Pill bottle
Friday, March 19, 2021
Amanda May

As the COVID-19 pandemic was spreading globally, University of Iowa master’s student Zachary Veigulis (BA14/MS20) spent the summer of 2020 studying Veteran’s Affairs (VA) patient data to determine what factors affected patient outcomes when they contracted the virus.

Veigulis was already a clinical informatics analyst for the VA Health Care System when he decided to pursue a Master of Science in Business Analytics (MSBA) degree at the Tippie College of Business. His analysis counted towards his coursework but has reached an audience far outside of the classroom (including a mention on CNN) because of his discovery of better outcomes in COVID-19 patients on low dose aspirin regimens. In February 2021, Veigulis’ analysis was published in the peer reviewed journal PLOS ONE.

According to the paper, “Among COVID-19 positive veterans, a preexisting aspirin prescription was associated with a statistically and clinically significant decrease in overall mortality at 14 and 30 days, cutting the odds of mortality by more than half.”

As the vaccine rollout gains speed in the United States, much of the world’s vaccine distribution is far behind, making affordable, accessible treatments like aspirin a potentially important global initiative.

Associate Professor of Instruction and Director of the Business Analytics Part-Time Graduate Program at Tippie, Patrick Johanns, oversaw the project.

Capstone projects are a requirement of the MSBA and are usually worked on in teams within the graduate program, according to Johanns. But because of the sensitive nature of the patient data, Veigulis worked with fellow analytics professionals at the VA to complete this project. While it added a level of complexity to the logistics of the project, the patient data was also the key to discovering the correlation.

“The project started looking at admitted patient’s Care Assessment Needs (CAN) scores (a metric the VA created several years ago as a measure of overall health) to determine if it would be a good predictor of mortality rates specifically for COVID-19 patients,” Johanns explained. “But because the VA provides over-the-counter medications to their patients, they had data on what of thousands of people were taking. Others have speculated on this, but nobody else had the data to test that in the United States.”

“We were uniquely positioned, due to access to national datasets, technical resources, and healthcare analytics experience, to assist with the pandemic response,” Veigulis said.

The team’s first assessment saw that CAN scores were indeed good predictors but they wanted to know which variables specifically were impactful.

“For the capstone project, we built the model from scratch using 70 variables, including aspirin use.”

Veigulis uncovered aspirin’s protective effect on the fourth of July weekend.

“Aspirin use kept showing up as significantly reducing risk of poor outcomes. I must have run it 20 times and it kept happening, he said. “I went over every single line of code to make sure it wasn’t something I was doing. That’s when I called my boss, VA physician Thomas Osborne, and said, ‘I think we’re on to something here.’”

“After rigorous analysis, we were able to conclude that this was, in-fact, a significant finding. It was exciting. We said, ‘Wow, we can really have an impact.’ This is novel and important—especially for this demographic (older, aspirin-taking patients),” Veigulis said.

It was a moment that he had envisioned since he was an undergrad intern with the VA.

“As an intern, I enjoyed using health-related data and translating that into insights that could assist with treatment,” he said. “Our projects now directly relate to addressing care-gaps that have been identified by VA clinicians, to ensure our work is relevant and addresses a real problem. Many of our projects result in direct patient care changes, which have immediate impact.”

While Veigulis hasn’t yet analyzed whether or how much VA physicians are prescribing more aspirin since his assessment was released in February, it has had at least one immediate effect. 

Associate Professor Johanns began taking low dose aspirin.

“My wife had been asking me to take it for years,” Johanns said. “And I believe in data.”

 

Zacharay Veigulis is now a data scientist with the VA’s National Center for Collaborative Healthcare Innovation (NCCHI).

Speak to your physician before beginning any drug regimen.