ABA, Developmental Approaches, and Minnesota's Autism Treatment Landscape

Written By Hillary Duarte

What the Evidence and the Policy Actually Show

A primer for families, clinicians, and providers navigating autism intervention choices in Minnesota.

Autism treatment in the United States is shaped by two things that don't always move in sync: what the research evidence actually supports, and what insurance systems and state policy actually fund. Minnesota offers a particularly interesting case study right now, because its Medicaid autism benefit formally recognizes a range of treatment models — not just Applied Behavior Analysis (ABA) — at the same moment the program is undergoing one of the largest Medicaid fraud crackdowns in the state's history. This post walks through where ABA's "gold standard" reputation comes from, what the evidence base for structured ABA actually shows once you separate it from its naturalistic developmental cousins, and what's happening right now in Minnesota's policy landscape.

Where the "gold standard" framing comes from

Much of ABA's reputation as the default, evidence-based autism treatment traces back to a single 1987 study by O. Ivar Lovaas, which reported that 47% of children receiving roughly 40 hours a week of intensive behavioral treatment for two or more years achieved "normal intellectual and educational functioning," compared to about 2% of a comparison group [1]. That's an extraordinary effect size for a child psychiatric intervention, and it became the template for what's now called Early Intensive Behavioral Intervention (EIBI).

It's also one of the most heavily critiqued studies in the field. The treatment and comparison groups weren't randomly assigned, outcome assessors weren't blinded, and the comparison group received minimal active treatment rather than an equivalent alternative [1]. A more rigorous attempted replication by Sallows and Graupner in 2005 found real but far more modest gains, and notably found no significant difference between a clinic-directed and a parent-directed treatment group — undercutting the idea that only intensive, expert-delivered ABA produces benefit [2]. Later policy documents that cemented ABA's "gold standard" status, including the 1999 U.S. Surgeon General's report, largely built on aggregations of Lovaas-style, non-randomized studies sharing the same structural weaknesses.

What the evidence actually shows about structured ABA (apart from developmental hybrids)

This distinction matters more than it might seem. Over the past two decades, a separate category of intervention has emerged — Naturalistic Developmental Behavioral Interventions (NDBIs), including the Early Start Denver Model (ESDM), Pivotal Response Treatment, and JASPER, which blend behavioral reinforcement principles with child-led, play-based, developmentally informed delivery. Many recent systematic reviews now classify NDBIs as a subtype of "ABA-based intervention," which tends to blur the comparison researchers, clinicians, and families actually want to make: structured, discrete-trial-style ABA versus genuinely developmental models [3].

When you isolate structured ABA (EIBI/Discrete Trial Training) from its NDBI cousins, a fairly consistent pattern shows up across multiple recent meta-analyses.

Where it holds up: A 2018 Cochrane review found EIBI produced a meaningful average gain of about 15 IQ points and about 9.6 points on adaptive behavior composite scores compared with control groups, though the certainty of this evidence was rated low due to small samples and lack of blinding [4]. A 2023 meta-analysis similarly found moderate effects on intellectual functioning (SMD = 0.51) and adaptive behavior (SMD = 0.37) for comprehensive ABA-based interventions [5]. Structured ABA's manualized, highly trainable protocol and rigorous data-collection systems are also genuine practical strengths for service delivery and fidelity monitoring.

Where it consistently struggles: The same syntheses that find cognitive and adaptive behavior gains tend to find weak, inconsistent, or non-significant effects on the things closer to autism's core features. The 2023 meta-analysis above found no significant effect on language ability or symptom severity once methodological quality was accounted for [5]. A 2022 meta-analysis found effects on cognitive ability, communication, and socialization that became non-significant after excluding high-risk-of-bias studies [6]. Generalizing skills learned in structured, adult-directed teaching sessions to real-world settings has been a long-acknowledged weakness — arguably the entire reason NDBIs were developed as a corrective in the first place. Methodologically, the same problems recur across nearly every review: non-random allocation, impossibility of blinding behavioral therapists, small samples, and a near-total absence of systematic tracking of adverse effects, with multiple reviews explicitly calling on the field to start measuring harms rather than assuming their absence reflects their non-occurrence [3], [7]. There's also a documented ethical literature here: a 2025 mixed-methods systematic review aggregating qualitative studies found ABA program staff describing internal conflict and discomfort, with one practitioner reporting feeling "like I was punishing the children" during reinforcement-based sessions, alongside broader concerns from autistic self-advocates about compliance-focused, high-intensity programming [8].

The one direct head-to-head comparison available in the recent literature — a 2022 retrospective chart review comparing ESDM, EIBI, and non-specialized treatment as usual in toddlers receiving similar intervention hours — found all three groups improved similarly on autism symptom severity, but only the developmental (ESDM) group showed significant gains across broader developmental domains like personal-social functioning and hearing/speech [9]. That's a notable data point against the idea that structured ABA outperforms developmental models across the board.

