Depression and PTSD are mental illnesses that are often missed and go undiagnosed. Victims do not always seek help and friends may not feel empowered to intervene. However, at a conference of statisticians earlier this month in Boston, there is significant potential that social media posts can provide evidence of depression and PTSD.
Currently, researchers are mining through social media posts for clues about mental illness, using only posts that users have agreed to make public. There is hope that doing this could help to create an algorithm that would help friends and family identify potentially depressed individuals. While this seems far-fetched, that one emotional social media post could diagnose depression and there is the notion of invasion of privacy, the benefit of the diagnosis could significantly outweigh the intrusion.
With about 270 million active monthly users around the world, Twitter seems to be of interest to scientists to serve as “the largest observational study of human behavior we’ve ever known.” At the conference, researchers presented preliminary findings on a study that found a correlation between depression and tweets in areas of high unemployment. In addition, they traced PTSD in Twitter users’ posts from regions with high populations of veterans. Earlier this year, researchers created an algorithm that scans tweets for depression and was 70 percent accurate.
While Twitter seems to be a valuable public health tool, more work needs to be done in considering how this information can be used while still preserving privacy.
