People can rely on social media like Facebook to showcase the highlights of their lives, such as vacations. New research suggests, however, that the language they use in posts can also help in predicting depression.
Using sophisticated software, researchers were able to scan social media posts and study for depression months before clinical trials.
"Social media has enabled people to share some of their daily lives with researchers," said study author Andrew Schwartz, a lecturer in computer science at Stony Brook University in New York.
"Basically, we used the language that people wrote every day and referred them to whether they diagnosed depression," he explained.
A look at Facebook posts "was slightly more accurate than standard screening questions in finding depression," said Schwartz.
So what types of language could someone with depression show?
The use of first-person pronouns was one of the patterns that the researchers saw. This means people often use "I" or "I" in their social media posts.
Schwartz said that people who were eventually diagnosed with depression often talked about their feelings, physical pain and pain and were alone.
But he warned against trying to diagnose your friends or family based on some social media posts.
"One single contribution is not enough to see a depression, we have been worried about depression for six months, so I would not support people trying to condemn their friends and family," he said.
Every year, more than 6% of Americans suffer from depression, the study authors noted. But less than half receive treatment for the disorder. These high levels of underdiagnosis or underdiagnosis suggest that current methods of detecting depression may be improved.
The research team was led by Johannes Eichstaedt, a Ph.D. student at the University of Pennsylvania.
Investigators entered the Facebook posts of nearly 700 people who had gone to an emergency room at an academic center, including 1
The researchers studied more than half a million Facebook posts to create the software algorithm for detecting depression. They identified the most commonly used words and phrases to identify depression-associated language markers.
Using these language markers, researchers were able to predict future depression as early as three months before being included in medical records.
"Social media gets a lot of negative attention, but there is a downside, it could be a very powerful tool for the overburdened mental health industry," said Schwartz.
The researchers imagine that this could be a screening tool for clinicians to use to detect depression earlier. But Schwartz also said more study is needed.
Dr. Alan Geller is a psychiatrist at Gracie Square Hospital, New York City, who was not involved in the new study. He said, "Depression is a real problem and preventative treatment is better than saving someone."
"The idea of capturing someone's risk on Facebook is appealing. Any kind of mental health technology – especially because we do not have tests like labs or imaging and rely on what people tell us – could help, "he added.
Published online Proceedings of the National Academy of Sciences.