Professor builds mathematical simulations to save lives

May 31, 2022

RESEARCH & INNOVATION

“What if?” That type of question dogs the pros trying to solve the country’s opioid crisis. What if we opened an overdose prevention site here — or there? How many lives could we save?

Joanna Wares, associate professor of mathematics and chair of the new mathematics and statistics department starting in July, knows these urgent questions well. She builds models for comparing different interventions that would take years to pilot and study in the real world. 

“The numbers keep rising,” Wares said, pointing to grim federal data that more than 80,000 Americans died from opioid overdoses in 2021. “Part of the reason is the introduction of fentanyl into street drugs, which causes even more overdose deaths than just heroin alone.”

Wares’ research originally focused on infectious disease policy until University of Pennsylvania contacts asked her about predictive opioid intervention modeling for a collaborative effort in Philadelphia. She learned that in the Kensington neighborhood, kids witness passers-by with needles still in their arms, and residents are administering overdose reversal drugs on the street. 

Lawyers with a local legal services nonprofit proposed an overdose prevention site, a clean, well-lit space staffed by trained health workers where people can inject illegal drugs. Such facilities in Canada and Europe have been extremely effective at reducing overdoses, Wares said. Communities that have the have these prevention sites also see overall drug use go down, because workers can link clients to treatment and services.

Harm reduction advocates sought to measure the specific effects a prevention site could have in Philadelphia, but there weren’t any legal sites open in the U.S. at that point.

The numbers keep rising. We're trying to figure out what type of interventions might help.
headshot of Joanna Wares
Joanna Wares

Associate Professor of Mathematics

Wares collaborated with The College of New Jersey, Pomona College, Philadelphia’s Department of Public Health, and the University of Pennsylvania to create models with drug use and overdose datasets from the city.

Based on the simulation results, Wares said, “We suggest that you need to have multiple centers in multiple places with larger capacities to really make an impact.”

Recently Wares and associate professor Zeynep Teymuroglu at Rollins College received grant support from the Center for Undergraduate Research in Mathematics to teach undergraduates differential equations models of opioid interventions. These models are much faster than the ones currently being used for opioid intervention research, Wares explained. 

Most students learning the equations are computer science and math majors who are more familiar with theoretical studies.

“We’re going to be looking at some Virginia data and some Florida data, trying to figure out what type of interventions might help,” Wares said. “This shows them how to apply what they’ve learned at the University of Richmond and at Rollins College to answer questions being asked in the real world.”