A faculty member from Oakland University William Beaumont School of Medicine played a significant role in a recent study that found face masks to be effective in limiting the spread of COVID-19.
Ramin Homayouni, Ph.D., professor, Department of Foundational Medical Studies and founding director, Population Health Informatics, was a co-author on COVID-19 seropositivity and asymptomatic rates in healthcare workers are associated with job function and masking.” The paper was published in Clinical Infectious Diseases.
The study was based on aggregate data collected from 20,614 employees who work at Beaumont Health’s eight hospitals in Southeast Michigan.
The research concluded that people who were exposed to COVID-19 with no masks had an 18 percent risk of getting infected. That dropped to 10 percent for people wearing N95 masks.
Homayouni’s role was ensuring that the data was accurately aggregated from multiple sources and validated before being securely passed to the statisticians for further analysis.
“You hear a lot about the era of interdisciplinary research and the importance of interdisciplinary collaborative teams, but it’s really hard to do effectively,” he said.
“What we’ve created is the true essence of interdisciplinary science, involving IT experts, data scientists, and clinicians. It’s very gratifying.”
Having such data — and having it quickly — was important, especially as the COVID-19 pandemic continues, said lead author of the research paper, Matthew Sims, M.D., director, Infectious Diseases Research, Beaumont Health, and associate professor of Internal Medicine, OUWB.
“It’s nice to have proof masking really works,” said Sims. “Masks play a vital role in protecting people and can dramatically reduce the risk of contracting COVID-19.”
Delivering information
Homayouni |
Homayouni joined OUWB in December 2018, as the founding director of Population Health Informatics. He is responsible for developing both graduate and research programs focused on Population Health Informatics in collaboration with various departments at Oakland University and Beaumont Health.
Informatics is the process of deriving information from data. In Population Health Informatics, data comes from sets of individuals from different communities. The information revealed by the data is used in many ways, including determining best practices with regard to health care.
“Our goal is to deliver information to the provider at the point of care...what are they going to do about a specific patient or community of patients?” he said. “It’s clinically focused.”
With the COVID-19 study, Homayouni said the research team received support from the highest levels of Beaumont Health as well as donors who contributed more than $5 million in just a few weeks.
Such support opened the door for rapid collection of blood samples from more than 20,100 employees across Beaumont Health, which includes eight hospitals in Southeast Michigan.
Homayouni’s job was to oversee the process of managing the data associated with the blood samples so that it was not only organized and uniform, but valid.
“The data didn’t get passed to the statisticians until my team said it was accurate and usable data,” he said.
A team of sophisticated data scientists as well as machine learning and database experts from a company called Quire were contracted to integrate and securely store the data and to support the research team. (Homayouni started Quire about 12 years ago and said the contract cleared all legal hurdles without issue in late March, during the first surge of COVID-19.)
“The motivation was there, the money was there, and we had the data and computing people —my job was to connect the dots,” he said. “It’s all about creating the pipelines to maintain data integrity so that people can ask questions and get to the answers they want as fast as possible.”
In addition to Sims and Homayouni, other authors of the paper with connections to OUWB were Gabriel Maine, Ph.D., an immunologist and associate professor of Pathology, and Richard Kennedy, Ph.D., professor and associate dean for Research.
The team engaged in labor intensive work, said Homayouni, that included a 20-day stretch where the team was working 10-12-hour days with no weekends. It wasn’t uncommon for the team to have conference calls of up to four hours a day, especially early in the process. Regular calls continue to this day.
“There was so much data and it was so pressing…we felt it was important to get the information out as soon as possible,” said Homayouni
First of many
According to Sims, the study yielded three primary takeaways (in addition to the specific stats regarding mask use):
- Masks do play a major role in protecting people and dramatically reduce the risk of infection.
- For frontline caregivers, job duties played a significant role in defining who was at greater risk of contracting COVID-19.
- People with higher levels of exposure were more likely to get it.
Sims said to expect more findings from the data to soon be published.
“Our research team had the first of what we hope will be many BLAST COVID-19 (Beaumont’s Large-scale Analysis of Serological Testing in COVID-19) papers accepted for publication, sharing with the medical and scientific communities some of our initial findings and how they may apply to the treatment and prevention of the novel coronavirus,” said Sims.
“We continue to track and test nearly 2,000 study participants who were antibody positive in the initial round of assessment, in pursuit of a better understanding of how long antibodies last and how much protection they are providing.”
Homayouni said he’s excited about the continuation of the work, as well as the possibility it holds for OUWB’s Population Health Informatics program.
“We’ve assembled a kind of Dream Team…we have the right mix of IT experts, statisticians, data and laboratory scientists, and clinicians. We continue to meet regularly and talk about everything from detailed logistics to where we are heading next,” he said. “That’s the beauty of science…when you answer one question it opens up even more questions.”
For more information, contact Andrew Dietderich, marketing writer, OUWB, at [email protected].
To request an interview, visit the OUWB Communications & Marketing webpage.
NOTICE: Except where otherwise noted, all articles are published under a Creative Commons Attribution 3.0 license. You are free to copy, distribute, adapt, transmit, or make commercial use of this work as long as you attribute Oakland University William Beaumont School of Medicine as the original creator and include a link to this article.
