Is Twitter the next big thing in ADHD diagnosis?

anna-demianenko-12400 (2)

Yesterday, I received my daily Google Alert email update for the search terms “attention deficit.” I’ve had this set up for several years, but I must admit, I don’t often check it; I can offer no better excuse than that it is rather hard to keep abreast of daily news about attention deficit if you suffer chronically from it. In any case, this time, something compelled me to read through the headlines compiled in the email, and one caught my attention (no pun intended): “Searching for ADHD in a Million Tweets.” It states: 

After analyzing 1.3 million tweets written by nearly 1,400 Twitter users, researchers at the University of Pennsylvania have created a machine learning model that can predict which of the site’s users are affected by attention deficit/hyperactivity disorder (ADHD). […] Researchers monitored Twitter users with self-reported ADHD, comparing them to a set written by age- and gender-matched controls. The team then used various machine learning models to search for patterns in the two sets of tweets.

It might come as a surprise to you, my readers, but I must say, I find the whole thing terribly problematic. Why? Because it fails to take into account the fact that there are as many manifestations of ADHD as people living with it. The effects of ADHD in the life of one person may be entirely different from that of another.

For example, Lyle Ungar, one of the study co-authors, was quoted as saying, “I didn’t realize how common it was for patients to use marijuana to treat their symptoms, so you see people talking more about dope and weed.” But in contrast to these ADHDers, I avoid pot like the plague, not wanting to be any more dumbed down than I already am.

Meanwhile, according to the abstract of the article accompanying the research, published Nov. 8 in the Journal of Attention Disorders, “Users with ADHD are found to be less agreeable, more open, to post more often, and to use more negations, hedging, and swear words. Posts are suggestive of themes of emotional dysregulation, self-criticism, substance abuse, and exhaustion.” It seems to me the way these results have been reported is liable to reify negative stereotypes associated with ADHD. 

ADHD: not ‘one-size-fits-all’

I commend anyone wishing to shed light on ADHD, which after all is still arcane to most people who neither live with it or know someone who does. Also, for the record, I’m well aware of the inherent value of the internet, and social media, in particular, having minored in communication studies as an undergrad at the University of Oregon.

Even so, this research is misguided. It assumes there is one, and only one, kind of life ADHDers live, regardless of gender identity, socioeconomic status, race, etc. 

True, some health outcomes are more likely for people with ADHD. Depression, eating disorders, and self-injury are among a slew of comorbidities, or co-occurring conditions, in ADHD women, for instance. But therein lies the salient point: These aspects are common in ADHD women, specifically. As Devon Frye notes in a blog post on ADDitude, “men with ADHD are more likely to have substance-abuse problems while women with ADHD are more likely to have personality or mood disorders.”

Thinking outside the ADHD box

Cat-out-of-a-box

Far from demarginalizing ADHDers, the reasoning driving this research homogenizes the ADHD experience. I love the idea of a group of researchers at an Ivy-League institution setting out to make things easier for people who may have ADHD. This is especially so because Ungar has said while ADHD is less well studied than conditions like depression, “Understanding the components that people have or don’t have, the range of coping mechanisms that people use all leads to a better understanding of the condition.”

What we really need

What people with ADHD — those who have been diagnosed with it, those who suspect they have it, and especially, perhaps, those who live with it unknowingly — really need is for the medical community to realize there is no one, singular ADHD experience.

Yes, this does make it harder to single out people who are likely to have undiagnosed, and thus, untreated ADHD. Yes, this does mean there would need to be a concerted effort to map all tweets from studied ADHDers, and not just those that appear so frequently as to be acceptable to make generalizations about said ADHDers.

But the effect — newfound insight into ADHD in all its countless iterations — would be well worth it.

 

One thought on “Is Twitter the next big thing in ADHD diagnosis?

  1. This study sounds like it could actually be helpful in the long run. The cool thing about machine learning models is that the more data you give them, the more accurate they become. So theoretically these researchers could continue to feed more tweets from ADHDers and non-ADHDers into this model and it would become better at “diagnosing” over time. It could create a profile of the “average” person with ADHD. Not every person with ADHD will fit that model and get “diagnosed” by it. It’s not the be-all-end-all diagnosis tool, but I think it could still be helpful.

    The weird thing that article didn’t really address is that when people post on social media, they’re often curating a version of themselves how they want others to see them. If only you could take this model and apply it to people’s private journaling of their thoughts, it might be more useful. But also way more creepy.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s