ABA therapy for toddlers: When to start and what to expect.

Much of the research on early autism intervention was done with toddlers. The landmark 2010 trial of the Early Start Denver Model enrolled children between 18 and 30 months old, and a 2021 multisite trial comparing teaching styles and weekly intensity enrolled children from 12 to 30 months. If your child is 2 and you are being told about ABA, you are well within the age range these programs were designed for. For the basics of how assessment, goals, and data collection work at any age, start with our guide to what is ABA therapy.

The American Academy of Pediatrics' 2020 clinical report on autism, written by Hyman, Levy, and Myers, explains why age matters. It notes that autism can be diagnosed as young as 18 months and says intervention for communication, adaptive, and behavioral needs should take place as soon as the need becomes evident. It also says intervention is most effective when it is early, intense, and involves the family, which is a useful lens for judging any program you are offered for a toddler.

Many parents start looking into ABA for toddlers because something about their child's communication or play has them worried. CDC lists social communication differences that can show up early, such as not responding to their name by 9 months, using few or no gestures like waving goodbye by 12 months, not sharing interests with you by 15 months, for example by bringing you a toy they like, and not pointing to show you something interesting by 18 months. CDC also notes that an autistic child may not have all or any of the behaviors on its lists, so a single missing skill is a reason to ask questions and not a diagnosis.

The AAP recommends that all children be screened specifically for autism at their 18 and 24 month well-child visits, in addition to general developmental screening at 9, 18, and 30 months. CDC's Learn the Signs. Act Early. program adds that autism screening should also happen whenever a parent or provider has a concern, so you do not have to wait for the next scheduled visit. A common screening tool is the Modified Checklist for Autism in Toddlers (M-CHAT), a questionnaire you fill out about your child. A positive screen means your child should have a full evaluation, while a negative screen does not override a concern you still have.

CDC's free milestone checklists and Milestone Tracker app can help you organize what you have noticed before an appointment, though CDC is clear that these checklists are not a substitute for a standardized, validated screening tool. If you leave the visit still worried, CDC's advice is direct: don't wait, ask for a referral to a specialist, and call your state's early intervention program yourself.

You do not need an autism diagnosis to get help for a child under 3. Part C of the Individuals with Disabilities Education Act (IDEA) funds an early intervention program in every state for infants and toddlers from birth through age 2 who have developmental delays or disabilities, along with their families. Each state's governor names a lead agency to run it, often the health department, and each state writes its own definition of developmental delay, so eligibility rules differ from one state to the next. CDC says plainly that you do not need to wait for a doctor's referral or a medical diagnosis to contact the program, and the AAP report says children should be referred for intervention for any identified delay without waiting for an autism evaluation to take place.

Once the program receives a referral, federal rules give it 45 days to complete the initial evaluation, the initial assessments of your child and family, and the first meeting to write an Individualized Family Service Plan, or IFSP, with some exceptions such as when a family is unavailable. The IFSP is an agreement between your family and the providers that spells out the services and supports your child will receive. CDC describes the evaluation as free, and according to the ECTA Center, services are provided at no cost to families except where federal or state law allows a system of payments, such as sliding fees, which the state has to spell out in its written policies.

Part C services are meant to be provided in natural environments to the maximum extent appropriate, so expect many visits to happen in your home or another everyday setting. The services each state must make available include special instruction, speech-language pathology, occupational therapy, mental health services, and family training, counseling, and home visits. ABA is not named on that list, though the list is not exhaustive, and whether a state's program includes behavior-analytic services varies, so ask your service coordinator directly.

Part C ends at age 3. If your child may qualify for preschool special education, the program must notify your school district and, with your approval, hold a transition conference at least 90 days before the third birthday, and it can happen as early as nine months before if everyone agrees. After that, evaluations and services come through your local school district, and CDC notes you can call any public elementary school to ask for a preschool special education evaluation, even if your child will not attend that school.

Many of the programs studied with toddlers are what researchers call naturalistic developmental behavioral interventions (NDBI). A 2015 paper by Laura Schreibman and researchers behind several of these models defined them as interventions delivered in natural settings, with shared control between child and therapist, natural rewards, and a range of behavioral strategies used to teach developmentally appropriate skills. Named models include the Early Start Denver Model, JASPER, Project ImPACT, and pivotal response treatment. In practice, the therapist follows your child's interest in a toy, sets up a reason to communicate, such as holding the bubbles until your child looks, gestures, or tries a sound, and then makes the fun continue as the natural reward.

