Adult Autism Insights

When autism isn’t coded (specifically ICD-10-CM and the emerging ICD-11) into autistic adults’ medical charts, that can affect health care. It leads to clinical invisibility and puts health at risk. Especially sleep medicine and GI care. It also affects medication safety, communication accommodations, and population studies.

Providers will say, “I’m not treating autism.” Well, no, they likely aren’t. But does that mean that autism isn’t relevant?

Why Adult Autism Medical Coding Gaps Hurt Patient Care

Example: Sleep studies. Sleep issues are a common issue for autistic people. But sleep doctors rarely code it. Adult sleep research already suggests autistic adults and those with ADHD take longer to fall asleep, have poorer sleep efficiency (the actual time spent asleep while in bed), more awakenings, and poorer perceived sleep quality.

If it isn’t coded on the medical chart, how can researchers easily test whether autistic adults get “insufficient sleep time” in sleep studies (at the lab) more often? I got less than an hour of sleep for my sleep study and got inconclusive results. It would be nice to know if this is common for autistic adults. Guess what? There’s no research on this.

Without accurate coding, researchers cannot determine whether a patient’s inability to sleep in a laboratory setting is a procedural exception or a clear pattern of their neurodevelopmental profile.

Because more kids have Medicaid, IEPs, and disability services, researchers have more ways to identify children.

Adults often no longer have the school/IEP/early-intervention infrastructure attached to them. If they are not on Medicaid, Medicare, disability services, or in a specialty autism clinic, they can be much harder to find in data since doctors often don’t code for it.

I know diagnoses in the chart aren’t always final or correct. And even with medical charts, some people will still be left out. It misses uninsured adults, undiagnosed adults, people whose doctors do not code autism, and people whose autism is only noted informally.

Fixing Medical Charts for Autistic Adults

For people who insist that doctors must follow the coding guidelines, they could document autism as a confirmed history or active neurodevelopmental condition when it affects the encounter, even if autism (ICD-10-CM F84.0) is not the treatment target. However, doctors will claim they know nothing about autism and continue not to code it.

So, I say break the rules as they do for other things. I don’t mean insurance fraud or inaccurate documentation. I mean breaking the rigid habit of only coding the primary complaint and instead using the existing system to document full medical histories. Happens all the time. Look at your medical chart over the years. Did the doctor treat everything they coded that day? Did another doctor make any diagnoses, and isn’t your doctor treating it at all?

My now ex-PCP never coded gastroparesis, and I would have to remind him every visit that I have it. Sure, he’s not a GI doctor.  He wasn’t “treating gastroparesis,” but if I’m prescribed or already taking medications that slow GI motility, then delayed gastric emptying can be relevant to side effects, weight, nausea, whether a new medication is a bad fit, and more.

A condition can be relevant even if the doctor is not currently treating it, because it changes risk, test validity, medications, or patient instructions.

Why Doctors Must Code Adult Autism

Adult autism is often treated in routine medical visits as irrelevant, unless the clinician is directly providing autism-related services. But that misses how autism can affect sensory tolerance, sleep testing, GI symptoms, adherence, and the validity of procedures such as in-lab sleep studies.

So, when autism is not documented or coded, autistic adults become harder to identify in medical records and claims research, which can warp what we know about adult outcomes. Detailed records enable proper medicine. It’s not just an optional administrative duty.

By mandating comprehensive neurodevelopmental documentation and leveraging EHR coordination, we can move toward a healthcare system that sees the patient as a whole. Z codes or secondary ICD-10-CM codes can be used to reflect the patient’s whole clinical context.

Maybe autistic adults and nonautistic adults get inadequate sleep data at the sleep lab at the same rate. It’s possible. But how do we know? Is it just the “first-night effect” (the natural tendency to sleep poorly in an unfamiliar lab environment)?  Or could it be related to autism? Who knows?