The number arrives early and it arrives with confidence. Twenty-five hours a week. Thirty. Sometimes forty. It is usually said in a way that makes questioning it feel like questioning whether you want your child to do well.
We kept adding hours for years, mostly because insurance made it the easiest thing to say yes to. Nobody ever explained where the number came from. So I went and found out, and the answer genuinely surprised me.
The short version
As of 2024, the best available evidence does not show that more hours produce better outcomes. That is not a fringe position or a parent-forum opinion. It is the conclusion of the largest analysis anyone has done on the question.
What the research actually found
In 2024, Sandbank, Pustejovsky and colleagues published a meta-analysis in JAMA Pediatrics looking at exactly this question. They pooled 144 studies covering 9,038 autistic children and tested three different ways of measuring how much intervention a child received:
- Daily intensity. Hours per day.
- Duration. Total days the intervention ran.
- Cumulative intensity. Total hours received overall.
None of the three showed a significant positive association with outcomes when analysed within intervention type. Their stated conclusion is that the findings “do not support the assertion that intervention effects increase with increasing amounts of intervention.” The authors go further and note there is little robust evidence supporting intensive intervention at all, and that the amount should be calibrated to the individual child rather than defaulted to a standard high-intensity recommendation.
This sits alongside Project AIM, the 2023 BMJ review of 252 studies and 13,304 children, which found that adverse events in this literature were “poorly monitored, but possibly common.”
Put those together and the honest summary is: we do not have good evidence that more is better, and we have not been carefully looking for the cost of more.
So where does 40 hours come from?
The number traces back to research from the 1980s, and it has been repeated so many times since that it now gets presented as settled. It is worth knowing that the figure predates most of the evidence base people cite when defending it, and that the 2024 meta-analysis above was specifically designed to test whether it holds up. It did not.
I am not saying the number is arbitrary in origin. I am saying that “this is the standard recommendation” and “this is what the evidence supports” are two different claims, and they get said in the same breath.
Why the number is often about insurance, not your child
This was the part I did not understand for years. Our ABA referral was fast, the pediatrician suggested it, and insurance approved it almost immediately. Meanwhile getting speech therapy approved was a fight, and occupational therapy was worse.
That asymmetry is not a clinical judgment about what my child needed. It is a coverage structure. In many plans ABA sits under a different benefit category with far more generous limits than speech and occupational therapy, which are often capped at a set number of visits per year. When one service is nearly unlimited and the others are rationed, the plan that gets built is the plan the coverage allows, and it can look identical to a plan built around the child.
This is worth checking rather than assuming. Ask your plan directly what the annual visit limits are for speech and OT, and what the limit is for ABA. The gap is often startling once you see it in writing.
The cost nobody put in the plan
Hours are not free even when insurance covers them. My child said this to me in the middle of a week that had no gaps in it:
Can we have a day with no clinics? I miss pancakes.
That is a real cost. So is the drive time, the missed unstructured play, the sibling who gets less of you, the dinner that happens at eight because the session ran late. None of it appears on a progress report, because a progress report measures the session, not the week.
The research on autistic burnout is about adults rather than children, so I want to be careful with it. But it describes a pattern worth knowing: chronic stress plus a mismatch between what is expected and what a person can sustain, without adequate support, producing long-term exhaustion and loss of skills the person previously had. If your child is losing skills at home while gaining them in the clinic, that is not a contradiction to explain away. It is the whole question.
How to ask for fewer hours without a fight
You are allowed to say no. In practice, saying no goes better when it is specific rather than general. These are the framings that got me actual answers instead of a warning.
Ask what the recommendation is based on
“What is this number based on for my child specifically, as opposed to the standard recommendation?” It is a fair question and a good provider will have an answer.
Ask what would be lost
“If we went from 25 hours to 15, what specifically would you expect us to lose, and how would we know?” This forces a concrete prediction instead of a general concern, and it gives you something you can actually check in three months.
Propose a trial, not a permanent change
“Can we try 15 hours for eight weeks and review it against these measures?” Almost nobody refuses a time-limited trial with a review date, and it changes the conversation from a confrontation into an experiment.
Bring the home data
Sleep, eating, meltdown frequency, whether your child still does the things they used to enjoy. Write it down for two weeks before the meeting. Clinic data covers the clinic. You are the only person collecting the rest.
Ask for the hours to move, not just to go
“We would like to reallocate ten of these hours to speech and OT.” That is a different request from “we want less therapy,” and it is harder to characterise as a parent giving up.
What I would tell myself in that meeting
The knot in your stomach during a meeting where everyone is kind and the graphs look impressive is not irrational. It is often you noticing a gap between what is being measured and what you are living with.
None of this means your child should have zero hours, and none of it means your provider is acting in bad faith. Most of the people in those rooms are doing what they were trained to do. It means the number is a starting point for a conversation, not a prescription you are failing to comply with.
The AAP clinical report on autism is worth reading before your next meeting, if only because it makes clear how much depends on the individual child and family. Bring it, bring your two weeks of home notes, and pick one question from the list above.
