The problem with comparing self-diagnosis and professional diagnosis is that both categories combine a wide variety of different things with hugely varying levels of accuracy.

Self-diagnosis can be thoughtful research, using both the experiences of other neurodivergent people and professional journals and books, taking advantage of your devotion to figuring out your own case and your privileged vantage point about how your own brain works. Or it can be someone taking a PsychCentral quiz and diagnosing themselves with twelve personality disorders.

Professional diagnosis can be someone with years of training and clinical experience across a wide variety of neurodivergences who listens thoughtfully to the patient’s experiences, uses medical tests to rule out possible physical and neurological causes, and understands how the patient’s culture and social experiences affect their neurodivergence. Or it can be a general practitioner saying “are you sad? Tired a lot? Here, have some antidepressants.”

Now, the best professional diagnosis is almost certainly better than the best self-diagnosis. And the worst professional diagnosis is probably better than the worst self-diagnosis, if only because there’s a higher base rate of depression than there is of twelve comorbid personality disorders. But these two categories overlap a lot, and I think a thoughtful, careful self-diagnosis should be considered far more credible than our friend the Indiscriminately Antidepressant-Prescribing GP.

Fun fact: according to one study, 38% of people currently using antidepressants have never met criteria for a disorder that antidepressants are prescribed for. This might be a reasonable state of affairs– after all, antidepressants are pretty safe drugs, depression is awful, and even under current standards about half of depressed Americans get no treatment— but it does suggest that professional diagnosis of depression is not great evidence that you were actually, at any point, depressed.

One thing that often gets conflated with self-diagnosis but should, in my opinion, be thought about separately is peer diagnosis: people who have a particular neurodivergence– particularly if they’re also a member of the community of people who have that neurodivergence– and who know you well suggesting that you have that neurodivergence. One of the biggest problems with self-diagnosis is that just reading the DSM symptoms doesn’t let you know what a neurodivergence looks like in the same way that interacting with a lot of people with that neurodivergence does. You might not know what “often has trouble organizing activities” looks like unless you know a lot of ADHD people. Clinical experience solves that problem, but so does knowing a hell of a lot of autistic people.

The other problem is that some neurodivergences are much, much easier to self-diagnose than others. Gender dysphoria is probably Self-Diagnosis Georg, insofar as the diagnostic criteria are basically:

  1. Says they have gender dysphoria
  2. Repeats that they have gender dysphoria in a very loud and emphatic voice
  3. Gets agitated at the suggestion that they don’t have gender dysphoria
  4. Demands hormones to treat their gender dysphoria, which they have

But even neurodivergences where the criteria aren’t purely self-referential can be pretty easy to diagnose. “Hm, I’m not eating very often because I’m afraid of getting fat. I wonder what I could possibly have.” “Huh, I experienced a traumatic event and now I’m on edge all the time and sometimes things remind me of my trauma and I have flashbacks. What could it be?” Similarly, people who suspect they have depression can find out pretty accurately if they’re depressed by taking the Beck Depression Inventory.

On the other hand, something like autism or borderline personality disorder has more complex symptoms, and someone who isn’t familiar with what they look like can be seriously misled. And some conditions like schizophrenia, while often pretty obvious to an outside observer, tend to leave people unable to figure out that they’re schizophrenic. In those cases self-diagnosis is going to be unreliable.

However, one thing that is usually easy to figure out (schizophrenia aside) is that something is wrong. Once you get beyond the PsychCentral quiz level of self-diagnosis, most people who consider the possibility that they’re neurodivergent are going to have some sort of problem. It might not be the problem they think they have! If someone self-diagnoses as having generalized anxiety disorder, they might actually have another mood or personality disorder, they might be in an abusive relationship or social group, they might be autistic and continually in a state of proto-meltdown, or they might just be under inhumane levels of stress.

I don’t mean to say that it isn’t important to figure out which you have. If you’re anxious, you might want to try exposure therapy, while if you are autistic and in a state of proto-meltdown, you probably want to remove stimuli that cause you to melt down from your environment. However, this does mean that a lot of self-misdiagnosis isn’t a problem of special snowflake fakers: it’s a problem of people who legitimately have something wrong with them but are mistaken about what.

Finally, I found in my own case that just having a professional diagnosis wasn’t very helpful in understanding myself. I had to go through something a lot like a self-diagnosis process– researching my symptoms, understanding coping mechanisms, talking to other people with similar brain issues to mine. So even if you have a professional diagnosis, you might be advised to research your diagnosis yourself or talk to someone who has. It pays off.