Study claims Facebook statuses can reveal people’s medical conditions

A new study claims that people’s Facebook statuses can be used to determine whether they suffer from certain diseases.

Carried out by researchers from the University of Pennsylvania, the peer-reviewed study analyzed status updates that contained more than 500 words from 999 individuals.

“Over two billion people regularly share information about their daily lives over social media, often revealing who they are, including their sentiments, personality, demographics, and population behavior,” the study states. “Because such content is constantly being created outside the context of health care systems and clinical studies, it can reveal disease markers in patients’ daily lives that are otherwise invisible to clinicians and medical researchers.”

Ranging from March 2009 through October 2015, the 950,00 Facebook posts, which totaled over 200 million words, helped the researchers predict whether an individual had health issues such as diabetes, anxiety, depression, and psychoses.

The researchers found that all 21 medical condition categories used in the study “were predictable from Facebook language beyond chance” and that 18 of those categories were better predicted with both demographic information and Facebook language than just demographic information alone.

Repeated use of the words “drink,” “drunk,” and “bottle” helped predict alcohol abuse while the words “pain,” “crying,” and “tears” were linked to issues such as depression. The study even claims that those who used the words “god,” “family,” and “pray” were often found to have diabetes.

“This does not mean that everyone mentioning these topics has the condition, but just that those mentioning it are more likely to have it,” the study notes. “For example, the top 25% of patients mentioning the (godfamilypray) topic were 15 times more likely to have been diagnosed with diabetes than those in the bottom 25% of mentioning that same topic.”

The study is not without its limitations. Researchers noted that 76% of participants were female while 71% were African-American and therefore “not representative of the general population.” The study also relied on just one social media site, Facebook; incorporating status updates from other social networks like Instagram and Twitter could lead to different conclusions.

This research is not the first to infer a link between Facebook user activity and personal traits. In 2013, researchers said they could predict a user’s race, IQ, sexuality, and political views based only on their Likes. For example, the study alleged that users who Liked Harley Davidson motorcycles and actor Tyler Perry were more likely to have a lower IQ.

READ MORE:

Got five minutes? We’d love to hear from you. Help shape our journalism and be entered to win an Amazon gift card by filling out our 2019 reader survey.

Mikael Thalen

Mikael Thalen

Mikael Thalen is a tech and security reporter based in Seattle, covering social media, data breaches, hackers, and more.