Goals for a toddler usually center on the building blocks of social learning. Typical targets include imitating actions and sounds, sharing attention with an adult around a toy or picture, requesting with gestures, signs, words, or pictures, playing back and forth, and getting through daily routines like meals, diaper changes, and bedtime with less distress. Schreibman and colleagues describe these as developmentally appropriate and prerequisite skills, meaning the early abilities that later learning builds on, and in a good program they are taught inside activities your child already enjoys.

Caregiver coaching is a large part of early intervention ABA. Schreibman and colleagues reported that randomized trials suggest adding parent coaching speeds developmental progress, and they point out that parents can use these strategies during meals and bath time, when a therapist is not around. In the Early Start Denver Model trial, therapists delivered about 15 hours a week while parents reported using the strategies for about 16 more hours a week during daily life. The Council of Autism Service Providers (CASP) also says caregivers should be involved as much as is feasible, while adding that a child's access to ABA should not depend on whether parents can deliver treatment themselves.

Structured teaching such as discrete trial training can still have a place, especially for practicing a skill a child is close to mastering, but a toddler's day should not be built around it. If you are weighing where sessions should happen, our comparison of in-home vs center-based ABA covers the trade-offs.

Early intensive behavioral intervention, or EIBI, is the name researchers use for comprehensive ABA delivered to young children for 20 to 40 hours a week, usually over multiple years. The best-known summary is a 2018 Cochrane review by Reichow and colleagues. It found only five studies with 219 children who started treatment at roughly 30 to 42 months old, and only one of those studies randomly assigned children to groups. Compared with generic special education, children in EIBI groups showed better adaptive behavior, IQ, and expressive and receptive language at the end of treatment. The review found no clear evidence of improvement in autism symptom severity or problem behavior, and it rated the overall quality of evidence as low to very low because the studies were small and mostly not randomized.

The Cochrane authors concluded that the evidence supports EIBI for some children with autism while urging caution, and they called for more rigorous research. The children in that review were about two and a half to three and a half years old at the start, so it says little directly about younger toddlers. For toddlers specifically, the strongest single trial is the 2010 Early Start Denver Model study by Dawson and colleagues in Pediatrics, which randomly assigned 48 children aged 18 to 30 months to two years of the program or to community services. Children in the program gained an average of 17.6 points on a standard developmental measure compared with 7.0 points for the comparison group, and they held steady on adaptive behavior while the comparison group's scores declined. The study did not find a difference between groups on its measure of autism severity.

Large reviews of the whole field add caution. In a 2023 BMJ update of the Project AIM meta-analysis, Sandbank and colleagues pooled 252 studies of more than 13,000 autistic children younger than 8. Among randomized trials, naturalistic developmental behavioral interventions showed small positive effects on several outcomes, including social communication, language, and play, and behavioral interventions showed an effect on social-emotional and challenging behavior outcomes. Once the authors set aside outcomes reported by caregivers or teachers and those at high risk of biased measurement, only one effect remained, for naturalistic developmental behavioral interventions on measures of autism characteristics. They also found that adverse events were poorly monitored but possibly common, which means research on early intervention has rarely tracked harms carefully.

Recommendations for toddlers vary, and you will hear different numbers depending on who you ask. CASP's practice guidelines say a comprehensive ABA program for a young child should average 30 to 40 hours a week of direct intervention for two or more years, and its 2025 paper on treatment intensity argues that more intensity produces better outcomes when goals are comprehensive. Focused programs aimed at a few specific goals typically use fewer hours. CASP is a trade group of ABA providers, so its guidance reflects the provider side of this debate, and our guide on how long ABA therapy lasts explains the focused and comprehensive categories in more detail.

Research with toddlers has complicated the picture. In a multisite randomized trial published in 2021, Sally Rogers and colleagues assigned 87 autistic toddlers with an average age of 23.4 months to either a naturalistic developmental behavioral approach or discrete trial teaching, delivered for either 15 or 25 hours a week of one-to-one intervention over 12 months, and every family also received regular caregiver coaching. Neither the teaching style nor the number of hours had an overall effect on progress in autism symptoms, expressive communication, receptive language, or nonverbal ability. CASP has criticized other studies that found little effect of intensity, so the question remains open, and our guide to criticism of ABA covers the debate over high-hour programs in more depth.

