Why CBT May Not Work for Autistic and ADHD Adults

Have you ever been told that you're treatment resistant?

Maybe you've tried therapy. You've tried cognitive behavioral therapy, or CBT. You've learned coping skills, challenged your thoughts, completed worksheets, and tried to do everything you were supposed to do.

And it still didn't seem to work.

Over time, the message can start to feel like:

Maybe I'm just difficult to treat.

If you're autistic, have ADHD, or are AuDHD, there may be another explanation.

Sometimes the person isn't resistant to treatment.

Sometimes the treatment model doesn't fit the person.

I want to make something clear before going any further: I like CBT.

CBT is an evidence-based approach, and I have used its principles throughout my career. The problem is not CBT itself.

The problem is what happens when CBT becomes rigid, standardized, and disconnected from the actual nervous system, goals, and life of the person sitting in the therapy room.

Does CBT Work for Neurodivergent People?

There isn't a simple yes-or-no answer.

CBT isn't one single treatment delivered exactly the same way every time. It is a broad therapeutic framework that can be adapted to different people and different problems.

And therapy outcomes in general are complicated.

For example, a large meta-analysis by Fernandez and colleagues examined 115 CBT studies involving 20,995 participants across a range of mental health conditions. The researchers found that approximately 15.9% dropped out before treatment really began, while 26.2% dropped out after beginning treatment (Fernandez et al., 2015).

That study was not specifically about autistic or ADHD clients, so we shouldn't use those numbers to claim that neurodivergent people leave CBT at those rates.

But it illustrates an important point:

Therapy does not work simply because someone walked into an office and received an evidence-based treatment.

Treatment format matters.

The problem being treated matters.

Life circumstances matter.

And the fit between the therapist, the treatment, and the person matters.

If CBT didn't work for you, that does not automatically mean you are "treatment resistant."

It may be worth asking why it didn't work.

Why Can CBT Feel Like It Fails Autistic and ADHD People?

Over the years, I've noticed several recurring problems when standard CBT is applied too rigidly to neurodivergent clients.

1. Therapy Starts With the Thought When the Problem Starts With Overload

A common CBT sequence is:

"What's the thought you're having?"

Then:

"Is that thought accurate?"

And finally:

"How could we challenge or reframe that thought?"

There is nothing inherently wrong with that.

But sometimes we are starting in the wrong place.

An autistic person may already be dealing with sensory overload, social exhaustion, masking fatigue, unexpected changes, or burnout.

A person with ADHD may arrive after spending an entire day managing attention, executive functioning, deadlines, interruptions, and competing demands.

Their nervous system may already be at capacity.

And now we're asking them to complete a perfect thought record.

At a certain point, that can feel like asking somebody to solve a math problem while they're drowning.

It isn't that thoughts don't matter.

They do.

But the state the person is in matters too.

Sometimes we need to address overload and regulation before asking someone to engage in complicated cognitive work.

2. CBT Becomes a Script Instead of a Tool

Another problem occurs when CBT becomes something the therapist performs at the client.

Identify the cognitive distortion.

Challenge it.

Choose a coping strategy.

Complete the worksheet.

Repeat next week.

That might work beautifully for some people.

For others, it won't.

A neurodivergent client may need more concrete language. They may need visual information or explicit examples. They may need executive-function supports built into therapy. Sensory factors may need to be considered. Homework may need to be changed so the person can realistically complete it.

Most importantly, treatment goals should come from the person's actual life.

The goal shouldn't be:

"How do we make you act more normal?"

The better question is:

"What is making your life harder, and what would you actually like to be able to do?"

3. The Therapist Doesn't Accurately See the Person

This part does not get enough attention.

Therapy is going to be much harder if you consistently feel misunderstood.

If you feel judged, corrected, or as though your therapist is trying to train the neurodivergence out of you, your guard is probably going to stay up.

A therapist might interpret sensory overload as avoidance.

They may see fidgeting as a behavior that needs to stop rather than a regulation strategy.

They may interpret direct communication as hostility.

They may focus on eliminating behaviors because they look unusual instead of asking whether those behaviors are actually causing a problem.

Before we change something, we need to understand what it is doing for the person.

