Social media has evolved tremendously over the years, and its capabilities are ever-growing and changing. While still a way for people to keep in touch and connect with their friends, social media has also grown to do much more than that.
A new study, conducted by researchers from the University of Pennsylvania and Stony Brook University and recently published in Proceedings of the National Academy of Sciences of the United States of America, explores a new algorithm that can predict a depression diagnosis based off of language used in people’s Facebook posts.
“There’s a perception that using social media is not good for one’s mental health, but it may turn out to be an important tool for diagnosing, monitoring, and eventually treating it,” said lead researcher H. Andrew Schwartz. “Here, we’ve shown that it can be used with clinical records, a step toward improving mental health with social media.”
The University of Pennsylvania’s Positive Psychology Center and Stony Brook University’s Human Language Analysis Lab have been working on the World Well-Being Project since 2012. The initiative strives to make breakthroughs in psychological wellness based on language used on social media.
The researchers involved in this study wanted to take the World Well-Being Project to the next level and see if social media could be an effective tool in diagnosing depression. Over 1,000 people joined the study -- 114 of whom had depression diagnoses in their medical records -- and gave the researchers full access to their Facebook statuses and medical records.
To compare the participants with depression to a control group, the researchers used five participants without depression for every one participant with depression for a total of 683 study participants. The researchers then used the participants’ Facebook statuses -- which totaled over 500,000 posts -- to determine the words and phrases that they deemed “depression-associated language markers.”
In putting the algorithm to the test, the researchers found that it was successful in predicting depression up to three months before a formal diagnosis, based entirely off of language used in Facebook posts.
The authors concluded that the users most likely to develop depression were found to use language expressing loneliness, anger, sadness, and self-reflection in their Facebook statuses.
“The hope is that one day, these screening systems can be integrated into systems of care,” said researcher Johannes Eichstaedt. “This tool raises yellow flags; eventually the hope is that you could directly funnel people it identifies into scalable treatment modalities.”