TL;DR

• One in four autistic adults, and one in three autistic women, have been given at least one psychiatric diagnosis before autism that they now consider a misdiagnosis.
• Personality disorders, anxiety, depression, and chronic fatigue or burnout-related conditions are the most common stand-ins.
• Camouflaging (also called masking) delays diagnosis, especially for women, and race and ethnicity shape the timeline in their own ways.
• What’s almost always missing from the earlier diagnoses: any real look at the nervous system and how it processes sensory information.
• Misdiagnosis isn’t neutral. It means years of the wrong treatment while the real source of distress goes unaddressed, and autistic adults already face significantly elevated rates of suicidal ideation and attempts.

Ellie (a composite of clients I’ve worked with, not any one person) was 34 the first time anyone said the word autism to her. By then she’d already collected a small stack of other labels. Generalized anxiety disorder at 19. Major depressive disorder at 24. A brief, unsettling stint with a borderline personality disorder diagnosis at 29, given after a particularly hard year, that she carried around like a verdict for nearly half a decade.

None of it ever quite fit. The anxiety medication helped a little, then stopped helping. The depression lifted and came back on its own schedule, not really tied to anything in her life. The personality disorder diagnosis, she told me, felt less like an explanation and more like an accusation, as though her own nervous system were a character flaw.

By the time she found her way to an autism assessment, she’d spent fifteen years in rooms with well-meaning clinicians who were looking at the right symptoms and reaching for the wrong explanation.

Ellie’s story isn’t unusual. It’s closer to the norm.

The scale of the problem

Research following adult autism diagnoses found that one in four autistic adults, and one in three autistic women, report receiving at least one psychiatric diagnosis before autism that they now consider to have been a misdiagnosis (Kentrou et al., 2024). Personality disorders topped the list, followed by anxiety disorders, mood disorders, and chronic fatigue or burnout-related conditions.

The gender gap here is striking. Women in the study reported a prior psychiatric diagnosis of some kind at a rate of 65.8%, compared to 34.2% for men, and women were specifically more likely to walk away with a personality disorder, anxiety disorder, or mood disorder label before anyone considered autism as the underlying explanation.

That’s not a small discrepancy. That’s an entire generation of autistic women who spent years, sometimes, decades, being treated for the symptom instead of the source.

Why the wait is longer for women

Globally, autism is identified in males roughly four times more often than in females (Zeidan et al., 2022). Some of that gap is almost certainly diagnostic bias built into research and assessment tools developed primarily around how autism presents in boys. But part of it is also something autistic women often do without even realizing they’re doing it: camouflaging.

Camouflaging, or masking, is the conscious or unconscious effort to mask autistic traits well enough to move through a neurotypical world without standing out. Eye contact that’s been practiced and rehearsed. Conversational scripts memorized and deployed on cue. Stimming saved for the car, or the bathroom, or the hour after everyone else has gone to bed. Research shows a clear relationship between higher camouflaging and later age at diagnosis, and this relationship is notably stronger for women than for men (Milner et al., 2024). The better someone has learned to perform neurotypical, the longer it tends to take for anyone, including the clinicians trained to notice, to see what’s underneath the performance.

I want to be direct about something here: this isn’t a personal failing. Masking is rarely a conscious choice in the way it sounds. It’s a survival strategy, often built in childhood out of necessity, that the world never recognized as one.

Race and ethnicity shape the timeline, too

Diagnostic delay doesn’t land evenly across every population. Research has found a more pronounced diagnostic lag specifically for autistic females within certain ethnic groups, even as racial and ethnic minority youth overall were, in some cases, diagnosed earlier than non-minority youth, a pattern the researchers attribute to phenotypic differences in how autism presents across populations (Goldblum et al., 2023).

For BIPOC autistic adults more broadly, the research points to compounding barriers: clinician bias, cultural assumptions layered on top of autistic traits, and a justified wariness of mental health systems that haven’t always served these communities well. The result is a population whose true rate of autism, and whose true level of unmet need, is almost certainly undercounted in the data we currently have.

