In a world of digital addicts, oversharing on social media is often seen as a character flaw — but a new study reveals it might actually be good for your health. Researchers from Penn Medicine and Stony Brook University analyzed the Facebook statuses of 999 participants whose posts were more than 500 words — amounting to 949,530 posts on the social network. From there, scientists were able to predict 21 medical conditions — from pregnancy to skin issues — from the Facebook users’ updates. “People’s personality, mental state, and health behaviors are all reflected in their social media and all have tremendous impact on health,” the authors write in their report to be published Wednesday in the peer-reviewed journal PLOS One. Facebook statuses were “particularly effective at predicting diabetes and mental health conditions including anxiety, depression and psychoses.” By identifying language associated with behaviors or symptoms of diagnoses, researchers were able to predict the users’ underlying health issues. For example, words such as “drink,” “drunk” or “bottle” were flagged by researchers for alcohol abuse. The more those words turned up, the more likely the poster was to have a real problem. On the other hand, some word associations were not so straightforward. Those who most frequently referenced religion through words such as “pray” and “God” were 15 times more likely to have diabetes than participants who used religious language the least. Over the course of the study, many of the scientists were able to correctly predict depression, for example, up to three months earlier than an official clinical diagnosis. They found that 10 of 21 conditions studied were actually better detected through Facebook than by using demographic information. While these hints cannot be used to prove any conclusions on a person’s health, researchers say they can indicate a pattern of health concerns — and help doctors devise more personalized health care. “As social media posts are often about someone’s lifestyle choices and experiences or how they’re feeling, this information could provide additional information about disease management and exacerbation,” says lead author Dr. Raina Merchant, director of Penn Medicine’s Center for Digital Health. Senior author and computer science professor Andrew Schwartz adds, “Our digital language captures powerful aspects of our lives that are likely quite different from what is captured through traditional medical data.” However, experts warn that using social media in this setting “raises parallel questions about privacy, informed consent, and data ownership.”
In a world of digital addicts, oversharing on social media is often seen as a character flaw — but a new study reveals it might actually be good for your health. Researchers from Penn Medicine and Stony Brook University analyzed the Facebook statuses of 999 participants whose posts were more than 500 words — amounting to 949,530 posts on the social network. From there, scientists were able to predict 21 medical conditions — from pregnancy to skin issues — from the Facebook users’ updates. “People’s personality, mental state, and health behaviors are all reflected in their social media and all have tremendous impact on health,” the authors write in their report to be published Wednesday in the peer-reviewed journal PLOS One. Facebook statuses were “particularly effective at predicting diabetes and mental health conditions including anxiety, depression and psychoses.” By identifying language associated with behaviors or symptoms of diagnoses, researchers were able to predict the users’ underlying health issues. For example, words such as “drink,” “drunk” or “bottle” were flagged by researchers for alcohol abuse. The more those words turned up, the more likely the poster was to have a real problem. On the other hand, some word associations were not so straightforward. Those who most frequently referenced religion through words such as “pray” and “God” were 15 times more likely to have diabetes than participants who used religious language the least. Over the course of the study, many of the scientists were able to correctly predict depression, for example, up to three months earlier than an official clinical diagnosis. They found that 10 of 21 conditions studied were actually better detected through Facebook than by using demographic information. While these hints cannot be used to prove any conclusions on a person’s health, researchers say they can indicate a pattern of health concerns — and help doctors devise more personalized health care. “As social media posts are often about someone’s lifestyle choices and experiences or how they’re feeling, this information could provide additional information about disease management and exacerbation,” says lead author Dr. Raina Merchant, director of Penn Medicine’s Center for Digital Health. Senior author and computer science professor Andrew Schwartz adds, “Our digital language captures powerful aspects of our lives that are likely quite different from what is captured through traditional medical data.” However, experts warn that using social media in this setting “raises parallel questions about privacy, informed consent, and data ownership.”
0 Comments