How opioid-related messaging goes viral: Biocomplexity Institute of Virginia Tech researchers share findings with public health experts
April 4, 2018
What does the U.S. opioid epidemic look like from the front lines? An estimated 116 overdose deaths a day and more than a thousand trips to the emergency room.
Amidst all this tragedy and confusion, public health officials are working hard to circulate potentially lifesaving information. Ads for addiction treatment resources and warnings where local street drugs have been laced with ultra-potent synthetic opioids are becoming increasingly common sights in Americans’ social media newsfeeds. Still, public service organizations are concerned that these messages may be reaching their intended audience too late — or not at all.
To maximize the reach of this vital messaging online, Biocomplexity Institute of Virginia Tech researchers recently shared their latest findings on the factors that have proven to help opioid-related social media posts go viral to policymakers and emergency responders from across the state at the 2018 Virginia Public Health & Healthcare Preparedness Academy in Roanoke.
Pulling data from more than 5,400 opioid-related Twitter posts made by users in Virginia and Georgia, the Virginia Tech team searched for common factors among the most-highly shared content. While this research is still in its early stages, three clear takeaways have emerged from the study so far.
For public service organizations, pose the opioid epidemic as a problem that community members can help solve. “Law enforcement agencies sharing positive messages that invited the community to take part in some fun, anti-drug-themed activity received 83 times more views than law enforcement posts that focused on the negative legal consequences of opioid use, such as reports on recent arrests,” said James Schlitt, a graduate research associate in the Biocomplexity Institute’s Network Dynamics and Simulation Science Laboratory and Ph.D. candidate in the genetics, bioinformatics, and computational biology program. “In general, the data seem to suggest that people like to see their local government organizations taking concrete steps toward decreasing drug use.”
Use news sources to emphasize emerging threats. “Posts from local news outlets received seven times more views than other sources when they highlighted the health risks posed by opioid abuse in their area,” said Schlitt. “Based on our current data, if public health officials want to get the word out about tainted street drugs that are putting citizens at an elevated risk for a fatal overdose, their message might get more attention if it’s relayed through a trusted local news source.”
Start your informational campaigns at midday and keep sharing. “When a tweet is first posted, it can easily reach audiences across the country, but after about five hours, the average post’s reach effectively falls to zero,” said Schlitt. “For urgent messages related to the local opioid situation, we recommend public health agencies make their posts when the highest number of active Twitter users are online — around noon — and re-share in the early evening.”
These preliminary findings build upon more of a decade of Biocomplexity Institute research developing tools to help policymakers identify populations that face the highest risk of infection during disease outbreaks like flu or Ebola. With more information on drug prescriptions, sales, and usage being made publicly available, researchers hope to apply these same predictive modeling techniques to the U.S. opioid epidemic.
“Opioid use is such a widespread problem, it’s been a challenge trying to narrow down the factors that put communities at risk of addiction,” said Van Truong, a graduate researcher based in Virginia Tech’s College of Agriculture and Life Sciences. “These first steps make us hopeful that we’re finding meaningful patterns in the data. With the stakes so high, any advantage gained is worth fighting for.”
Written by Dan Rosplock