Could tweets help clinicians diagnose adults with ADHD? That remains to be seen. But researchers recently found that adults with ADHD use language in characteristic ways on social media, with a tendency to be more open, less agreeable, and focus more on the past than adults without ADHD.
Their study, published in Journal of Attention Disorders, lays the groundwork for further research that may lead to using social media as a tool for screening, diagnosis, or even treatment of adults with ADHD.
For this investigation, researchers from the University of Pennsylvania used Twitter to analyze the language of adult users with ADHD to gain insights into how behavioral symptoms of ADHD manifested on social media. All told, the researchers analyzed some 1.3 million tweets posted by 1,032 Twitter users with self-reported diagnoses of ADHD between 2012 and 2016 and compared their tweets to those posted by a control group of 1,029 Twitter users matched by age, gender, and period of activity.
Their analysis revealed that adults with ADHD tend to post more frequently than controls, and tended to have more followers. Adults with ADHD were often less agreeable in their tweets, but also more open. They also used more negations, hedging, and swear words. The tweets suggested themes of emotional dysregulation, self-criticism, substance abuse, and exhaustion. Adults with ADHD on Twitter also often posted overnight, from midnight to 6 a.m.
In addition, the researchers found that those with ADHD tended to have low “future-orientation,” in that they tweeted more about the past compared to users without ADHD. Moreover, compared to controls, they posted significantly more words related to drugs and drug use, including drugs prescribed for treating ADHD or related mood disorders and recreational drugs, such as marijuana, that may be viewed by adults with ADHD as self-medication for ADHD symptoms.
“While the motivation of this work is to understand the manifestation of ADHD symptoms in online spaces, it is only a preliminary step toward considering the possible role of online platforms in the detection of and intervention with social media users with ADHD,” the authors noted.
Although the researchers found that social media language is predictive of ADHD, the type of data set they used was not sufficient for screening purposes.
“Future studies should examine the integration of both social media data and clinician judgment for ADHD screening and/or diagnosis, such as with a semi-structured interview delivered by a clinician,” the researchers stated.
Limitations and expectations
There are, of course, important limitations to this study. For their analysis, researchers used the social media language of individuals who self-identified as diagnosed with ADHD, and had no way of confirming whether these Twitter users were formally diagnosed by a mental health professional.
Also, Twitter posts were restricted to 140 characters, limiting the user’s expression in ways that might affect the content and other characteristics of the post, thereby limiting the conclusions that may be drawn from an analysis of such posts.
Moreover, while the researchers used established protocols for gathering their data and employed established models of ADHD for conducting their analysis, “interpretation of the meaning and intent of word clusters used in posts in this sort of research requires a degree of empirically- and clinically-informed conjecture,” they acknowledged.
The relationship between social media and mental health is an increasing concern among clinicians. The general approach and findings of this study may provide a novel route to a more profound understanding of adults with ADHD, the authors noted, as well as potential strategies for using social media not only for research but for screening and clinical uses that remain to be developed.