Vidcast:  https://youtu.be/TzbK0xNiGhM

The pattern of the words you use on social media generally and on Facebook specifically can help your doctors and you identify your unique set of medical problems including diabetes and a bevy of psychological issues including anxiety, depression, and frank psychosis.  Researchers at the University of Pennsylvania processed the language in nearly 1 million Facebook status updates from about 1000 participants, and they cross-referenced their words with healthcare data from their electronic medical records.

They focused on 21 separate conditions, and Facebook data accurately predicted each of them.  In about half of the conditions, Facebook did a better job of prediction than a patient’s raw profile data.  The posts most often predicted diabetes, pregnancy, and mental health issues.

The words ‘bottle’ and ‘drinking’ were associated with alcohol abuse.  Hostile words and stories most often occurred in the posts of drug abusers and those with psychiatric issues.  In fact, the posts predicted depression 3 months earlier than clinical data alone.

Social media posts have for years been used in a negative way by employers and colleges to screen out applicants.  This same information can be used in a positive way to give doctors a unique glimpse into the lives of their patients that may provide better ways to keep them healthy.

Raina M. Merchant, David A. Asch, Patrick Crutchley, Lyle H. Ungar, Sharath C. Guntuku, Johannes C. Eichstaedt, Shawndra Hill, Kevin Padrez, Robert J. Smith, H. Andrew Schwartz. Evaluating the predictability of medical conditions from social media posts. PLOS ONE, 2019; 14 (6): e0215476 DOI: 10.1371/journal.pone.0215476

#Facebook #diagnosis #lifestyle