By GREGORY ZELLER //
Social media is an undeniable mixed bag – global connector, presidential bully pulpit, unprecedented panic-spreading platform – but its potential as a real-time healthcare tool has hardly been tapped.
So says a new study published in Scientific Reports, an online open-access scientific “mega journal” published by Nature Research. The study, co-led by Stony Brook University researcher H. Andrew Schwartz, reveals the language people use in their Facebook posts subtly changes before they visit an emergency room – sometimes, months before they visit – evidence that “social media is often an unseen signal of medical distress,” according to Stony Brook University.
The study, “Variability in Language Used on Social Media Prior to Hospital Visits,” compared patients’ Facebook posts to their medical records and found that shifts to “more formal language” – including less swearing, fewer emojis and complete sentences where acronyms usually rule – “reliably preceded hospital visits,” the university said in a statement.
The language also featured a more “worrisome and depressed” tone, along with more detailed descriptions of physical pain – in some cases, as long as two months or more before patients finally wound up in the ER.
The study focused on 2,915 patients at an unnamed urban hospital who consented to sharing their Facebook posts and electronic health records, including 419 patients who made recent emergency-department visits for symptoms ranging from chest pain to pregnancy-related issues.
Posts from as early as two-and-a-half months before their ER visits were analyzed with a language-processing model that uses machine-learning protocols to detect subtle changes – less cussing, more references to health and family – over time.
While Facebook doesn’t necessarily speak for all social media, “people live a lot of their lives online and Facebook is that dominant platform right now,” according to Schwartz, who co-led the study with Sharath Chandra Guntuku, a research scientist at the Center for Digital Health, part of the Perelman School of Medicine at the University of Pennsylvania.
And according to their Facebook posts, imminent emergency patients often drop copious digital clues, noted Schwartz, an assistant professor in SBU’s Department of Computer Science.
“The decrease in informal language seems to go hand-in-hand with an increase in anxiety-related language,” he added.
Being able to detect those subtle changes weeks or months before a medical condition reaches the crisis stage helps providers understand the contexts in which patients ultimately seek care – the timing, the pain thresholds, the other societal factors influencing their decision.
“How does life affect personal decisions to seek care?” Schwartz said. “How does care affect life? These are the things I would hope that we could fully describe, how people’s everyday lives intermix with healthcare.”
These are important clues for the future of healthcare at any time, but especially during situations like the global COVID-19 pandemic. And while patient privacy must always remain an “utmost priority,” Guntuku believes the research is critical to detecting serious conditions before they become desperate – and to understanding the psyche of the 21st century patient.
“The better we understand the context in which people are seeking care, the better they can be attended to,” Guntuku added. “While this research is in a very early stage, it could potentially be used to both identify at-risk patients for immediate follow-up or facilitate more proactive messaging for patients reporting doubts about what to do before a specific procedure.”