recommended reading

HHS Official: Social Media Listening a ‘Routine’ Part of Emergency Response

Twin Design/Shutterstock.com

Tweets are essential data points when mapping out federal emergency response, a group of government officials told an audience Thursday.

In the first few days after Hurricane Sandy in 2012, government response teams scanned the microblogging platform for mentions of especially dire situations -- including hospitals without power, nursing homes that needed to evacuate residents, and elderly people trapped in high-rise buildings -- to target aid. There were about 20 million tweets related to the storm, said Nancy Nurthen, a director within the Department of Health and Human Services’ Office of Emergency Management, speaking at a monthly AFCEA Bethesda breakfast. 

On the 50th anniversary of Martin Luther King Jr.’s march on Washington, HHS also checked Twitter for geolocated tweets mentioning people suffering from heat exhaustion so Nurthen’s office could dispatch response teams to treat those patients.

Social media is a “routine” information source in emergency response, Nurthen said, noting that her office especially follows “verified sources” such as local aid organizations, news reporters and politicians for news.

The Agriculture Department's emergency response agency is training to understand social media listening techniques, according to Cora Russell, acting director at the Food and Nutrition Services’ Office of Emergency Management within USDA.

But the government must separate the relevant tweets from the rest, said Scott Shoup, the Federal Emergency Management Agency's chief data officer.

“Things can spin way out of control really quickly," he said.

A tweet about a fallen tree blocking a road may no longer be relevant if the tree has been cleared by the time a response team gets to the scene, he said.

Sometimes, he said, “no information is better than some information.”

Threatwatch Alert

Thousands of cyber attacks occur each day

See the latest threats

JOIN THE DISCUSSION

Close [ x ] More from Nextgov