For your family, the practical point is that the right schedule should come out of an assessment of your child, and the behavior analyst should be able to explain how they arrived at it. A toddler also naps, eats, plays with siblings, and needs downtime, so ask how sessions fit around your child's day and how the plan will change if your child is tired, upset, or making faster progress than expected. Time you spend using the strategies during meals, baths, and play also gives your child learning opportunities, even though it never appears on an authorization.

Part C early intervention is a public program that runs separately from your health insurance, although the two can overlap. A state may ask to bill your private insurance for Part C services, but federal rules say it generally needs your written consent first, and it needs consent again if the frequency or intensity of services increases. If you are found unable to pay and do not consent, the lack of consent cannot be used to delay or deny services. The state also cannot require you to enroll in Medicaid as a condition of getting early intervention.

ABA delivered by a clinic is usually paid for by health insurance or Medicaid, and those benefits generally require an autism diagnosis and a treatment plan before they pay. Many states have state autism insurance laws that require state-regulated plans to cover autism treatment. For children enrolled in Medicaid, a 2014 informational bulletin from the Centers for Medicare and Medicaid Services clarified that the EPSDT benefit requires states to cover medically necessary services for autistic children under 21, and that a child whose screening shows a need should be referred for diagnosis and treatment without delay. Our guide to Medicaid EPSDT coverage for ABA explains how that works in practice, and our breakdown of ABA therapy cost covers what families pay out of pocket. You can start Part C services while you wait for a diagnostic evaluation, and once a diagnosis comes through, your service coordinator can help Part C services and insurance-funded ABA work together without duplicating goals.

• How many of your clients are under 3, and what does a typical session with a 2-year-old look like?
• Which approach do you use with toddlers, and how much of the day is play-based versus structured table work?
• How will you coach me to use the strategies at home, and how often will the supervising behavior analyst see my child and me together?
• How did you arrive at the recommended weekly schedule, and what would make you increase or decrease it?
• How will you tell whether my child is enjoying sessions or wants a break, and what do you do when they do not want to take part?
• Will you coordinate with our Part C service coordinator, speech therapist, and pediatrician?
• If challenging behavior comes up, will you do a functional behavior assessment before changing the plan?

If behavior like head-banging, biting, or long meltdowns is part of the picture, a functional behavior assessment is how the team figures out what the behavior is doing for your child. Our guide on how to start ABA therapy walks through referrals, intake, and authorization. Ask who will supervise the case and who will run sessions, and see our guide to BCBA vs RBT for what each credential means. For a longer checklist, see how to choose an ABA provider, and when you are ready to compare options near you, you can find a clinic in our directory.

• Hours of table work or rapid-fire drills for a young toddler, with little time spent playing on the floor or in daily routines.
• No real role for you, such as a clinic that discourages you from watching sessions or offers no caregiver coaching.
• A weekly schedule set before anyone has assessed your child, or the same number of hours proposed for every family.
• Goals aimed mainly at making your child look less autistic, such as forcing eye contact or stopping harmless stimming, instead of goals that help them communicate and take part in family life.
• Ignoring your child's signs of distress or unwillingness, such as crying, turning away, or trying to leave, and pushing through anyway.
• Using restrictive or punishment-based procedures before less intrusive approaches have been tried.

The Behavior Analyst Certification Board's ethics code backs several of these points. It requires behavior analysts to involve clients and stakeholders in selecting goals and interventions, to prioritize positive reinforcement, and to use restrictive or punishment-based procedures only after less intrusive means have not worked or when the risk of harm from a behavior outweighs the risk of the procedure. It also makes behavior analysts responsible for obtaining assent when applicable, which the code defines as vocal or nonvocal behavior that can be taken to indicate willingness to take part from someone who cannot give informed consent, such as a toddler.

Many autistic adults who received ABA as children have described programs built around compliance and hiding autistic traits, and some describe lasting harm. Those concerns are a good reason to choose a toddler program that follows your child's interests, respects their signals, and keeps you closely involved. Our guide to criticism of ABA covers what autistic advocates object to and how the field has responded.