What are the most common psychological diagnoses in the tech field?And how common are they? Continuing in my effort to use people in the tech industry as a stand-in for nerds, we’re diving back into the Open Sourcing Mental Illness data set. Previously, we looked at the number of people had a disorder, who had a diagnosis, and who had gotten treatment, and most of what we saw was good news. This time, I wanted to dig in to the actual disorders and diagnoses that we see in tech, and see how much this might differ from the rest of the world.
What I love about writing for a blog versus writing for a journal article is that I can skip all the data cleaning and summaries and whatnot. I’m not going to lie, the primary purpose of this post is to give me some practice making sunplot diagrams, so voila!
I created this using the D3partitionR package, and I tell you no lie when I say that one of the first things I did when I made this is show my mom so she could tell me it was pretty.
My mom’s praise aside, this requires some explanation. After several hours of pounding my head into the computer monitor, I decided I wasn’t going to be able to figure out how to change the legend title to match the question responses I was looking at, so I’ll just have to tell you here.
- The innermost circle is just a circle, enjoy.
- The next ring out is the response to the question: “Do you have a mental illness?” (Yes, No, Maybe)
- The next ring out from there is the specific diagnosis they reported.
- The next ring out from there is for people who have multiple diagnoses, the number of diagnoses.
You can click on areas of the ring to zoom in, for instance to see the difference between the “Yes” and “Maybe” folks. We’re going to tear this apart layer by layer, like an onion.
Prevalence of Any Mental Illness
Unlike in the last post, I focused on current mental illnesses, which is probably most relatable to overall one year prevalence. In the general population, over the course of a year, approximately 1 in 5 adults experiences a mental illness. This includes illnesses that can occur and fade over the course of a few months, like Adjustment Disorder, and lifetime diagnoses like Schizophrenia. Of the people that filled out this survey, 42% have a diagnosed mental illness and a further 22% think they do (or answered maybe). So does that mean that 3 out of every 5 tech workers has a mental illness?
I doubt it. This is probably just heavy, heavy response bias. This survey wasn’t mailed out or distributed widely across the tech field; they depended on people stumbling across it on their website devoted to mental health. It’s like posting the survey outside of a psychiatrist’s office, you’re probably going to get mostly psychiatric patients to fill it out.
The yes/maybe split is pretty interesting. In the human world, more than half of all mental illness is undiagnosed and untreated. In the tech world, it looks like it’s more like 35%. Now, is that another artifact of response bias? Depends on whether you think that people who have a diagnosed mental illness are more likely to be actively searching for information and resources, which I don’t tend to believe. I take this as a good sign that people in tech are more likely to seek professional help. Maybe they’ve seen what well-meaning but uninformed people can do when trying to fix their computers, and they don’t want to do the same thing to their brains.
Most Common Mental Illnesses
So for the next ring down we see which mental illnesses people were reporting. I again don’t love the way this survey was designed, because they listed all Mood disorders as one category, so depression, Major Depression and Bipolar disorder are all in one category, even though they are so different in terms of severity and symptomology. Also, I didn’t include this in my graphic, but the survey had “addictive disorders” and “substance use disorders” as separate categories, and even I would have no idea how to answer that question. Are addictive behaviors just behavioral addictions, like gambling or sex? The survey does not say, but I digress.
There aren’t a whole lot of differences between the “Yes I was diagnosed” folks and the “Maybe, I think I have…” folks, which surprised me. I expected that some of the more popularized diagnoses, like PTSD or OCD, which show up a lot in the media would be more represented in the self-diagnosed crowd, but it’s almost exactly the same.
Overall, unsurprisingly mood and anxiety disorders reign supreme. The Anxiety and Depression Association of America says that anxiety disorders are the most common diagnoses in America, and they would know. In this survey, mood disorders are quite a bit more common. I tend to think that this means that depression is a little bit more common in tech workers than in other people, given that Bipolar probably isn’t contributing all that much.
The next most common diagnosis was ADHD, which is again not too different from the non-tech world, where about 4% of adults have ADHD. I might have expected this to be a little bit lower, as ADHD tends to impact academic performance, and many tech jobs require at least a college degree. However, at least one person thinks that careers in tech are actually perfect for someone with ADHD, so maybe he’s right.
The biggest category of diagnoses was multiple diagnoses, at over half of all people reporting any diagnosis at all. That outermost ring shows the number of diagnoses, and you can see that most people just listed two. The graph was already getting a little busy and I didn’t want to tease out which particular diagnoses people listed, but for most people it was mood disorder and anxiety disorder. This again tracks with the general population, where over half of all people with anxiety also have depression, and vice versa.
Beyond that, I was relieved to see that there weren’t a whole lot of people with 3 or more diagnoses. A lot of disorders, particularly personality disorders and bipolar disorder, can be really tricky to diagnose. People can end up with a whole alphabet soup of diagnoses before they get in front of someone who spends enough time to realize that all of their various constellations of symptoms are really just stemming from one galaxy, so to speak.
- These folks should revamp their survey. In addition to the poor wording of the questions I already mentioned, there were no continuous measures and no demographic information. In the parlance of our times, nerdiness is intersectional, and we know that race and ethnicity have a huge impact on mental illness and treatment.
- Nerds aren’t so different from normals after all. They seem a bit more likely to get into treatment, and mood disorders are a bit more prevalent, but overall it’s pretty similar.
- Self-diagnosis isn’t as bad as I thought. Going in, I was really curious about the folks who said they “maybe” have a mental illness, but these people reported almost exactly the same diagnoses as the people who had been professionally diagnosed. There’s no way to know if this is true at the individual level without getting each of these people in front of a professional, but I didn’t see the pull to the more exotic mental illnesses that I was expecting.
I think I’m about done with this dataset. For as much shit as I gave them about the particular questions and options, I’m mostly complaining because I’m trying to fit their square peg into my round hole. I’ve mostly sidestepped the main point of their survey, which is to examine how the tech field responds to mental illness, but their own website does a great job of doing that. But until I find a university willing to sponsor my Nerd Mind Laboratory and the associated research agenda, this is what I have to work with.