A new research, conducted by Penn Medicine and Stony Brook University, has revealed that your social media posts can be used for diagnosing illnesses faster than a clinical diagnosis. The research has concluded that people use words and dialects in posts shared on social media that point to conditions such as diabetes, depression, anxiety, and psychosis.
The researchers analyzed 949,530 posts by 999 participants using a natural language processing tool and each post had at least 500 words. The posts were vetted for 21 medical conditions and it was found that all the 21 conditions can be predicted from the posts.
In the research, it was revealed that people who use words like “family,” “god,” and “pray” are 15 times more likely to be suffering from diabetes.
Similarly, there are more chances of people suffering from alcohol abuse if their posts mention words like “bottle,” “drunk,” and “drunk”. However, this is easy to detect.
People who are suffering from anxiety mention words related to physical symptoms of it like “head,” “stomach” and “hurt.”
Researchers say that it cannot be said that all the people who mention such words are suffering from related medical conditions, but those who mention such words are more likely to be diagnosed with the related medical condition.
As per the findings of the research, it makes sense to devise a model that includes clinicians analyzing your social media posts to detect your medical conditions beforehand. However, it would come at the cost of your privacy.
Raina Merchant, the lead author of the research and the director of Penn Medicine’s Center for Digital Health says, “For instance, if someone is trying to lose weight and needs help understanding their food choices and exercise regimens, having a healthcare provider review their social media record might give them more insight into their usual patterns in order to help improve them.”
The fact that researchers were able to diagnose your illness from Facebook posts before a clinical diagnosis is indicative of the fact that we are sharing too much on social media.
We fail to understand that the words and bits that we share on social media unconsciously can be used to track our habits. These bits and pieces can be combined to form a bigger picture which could pose a risk to our privacy.
The research was published in Plos One and you can read the complete findings here.