What Adapted CBT Can Look Like for an Autistic Teen

Years ago, I worked with a teenager who had autism and ADHD.

He was a really sweet kid, and his therapy goal was remarkably straightforward:

"How do I make a friend?"

His goal wasn't to reduce an anxiety score on a questionnaire.

He wanted to know:

How do I talk to people?

How do I initiate a conversation?

How do I keep a conversation going?

How do I turn that into an actual friendship?

He was quite formal when he spoke. He rarely asked questions, and reciprocal conversation was difficult for him.

It wasn't because he didn't care about other people.

He didn't have a structure for what to do.

So we gave him one.

Making Therapy Concrete

We practiced while playing games.

Sometimes it was cards. Sometimes checkers. Sometimes chess.

I added a rule.

After certain turns or moves, he had to ask me a question.

But I didn't make it a complete free-for-all.

If generating a question felt difficult, I had cards with questions already written on them.

Maybe:

"What's your favorite thing to do on vacation?"

Or:

"What do you think about kids playing a lot of video games?"

The content itself wasn't terribly important.

The point was to make reciprocal conversation concrete and practiceable.

He would say something.

I would respond.

Then he would respond to what I said.

Then I would respond again.

Now we had a four-part reciprocal interaction.

Instead of telling him:

"You need to improve your social skills,"

we were practicing an observable skill that we could actually measure over time.

Turning Anxious Thoughts Into Behavioral Experiments

There was still traditional CBT involved.

He would think things like:

"Nobody is going to want to do what I want to do."

"The people at my lunch table think I'm weird."

"If I start a conversation, I'm going to get rejected."

But instead of spending all of our time debating those thoughts in my office, we turned them into behavioral experiments.

I asked him what he liked doing socially.

He liked bowling.

Great.

But we didn't immediately start with:

"Go ask everybody at lunch to go bowling."

That was too big.

First, he practiced making a comment at the lunch table.

Then he practiced asking a question.

Once that became more comfortable, he worked on initiating conversations.

Eventually, he asked:

"Do you guys want to go bowling?"

And the first attempt didn't really work.

Not because he'd failed.

People had scheduling conflicts.

Someone could go. Other people couldn't. Then another time, he couldn't make it himself.

So we tried again.

Eventually, they went.

Over time, he became much more capable of initiating and sustaining conversations.

And eventually, he didn't need therapy anymore.

I worked myself out of a job.

That's the goal.

He didn't become a different person.

Therapy became useful because it matched the person, the skills he needed, and the goals he actually cared about.

What Does the Research Say About CBT for Autistic Adults?

A 2026 systematic review and meta-analysis gives us an important reality check.

Perez-Castilla and colleagues reviewed 10 randomized controlled trials involving 537 autistic adults.

For most outcomes, the evidence was considered very uncertain. The studies were generally small, many had methodological limitations, and the results varied considerably across studies.

The review found moderate-certainty evidence that CBT probably reduced alexithymia in the short term. Alexithymia refers primarily to difficulty identifying and describing one's own emotions.

For outcomes including anxiety, depression, social anxiety, functioning, obsessive-compulsive symptoms, overall mental health symptoms, and quality of life, the evidence was much less certain (Perez-Castilla et al., 2026).

That doesn't mean:

"CBT doesn't work for autistic adults."

That's too strong.

It means we do not currently have evidence supporting one universal answer about how well CBT works across all of these outcomes for autistic adults.

And that makes individualized treatment even more important.

The Question Isn't Just "Does CBT Work?"

A better set of questions might be:

Does this version of CBT work for this person?

Is it targeting the problem they actually want help with?

Has it been adapted to their communication style and executive functioning?

Is sensory overload being considered?

Can they realistically use the strategies outside of therapy?

Do they feel understood by their therapist?

That is very different from simply handing everyone the same worksheet.

What About Radically Open DBT?

Another treatment you may encounter is radically open dialectical behavior therapy, usually called RO DBT.

RO DBT focuses on difficulties related to overcontrol.

Overcontrol can involve patterns such as excessive self-monitoring, rigidity, emotional inhibition, difficulty with spontaneity, social withdrawal, and intense concern about making mistakes.

