Back in 2002 a new health information service in Canada was monitoring news and public health reports across the world — mining them for data about possible outbreaks of communicable diseases. Late that year, the algorithm picked up disturbing reports of respiratory disease among young children in Southern China.

Months later, the Chinese government publicly disclosed that it was grappling with a frightening new virus called Severe Acute Respiratory Syndrome — SARS. With an almost 10 percent fatality rate, the virus would go on to kill more than 750 people in 17 countries.

“The Chinese government finally acknowledged they had SARS cases and I think it was February of 2003,” says Ronald St. John, the doctor who created the Global Public Health Intelligence Network (GPHIN). “However, GPHIN had picked it up November - months before that.”

Decades after it was created, GPHIN has become an important tool for global public health officials to monitor and track diseases quickly — sometimes before they are officially acknowledged. Some countries are reluctant to report epidemics in the earliest phases, mindful not to create a panic and also worried about possible economic impact.

But news organizations — and, increasingly, social media — are quick to report unusual occurrences of illness and especially death. GPHIN monitors thousands of local and national websites in nine different languages, scanning for keywords that indicate a communicable disease. Artificial intelligence programs crawl through the data and flag hotspots that are then analyzed and verified by human analysts. 

The World Health Organization (WHO) says that now “more than 60% of the initial outbreak reports come from unofficial informal sources,” adding that it then verifies the most troublesome reports it receives.

GPHIN primarily relies on news reports, but other health tracking networks acknowledge the growing role of social media. In 2014, the HealthMap system tracked disturbing reports on news websites and social media of a hemorrhagic fever in West Africa. It was later identified as Ebola. 

Founder Dr. John Brownstein says Twitter has also helped them map cases of food poisoning, cholera and influenza in the United States. “Digital traces are these breadcrumbs that people leave behind about their health,” he tweeted. 

All agree that official public health reports are the gold standard of verified and useful data. But some experts believe we should further ramp up efforts to mine health data from social media — as we increasingly overshare our health concerns with friends, family and the world at large. 

Ming-Hsiang Tsouis, a professor at San Diego State University, has studied how the spread of the flu is tracked on Twitter. 

“Traditional methods of collecting patient data, reporting to health officials and compiling reports are costly and time consuming,” Tsou told Government Technology magazine. “In recent years, syndromic surveillance tools have expanded and researchers are able to exploit the vast amount of data available in real time on the Internet at minimal cost.”

So next time you overshare about your sniffles, there might be an algorithm out there that's even more interested in the details than your followers.  

Published on Feb 06, 2020