Researchers Merge Social Media, EHR Data to Gain Health Insights

A database that merges social media with electronic health record data has the potential to provide insights about patients’ health and health outcomes, according to researchers at the University of Pennsylvania’s Department of Emergency Medicine.

Adult Facebook and Twitter users who sought care at Penn’s emergency department were queried about their willingness to share their social media data and EHR data so that researchers could build a database for the study. More than 1,000 patients consented to share their social media data to compare it with their EHR.

The aim of the study was to “learn from social media what we can take from these digital bread crumbs that people are leaving to better understand populations and population health,” says Kevin Padrez, M.D., formerly a fellow at the Penn Medicine Social Media and Health Innovation Lab and now a resident physician in the University of California-San Francisco Department of Emergency Medicine.

The study pairs social media data with EHR data to better understand the relationship between individuals’ health and their everyday lives, as well as their attitudes towards health and the perceived quality of healthcare services.

In particular, shared social media posts were searched online for select terms about health and healthcare. “We looked at what patients were talking about on social media and how that related to their actual health status,” Padrez says. “People are not just talking about health in general but about their health.”

Also See: Poll Finds Parents 'Overshare' on Social Media

Among the findings of the study published in BMJ Quality & Safety was that patients with a given diagnosis in their EHR were significantly more likely to use terms related to that diagnosis on Facebook than patients without that diagnosis in their EHR.

Pardez says one of the questions that remain to be answered by future research is whether a patient with a chronic condition who more actively posts on social media about their disease reflects a flare up in their symptoms and is more likely to be hospitalized. 

Participants in the study who shared social media data tended to be younger, more likely to post at least once a day and more likely to present to the emergency room via self-arrival mode and have private health insurance.  

“One of the most surprising take homes from our study was it was less difficult than anticipated to persuade patients to share their social media data for health research purposes,” Padrez concludes. “We were pleasantly surprised with their willingness to share.”