More than 16 million Americans suffer from depression in a given year, but less than half get treatment.
Researchers found that social media can potentially help shrink this gap and provide doctors with another way of identifying individuals at risk of the mental health condition.
Findings of a new study show that the language that a person uses in Facebook posts can predict the likelihood that this person will be diagnosed with depression three months later.
In the research, which was published in the Proceedings of the National Academy of Sciences on Oct. 15, researchers used an algorithm to scan more than 524,292 Facebook updates of 683 people.
Study researcher Raina Merchant, director of the Penn Medicine Center for Digital Health, and colleagues found that those who were later diagnosed with depression use first-person pronouns such as “I” and” “me” heavily. They also tend to use emotional languages such as “tears” and “feeling.”
The Telegraph said the algorithm was developed at the World Well-Being Project (WWBP), which is based at the Positive Psychology Center of the University of Pennsylvania, and the Human Language Analysis Lab of Stony Brook University.
“We found that language predictors of depression include emotional (sadness), interpersonal (loneliness, hostility), and cognitive (preoccupation with the self, rumination) processes,” Merchant and colleagues wrote in their study.
The model was found to work best when using Facebook data three months before a person received a diagnosis for depression. The model becomes less precise when longer periods of Facebook data were included.
The predictive model is moderately accurate but the approach holds promise in helping flag people who may be in need of help.
“We want to think of new ways to get people resources and identification for depression earlier,” Merchant told Philly.
“We’re at the very beginning of trying to understand how this data is sometimes people just saying hi to each other, but sometimes it can give us insight into the health of individuals and communities.”
The research is not the first to show that social media posts may provide clues about a person’s mental health problems. Other researchers also noticed patterns on Twitter and Instagram that may suggest the user suffers from depression.
In an earlier study, for instance, depressed Twitter users were found to more likely tweet during night hours. Another study also found that Instagram users who use black and white or muted colors in their posts may be suffering from depression.