What almost always gets missed: The nervous system underneath

Here’s the piece I think about constantly in my own work, and the piece that’s rarely part of the conversation when we talk about misdiagnosis.

Anxiety, depression, and personality disorders are typically diagnosed by looking at mood, behaviour, and relational patterns. What goes unexamined, again and again, is the nervous system itself; specifically, how it processes sensory information.

When I sit down with a late-diagnosed client, one of the first things I assess is sensory processing, because it so often explains what years of other diagnoses couldn’t. Autistic nervous systems frequently process sound, light, touch, movement, and internal body signals (interoception) differently than a neurotypical nervous system does. Not better or worse, differently. A fluorescent light that’s mildly annoying to one person can be genuinely dysregulating to an autistic adult. The hum of an air conditioner, a scratchy clothing tag, a crowded grocery store at 5pm; these aren’t quirks or preferences. They’re nervous system events, and the body responds to them as such.

When that goes unrecognized, the resulting exhaustion, irritability, shutdowns, or avoidance get reinterpreted through the only lens most clinicians are trained to use first: mood and anxiety. Ellie’s anxiety diagnosis wasn’t exactly wrong, she was anxious. But the anxiety was downstream of a nervous system working in overdrive in environments that were never built to accommodate it, and no amount of talk therapy aimed at her thought patterns was ever going to address that root cause.

This is part of why I think a proper sensory assessment matters so much when an adult is exploring a late autism diagnosis. It isn’t a box to check. It’s often the first time someone has the language to describe an experience that’s been misnamed, sometimes for their entire adult life.

The cost of getting it wrong

This isn’t an abstract conversation about diagnostic precision. Autistic adults experience suicidal ideation at a rate of 34.2% and suicide attempts at a rate of 24.3%, both substantially higher than the general population, with risk of death by suicide running as high as eightfold (Brown et al., 2024). The longer someone goes undiagnosed or misdiagnosed, the longer they’re moving through life, and through treatment, without an accurate framework for what they’re actually experiencing.

Misdiagnosis isn’t a neutral delay. It’s years of the wrong kind of support, while the actual source of distress continues, mostly unaddressed, underneath whatever label has been applied on top of it.

If this sounds like your own story

If you’ve spent years being treated for anxiety, depression, or a personality disorder, and something about the treatment never quite landed, you are not alone, and you are not imagining the mismatch. Ellie’s story is one version of a pattern that shows up constantly in this research, and constantly in my practice.

A correct diagnosis, even one that arrives at 34, or 44, or later, often brings something close to relief. Not because a label fixes anything on its own, but because it finally offers a framework, sensory experience included, that explains a lifetime instead of just naming a symptom.

References

Brown, C. M., Newell, V., Sahin, E., & Hedley, D. (2024). Updated systematic review of suicide in autism: 2018-2024. Current Developmental Disorders Report, 11, 225-256. https://doi.org/10.1007/s40474-024-00308-9

Goldblum, J. E., McFayden, T. C., Bristol, S., Putnam, O. C., Wylie, A., & Harrop, C. (2024). Autism prevalence and the intersectionality of assigned sex at birth, race, and ethnicity on age of diagnosis Journal of Autism and Developmental Disorders, 54(10), 3777-3791. https://doi.org/10.1007/s10803-023-06104-5

Kentrou, V., Livingston, L. A., Grove, R., Hoekstra, R. A., & Begeer, S. (2024). Perceived misdiagnosis of psychiatric conditions in autistic adults. eClinical Medicine, part of The Lancet, 71. https://doi.org/10.1016/j.eclinm.2024.102586 

Milner, V., Colvert, E., Hull, L., Cook, J., Ali, D., Mandy, W., & Happé, F. (2023). Does camouflaging predict age at autism diagnosis? A comparison of autistic men and women. Autism Research, 17, 626-636. https://doi.org/10.1002/aur.3059

Zeidan, J., Fombonne, E., Scorah, J., Ibrahim, A., Durkin, M. S., Saxena, S., Yusuf, A., Shih, A., & Elsabbagh, M. (2022). Global prevalence of autism: A systematic review update. Autism Research, 15(5), 778-790. https://doi.org/10.1002/aur.2696