Why Is It So Difficult to Diagnose Adults with Autism?

The importance of an accurate diagnosis for adults.

Posted Jan 01, 2019

On the first day of 2019, I found myself reading a thought-provoking opinion piece by Sara Luterman, published on Spectrum News. Sara is the founder of NOS magazine, which focuses on neurodiversity. Sara was diagnosed with autism in her early 20s, and her piece focuses on the issues that arise when adults seek a diagnosis for the first time. 

As a clinician, I know this frustration well. We often get calls from adults who are seeking a first-time diagnosis. This presents a diagnostic conundrum. We want to help, but often feel as though our hands are tied. The biggest issue stems from the diagnostic criteria for autism from the DSM-5. Criteria C states:

"Symptoms must be present in the early developmental period (but may not become fully manifest until social demands exceed limited capacities or may be masked by learned strategies in later life)."

I usually consider the "early developmental period" to be prior to 7-years-old, give or take. This leads to the crux of the issue: how do we know whether symptoms were present prior to age 7 in an adult, whose parents may be deceased, no longer involved in their health care, or unable to remember details of their child's behavior decades later? I personally don't remember much of my life or behaviors prior to age 7, especially not about details related to my social development. 

A related issue is that adults who are looking for a first-time diagnosis are often what we used to call "high functioning," and what most of us would now consider "mildly affected" with ASD. Individuals who did not get diagnosed as children are rarely the "clear-cut" cases. If they were, they likely would have been identified as having some kind of developmental disability as children. This only muddies the diagnostic waters, as the symptoms of ASD in these individuals are often subtle, and/or the adult has learned compensatory strategies for navigating the social landscape. 

However, the above issues don't negate the importance of getting an accurate diagnosis, or mitigate the frustration felt by individuals who are repeatedly turned away from diagnostic clinics because "we only see children." 

We, as clinicians, need to do better. We need to figure out how to help diagnose adults for the first time. Although far from ideal, I have started doing the following:

1. Asking adults who call the clinic whether they have anyone who could speak to their childhood behaviors (e.g. siblings, childhood caretakers, etc). This helps widen the scope of people who can tell us about childhood beyond parents. 

2. I try to be honest with adults about the difficulties involved in providing a first time diagnosis without reliable information about childhood behaviors. Especially if there is no one who can report on their developmental behaviors, I explain that I can see them and do a diagnostic screening, but I likely cannot officially "diagnose." What I can do is help them understand their own behaviors and history of social difficulties by giving a "provisional" or "probable" diagnosis. Though this doesn't help the adult receive services, it can be helpful for understanding why they experienced difficulty making friends, or with social interactions. 

The above isn't a great solution, and has multiple flaws, but it's the best I've come up with thus far. One of my hopes for the new year is that there are more services and supports for adults with autism and that diagnosis can improve for these individuals. 


Luterman, S. (2018, December 18). Why adults need an easier path to autism diagnosis. Retrieved from