Minnesota's EIDBI benefit: a policy landscape in flux

Minnesota's Medicaid autism benefit, the Early Intensive Developmental and Behavioral Intervention (EIDBI) program, is worth understanding on its own terms, separate from the national research debate, because the state's regulatory framework is unusually explicit about which treatment models it recognizes, and because the program is currently at the center of a major fraud investigation that's reshaping how it will be regulated going forward.

DIR/Floortime has formal standing alongside ABA

Under Minnesota Statute 256B.0949, the Department of Human Services (DHS) recognizes four approved EIDBI treatment modalities, with the Commissioner authorized to approve additional modalities grounded in behavioral and developmental science [10], [11]:

  • Applied Behavior Analysis (ABA)

  • Developmental, Individual Difference, Relationship-based (DIR)/Floortime model

  • Early Start Denver Model (ESDM)

  • Relationship Development Intervention (RDI)

Providers can use one modality as their primary treatment approach or combine several, provided they document fidelity to the model(s) used [10]. In other words, DIR/Floortime isn't a marginal or unofficial option within Minnesota's funding structure — it sits formally alongside ABA as a covered, recognized treatment modality.

Explosive growth, then a fraud scandal

EIDBI was established in 2014 and began receiving federal approval and Medicaid funding under the Early and Periodic Screening, Diagnostic, and Treatment (EPSDT) program shortly after, making Minnesota one of the first states to cover this kind of benefit [12]. Growth from there was dramatic: EIDBI claims rose from roughly $600,000 in 2018 to over $400 million by 2025 [13]. For its entire history until very recently, the benefit operated without a state licensing requirement for provider agencies [14].

That combination of rapid and unlicensed growth, created conditions for large-scale fraud. Through 2024 and 2025, federal and state investigators (the FBI, HHS Office of Inspector General, the IRS, and the Minnesota Attorney General's Medicaid Fraud Control Unit) uncovered and prosecuted multiple schemes tied to specific autism-center agencies, including:

  • A roughly $46.6 million scheme described by the U.S. Department of Justice as the largest Medicaid autism fraud case it has ever charged [13]

  • A separate $14 million scheme involving an agency called Smart Therapy LLC, whose owner was also implicated in the unrelated "Feeding Our Future" pandemic food-aid fraud case [15], [16]

  • A scheme run through Star Autism Center LLC that allegedly recruited children — in some cases without an actual autism diagnosis — paid cash kickbacks to parents, and hired unqualified relatives as "behavioral technicians," netting more than $6 million in fraudulent reimbursement [17]

As of late 2025, DHS reported having more than 80 open investigations into autism providers and had conducted on-site visits to over 340 EIDBI provider agencies [18]. This unfolded alongside, and is partly intertwined with, the broader Feeding Our Future scandal that has dominated Minnesota political news since 2022.

New licensing requirements are now in effect

In response, the 2025 Minnesota Legislature passed reforms requiring all EIDBI provider agencies to obtain a provisional license. New agency enrollment was paused effective November 1, 2025, and existing agencies must apply for provisional licensure by May 31, 2026 [19]. DHS is required to propose comprehensive, permanent licensing standards to the legislature by January 1, 2027, a process being developed through a multi-phase study with outside consultants and informed by a DHS-convened EIDBI Advisory Group that includes autism community stakeholders [19], [20].

What this means going forward

A few things are worth keeping in mind if you're a family weighing options or a provider operating in this space.

First, on the evidence side: the research doesn't support a clean, settled "ABA beats developmental approaches" conclusion, even setting aside which specific studies get cited in either direction. Structured ABA's strongest, most replicated advantage is on IQ and adaptive behavior composite scores; developmental and naturalistic models tend to perform comparably or better on language and broader developmental and social-communication outcomes, and direct head-to-head comparisons remain rare.

Second, on the Minnesota policy side: the fraud crisis driving the current licensing overhaul has been overwhelmingly an ABA-agency phenomenon, tied to the high-volume, one-to-one billing structure and rapid behavioral-technician staffing model that made the schemes possible — not a structural feature of lower-caseload, relationship-based models like DIR/Floortime. That said, the entire EIDBI benefit, including non-ABA modalities, is now operating under intensified scrutiny, a frozen agency-enrollment pipeline, and a legislature and press environment that doesn't always distinguish between "EIDBI fraud," "ABA fraud," and "autism services fraud" in public discourse. As DHS builds out permanent licensing standards through 2026 and 2027, this is likely where real definitional and credentialing fights over how different modalities get held to fidelity standards will play out — and it's a process worth tracking closely, since policy details here are very much still moving.

References

  1. Lovaas, O.I. (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55(1), 3–9.

  2. Sallows, G.O., & Graupner, T.D. (2005). Intensive behavioral treatment for children with autism: Four-year outcome and predictors. American Journal on Mental Retardation, 110(6), 417–438.