Some people may recognize themselves in that description, including some neurodivergent people.

Someone might constantly think:

"Am I saying the right thing?"

"Did I do that correctly?"

"Did I sound weird?"

Then they go home and replay the conversation repeatedly.

RO DBT may be worth discussing when overcontrol is genuinely part of the clinical picture.

But I would not describe RO DBT as the treatment for autism or ADHD.

One of its largest randomized trials involved 250 adults with refractory depression. Participants receiving RO DBT plus treatment as usual showed greater improvement in depressive symptoms immediately after treatment, but the difference between groups was no longer statistically significant at the 12- or 18-month follow-ups (Lynch et al., 2020).

So this is another example where fit matters.

The question isn't:

"What therapy is best for every neurodivergent person?"

There isn't one.

The question is:

"What treatment fits the difficulties this individual person is experiencing?"

Red Flags When Looking for a Neurodivergent-Affirming Therapist

There are a few statements I would be cautious about hearing from a therapist.

"I treat everybody the same."

That can sound fair, but good therapy should be individualized.

"Therapy is therapy."

The basic principles may be similar, but how those principles are applied should change depending on the person.

"The goal is to make you act more normal."

That isn't a treatment goal I would endorse.

The goal should be improving a person's functioning, relationships, autonomy, emotional well-being, or quality of life, based on goals that actually matter to them.

I would also be cautious with exposure-based interventions when no distinction is being made between anxiety-driven avoidance and genuine sensory overwhelm.

Exposure can be extremely useful when it is appropriate.

But the plan needs to make sense for the problem being treated.

And if a therapist automatically tells someone to stop fidgeting, pacing, stimming, or using another regulation strategy, I would want to know why.

The relevant question should be:

Is this causing a meaningful problem?

Not:

Does this look unusual?

Green Flags in Neurodivergent Therapy

Here are the kinds of things I like hearing:

"Let's adapt this to your nervous system."

"Let's make this more concrete."

"What would actually make your life better?"

"Let's build something you can still use when you're overwhelmed."

"That strategy isn't working. Let's figure out why."

Those statements reflect flexibility and collaboration.

And those qualities matter.

If CBT Didn't Work, Are You Treatment Resistant?

Maybe. But don't jump to that conclusion.

Instead, ask what actually happened.

Did the therapist understand autism or ADHD?

Did you feel understood?

Were your sensory needs considered?

Were executive-function difficulties considered?

Could you realistically complete the assignments between appointments?

Were your therapy goals actually important to you?

Was treatment adapted when you repeatedly struggled with part of the protocol?

Was the therapist helping you build useful skills?

Or did therapy mostly consist of someone repeatedly telling you that you were thinking about things incorrectly?

Sometimes a treatment legitimately doesn't work.

But sometimes the wrong treatment was delivered in the wrong way to the wrong person.

Those are very different conclusions

Neurodivergent-Affirming Therapy in Portland, Oregon

At Forest Psychological Clinic, we work with autistic, ADHD, AuDHD, and other neurodivergent children, teens, and adults.

We offer therapy, comprehensive autism and ADHD evaluations, and medication management in the Portland, Oregon area.

Our goal isn't to force every person into the same treatment protocol.

We want to understand what is actually happening, identify goals that matter to the individual, and develop practical strategies and supports that fit the person's real life.

If you're looking for neurodiversity-affirming therapy, autism or ADHD testing, or medication management in Portland, Oregon, you can learn more at ForestPsychologicalClinic.com.

Therapy Should Fit the Person

CBT isn't the enemy.

Rigid therapy is the problem.

CBT can give therapists and clients an incredibly useful set of tools.

But tools only help when they're being used for the right problem and in a way the person can actually access.

If you're autistic, have ADHD, or are AuDHD and therapy hasn't worked before, don't automatically interpret that as evidence that you're impossible to help.

Maybe the therapy needs to become more concrete.

Maybe sensory overload needs to be taken seriously.

Maybe executive functioning needs to become part of the treatment plan.

Maybe the goals need to change.

Maybe your therapist needs to slow down and actually understand your experience before trying to change it.

Therapy should adapt to the person who walks through the door, rather than forcing that person to adapt to the protocol.