Health data can be gleaned through social media, as some medical conditions were – as recently found by American research – easier predicted from Facebook data instead of demographic information.
Many people who take to social media in this digital age have more insightful posts, which surreptitiously provides additional information about disease management and exacerbation, according to Raina Merchant, MD and Director of Pennsylvania Medicine’s Center for Digital Health, US.
Merchant’s research adopted three automated data models of nearly 1,000 patients, who had their electronic medical records linked to their Facebook profiles – one model only analysed the patients’ Facebook post language, another used patient demographics, such as age and sex, and the last model combined the two datasets. Some of the Facebook data was found to be more intuitive than others. Words like “drink/bottle” were shown to be more predictive of alcohol abuse, while words expressing hostility indicated drug abuse and/or psychoses.
Andrew Schwartz, PhD, a visiting Assistant Professor at Penn in Computer and Information Science, thinks current digital language patterns are likely quite different from traditional medical data; the predictive language could enable new applications of AI for medicine.
Want to publish your own articles on DistilINFO Publications?
Send us an email, we will get in touch with you.
At present, an opt-in system for patients that analyses their social media posts for refined care delivery would be welcome, but, Merchant opines that care providers must learn to condense and summarise social media data in order to interpret it successfully.
“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,” Merchant said.
Date: June 26, 2019
Source: HealthCare Asia Daily