  3. Han, D-G., Lee, Y., Kim, H-S., et al. (2025). Effectiveness and experiences of early intensive behavioral and naturalistic developmental behavior interventions for autism spectrum disorders: a mixed-methods systematic review and meta-analysis. Child and Adolescent Psychiatry and Mental Health, 20:14. https://doi.org/10.1186/s13034-025-00997-z

  4. Reichow, B., Hume, K., Barton, E.E., & Boyd, B.A. (2018). Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD). Cochrane Database of Systematic Reviews, 5. https://doi.org/10.1002/14651858.CD009260.pub3

  5. Eckes, T., Buhlmann, U., Holling, H.D., & Möllmann, A. (2023). Comprehensive ABA-based interventions in the treatment of children with autism spectrum disorder – a meta-analysis. BMC Psychiatry, 23:133. https://doi.org/10.1186/s12888-022-04412-1

  6. Daniolou, S., Pandis, N., & Znoj, H. (2022). The efficacy of early interventions for children with autism spectrum disorders: a systematic review and meta-analysis. Journal of Clinical Medicine, 11, 5100. https://doi.org/10.3390/jcm11175100

  7. Rodgers, M., Simmonds, M., Marshall, D., et al. (2021). Intensive behavioural interventions based on applied behaviour analysis for young children with autism: An international collaborative individual participant data meta-analysis. Autism, 25(5), 1137–1153. https://doi.org/10.1177/1362361320985680

  8. Han, D-G., et al. (2025), qualitative meta-aggregation findings. Child and Adolescent Psychiatry and Mental Health, 20:14. https://doi.org/10.1186/s13034-025-00997-z

  9. Cucinotta, F., Vetri, L., Ruta, L., et al. (2022). Impact of three kinds of early interventions on developmental profile in toddlers with autism spectrum disorder. Journal of Clinical Medicine, 11, 5424. https://doi.org/10.3390/jcm11185424

  10. Minnesota Statutes, Section 256B.0949 (Early Intensive Developmental and Behavioral Intervention Benefit). https://www.revisor.mn.gov/statutes/cite/256B.0949

  11. Minnesota Department of Human Services (2025). EIDBI presentation to the Minnesota Senate Human Services Committee, January 29, 2025. https://www.lrl.mn.gov/archive/minutes/senate/2025/human/20250129/Human_20250129_DHS-EIDBI-Presentation.pdf

  12. Holt Law (2025). The Minnesota Early Intensive Developmental and Behavioral Intervention (EIDBI) Program: A Comprehensive Analysis. https://djholtlaw.com/the-minnesota-early-intensive-developmental-and-behavioral-intervention-eidbi-program-a-comprehensive-analysis/

  13. U.S. Department of Justice, Office of Public Affairs (2025). Minnesota Health Care Fraud Takedown Results in Charges Against 15 Defendants for Over $90M in Fraud. https://www.justice.gov/opa/pr/minnesota-health-care-fraud-takedown-results-charges-against-15-defendants-over-90m-fraud

  14. Minnesota Department of Human Services (2025). EIDBI Fact Sheet: Strengthening oversight of autism programs. https://mn.gov/dhs/assets/2025-03-eidbi-fact-sheet-ig-58_tcm1053-669207.pdf

  15. Internal Revenue Service, Criminal Investigation (2025). First defendant charged in autism fraud scheme. https://www.irs.gov/compliance/criminal-investigation/first-defendant-charged-in-autism-fraud-scheme

  16. Minnesota Reformer (2025). Federal prosecutors charge first person in Minnesota autism fraud investigation. https://minnesotareformer.com/2025/09/24/federal-prosecutors-charge-first-person-in-minnesota-autism-fraud-investigation/

  17. U.S. Department of Justice, District of Minnesota (2025). Six Additional Defendants Charged, One Defendant Pleads Guilty in Ongoing Fraud Schemes. https://www.justice.gov/usao-mn/pr/six-additional-defendants-charged-one-defendant-pleads-guilty-in-ongoing-fraud-schemes

  18. Minnesota Reformer (2025), citing Minnesota Department of Human Services statement on open provider investigations and site visits. https://minnesotareformer.com/2025/09/24/federal-prosecutors-charge-first-person-in-minnesota-autism-fraud-investigation/

  19. Minnesota Department of Human Services. Early Intensive Developmental and Behavioral Intervention (EIDBI) licensing. https://mn.gov/dhs/partners-and-providers/licensing/eidbi/index.jsp

  20. Minnesota Department of Human Services. EIDBI benefit overview. https://mn.gov/dhs/partners-and-providers/news-initiatives-reports-workgroups/long-term-services-and-supports/eidbi/eidbi.jsp

This post is intended as a general informational overview, not legal, clinical, or policy advice. Minnesota's EIDBI licensing framework is actively under development through 2026–2027; readers should consult the Minnesota Department of Human Services directly for the most current requirements. This information is accurate to the time of this posting and may not be representative of future changes within the EIDBI or the field of autism interventions as policies change and/or new research emerges.

Hillary Duarte