2015 American Journal of Gastroenterology Lecture

How Digital Health Will Transform Gastroenterology

Brennan Spiegel, MD, MSHS


Am J Gastroenterol. 2016;111(5):624-630. 

In This Article

Digital Health Technology No. 2: Social Media

Social media has forever changed how society communicates. In health care, there are thousands of online forums where patients share information about their illness, discuss how it affects their lives, and exchange notes about how they are managing and treating their symptoms. With more mobile phones in the world than toothbrushes,[13] trillions of text messages per year, and over a billion Facebook users, this is a permanent change. Researchers are now harnessing the power of social media to understand the illness experience of our patients.

Although social media is a new data source for health care, it presents a rich opportunity to learn about the lives of non-experimental patients outside the walls of health-care facilities. In addition, people of all education levels and ages, from adolescence to late adulthood, use social media. This allows for a sampling of perspectives over a wide geographic distribution. Social media is currently being used for patient recruitment for clinical trials,[14] measurement of consumer sentiment,[15] patient education,[15] formation of patient affinity groups,[14] patient monitoring, management of patient care, and epidemiological research.[15,16] Sentiment analysis and "digital ethnography" of social media data have been used to identify diurnal and seasonal mood variations across cultures,[17] monitor viral epidemics,[18] describe the spectrum of migraine headache symptoms,[19] and predict which postpartum mothers will develop depression.[20]

Social media can also be used to passively monitor patients with GI disorders. For example, we performed digital ethnography to identify patients with inflammatory bowel disease and learn about their illness experience. Working with computer scientists, we identified patients with Crohn's disease by trolling through the world's tweets.[21] We then developed a natural language processing algorithm to automatically identify patients without requiring hand-review of tweets; this allows analysts to track individual patients over time.

Whereas the confines of an office visit may inhibit some patients from expressing their true opinions, they often voice free-flowing, detailed, and open information when outside of a health-care facility. For example, we recently used social media to study how patients using opioids balance GI side effects, such as opioid-induced constipation and nausea, with pain management.[22] Computers parsed through 2,519,868 Tweets and >1.8 billion e-forum posts to eventually identify 3,003 individuals who experienced opioid-induced GI side effects. The resulting posts revealed a struggle to balance pain management with GI side effects. Often, patients expressed frustration that their prescribing physicians failed to warn them about GI side effects, and many altered their pain medication regimens without consulting their physician. This example reveals how social media can passively tap into the lives of patients outside the four walls of the clinic and can identify opportunities to improve physician–patient communication.