strictest and build to it. “Your most restrictive is going to be your safest,” she said.
She also pushed back on a reflex she hears from providers, the objection that a given practice is not spelled out in a plan’s medical necessity language. CASP’s practice guidelines and the BACB ethics code still apply, she said: “As BCBAs, we still have to uphold those things,” and treatment plans should reflect them.
The reader on the other end may not be a behavior analyst. Providers have told Albanese that physician reviewers, or utilization management staff without a BCBA, sometimes misunderstand what ABA does. The remedy is to make the request and its reasoning unmistakable, and to confirm plan-specific requirements such as whether a diagnosis of autism is needed and who is qualified to make it. “If you don’t meet those requirements,” she said, “you’re not going to get in the door.”
The denial letters themselves rarely help. Albanese said they are often written in generic terms because regulations require them to meet a set reading level, which can leave a provider puzzling over a sentence that seems to say very little. Her answer was to stop treating each denial as an isolated event. Keep a record, she said, and ask: “What is the rationale you’re getting?” A documented pattern is something a provider can bring to a champion at the plan, and something the plan can use to educate its own physician reviewers.
ABA Prior Authorization: What a Three-Month Approval Usually Signals
A Chief Financial Officer who registered for the webinar asked about weekly authorization maximums and approvals issued over short windows; another registrant described documentation audits that seemed to continue even after appropriate addendums. Albanese took the authorization piece first. The standard concurrent authorization, she said, is typically six months. A shorter one generally reflects one of two things: a member with extreme behaviors or a physical health condition complicating treatment, who may need a new comprehensive diagnostic evaluation and more frequent monitoring, or a record that is not supplying enough evidence of medical necessity and progress.
Her recommended response is to ask the plan for a call. She suggested a provider tell the plan’s utilization management contact something close to this: “We feel like we’re constantly getting three months. Like, where are we missing the mark?” Either the provider is misreading the plan’s feedback, she said, or something is consistently absent from the treatment plans, and a conversation is the fastest way to learn which.
Vadgama addressed the larger question of why the friction exists at all. He pointed to the Office of Inspector General’s audits of state Medicaid ABA programs and to the CMS toolkit. Payers, he said, are working from a simple heuristic: “If there’s smoke, there’s fire.” Most providers have done nothing wrong, he said, but the field is experiencing “guilt by association.” The OIG’s four published audits of Indiana, Wisconsin, Maine, and Colorado each found at least one improper or potentially improper claim in every sampled enrollee-month, with documentation the dominant problem.
He located part of the problem in the requests themselves. In his years overseeing utilization management, he said, treatment requests arrived with “zero overlap” from one to the next. Individualization is a principle of the discipline, he granted, but at that scale it becomes difficult for anyone outside a clinic to judge what level of care a given clinical profile warrants, and that uncertainty travels upward to state agencies and CMS.
The operational side of the friction is real, too. In Wisconsin, providers reported roughly 400 pending prior authorization requests before ForwardHealth said it would backdate approvals. In Georgia, Amerigroup moved ABA authorization requests to the state portal on September 1, with new submission windows and no backdating. Short authorization periods are also being written into law: according to the CMS toolkit, North Carolina now requires plans above 16 hours a week to be approved by the state or the managed care plan and reapproved every three months.
ABA Titration: Why Starting at Lower Intensity Does Not Cap Future Hours
To answer the titration question, Vadgama built a case. A nine-year-old switches providers and arrives with a diagnostic report from age five and an authorization for 10 hours of ABA a week. The child stops responding to treatment; problem behavior stays high or swings widely. Asked about the rest of the child’s life, the parents describe broken sleep and a growing fear of school. A new evaluation points to possible sleep problems and a possible mood disorder. The provider asks to move from 10 hours toward something like 16, with more supervision and more parent training, coordinated with the child’s other clinicians.
Vadgama said he could “almost guarantee” how a reviewer would respond: “Okay, I see what you’re doing there. That makes total sense.” Once the team regains instructional control, the hours come back down. He compared it to diabetes care, where a spike in a patient’s A1C calls for more insulin until the numbers stabilize. “ABA is not unique to say that once you’re low, you can’t go back up,” he said. The harder part, in his view, is producing the validated assessment results that show an increase is needed. The CMS toolkit points the same way, recommending that treatment plans specify titration and step-down.
The Payer Case Against a 40-Hour ABA Treatment Default
Albanese agreed, then turned the question around. The more common problem she sees runs the other way: a belief that treatment must begin at 40 hours a week. For a young child with severe behaviors, she said, a narrower start aimed at those behaviors often makes more sense, and leaves room for naps. Instead, “we see the kitchen sink thrown, and the child is now exhausted. We’re actually exacerbating behaviors.” She suspects some of it stems from the fear the registrant described, that hours given up cannot be recovered.
From the payer’s chair, that approach tends to invite the very questions providers want to avoid. “I’m seeing stagnation in progress. I’m seeing worsening in behaviors,” Albanese said, describing what a reviewer might conclude, and “that’s actually worse outcomes for the member.” An increase that is supported, by contrast, raises no particular concern. Gershfeld, who moderated, observed that a deliberate titrate-up model could become a differentiator for providers willing to build one.
Federal guidance leans the same way. The CMS toolkit released August 4 states that 40 hours of ABA a week is not a best practice, in part because children need time to eat, use the bathroom, and nap, and that higher-intensity services should not be authorized by default. It also notes that there is no consensus on a set number of hours, and that Medicaid and CHIP beneficiaries receiving ABA for autism averaged 17.33 hours a week in 2025, up 22 percent since 2021. Some states have gone further; Indiana has set a 4,000-hour lifetime cap on comprehensive ABA.
The research does not settle the matter, and the toolkit itself presents it as divided, citing two meta-analyses that found higher intensity associated with greater gains and two that found no difference. When seven providers benchmarked their own outcomes against CASP’s intensity bands this year, they reported results broadly consistent with a dose-response pattern, while a 2024 meta-analysis found no clear link between intervention amount and gains, a disagreement that continues. What the panelists described is compatible with either reading: hours set by the child in front of the clinician, and not by a default.
Validated ABA Assessments as the Common Language of Authorization
Both panelists returned repeatedly to measurement as the thing that makes a request legible. Albanese advised choosing one validated assessment and staying with it across authorization periods and across clients; switching tools, she said, produces outliers no one can compare. The CMS toolkit makes a related recommendation, describing it as best practice for states to require at least one standardized outcome assessment instrument and saying states should not allow provider-created outcome measures.
The choice of instrument is itself contested. One attendee noted that some skills assessments have to change as a child outgrows them, and Vadgama agreed. When another asked why several Medicaid plans now limit assessments to the VB-MAPP, the Vineland, and the ABLLS, Albanese said the plans are “just trying to limit variability,” and pick the tools that come up first when they look. Vadgama said the field should advocate for a set better suited to the population it serves.
How ABA Providers Get a Denial Pattern in Front of the Right Payer Contact
Documentation, however careful, eventually needs a listener. Albanese suggested starting with the plan’s utilization management review team. They may not be the decision makers, she said, but they can connect a provider to the people who are, and a relationship built there tends to open further conversations. Vadgama was candid that the decision maker varies by plan, and argued that providers who adopt shared measures give plans more reason to engage with them together.
Albanese ended her answer on denials with the plan’s own responsibilities. A provider who arrives with a documented pattern of denials, she said, gives the plan something to act on, including conversations with its physician reviewers about what ABA can do. As she put it, “there’s a lot of education we can do on our side as well.”
