Tofunmi Omiye

High Schoolers Show How Data Analysis Can Shape Public Health Policy

Tofunmi Omiye

High Schoolers Show How Data Analysis Can Shape Public Health Policy

Remember the game where you’re given several disparate items and you get two minutes to make up a skit using all of them? Well, that’s not too different from what happened to Nigam Shah, MBBS, PhD, during late spring 2020.

Shah is professor of biomedical informatics at the Stanford University School of Medicine and associate chief information officer for data science at Stanford Health Care. In June, he received emails from four high school students looking for research experience during their summer vacation. The students approached Shah because the pandemic forced a temporary shutdown of programs such as the Stanford Institutes of Medicine Summer Research Program (SIMR), which is the primary mechanism by which faculty accept high school interns.

At the same time, Shah was in touch with Tofunmi Omiye, a physician in Nigeria, who had been admitted to a master’s degree program in health policy at Stanford but was delayed entrance because of the pandemic. Omiye said he was seeking a research assistant position to help fund his Stanford education and asked if Shah had a research project he could work on.

“At that point, I wondered if I could combine these two problems: Here’s a master’s student doing a research project, and here’s a bunch of kids wanting to do something so they’re not bored out of their minds sitting at home all summer long. So I asked each of them if they would be OK working on a team project as opposed to me working with them one-on-one, and they all said yes,” Shah says.

The result exceeded all expectations and led to a March 2021 presentation during the American Medical Informatics Association (AMIA) 2021 Virtual Informatics Summit.

The Backstory

The four students had written to Shah completely on their own to inquire about summer research opportunities in his lab. Logan Pageler (son of clinical professor of biomedical informatics research Natalie Pageler, MD) and Nikhil Majeti (son of professor and division chief of hematology Ravi Majeti, MD, PhD) were given Shah’s name by their parents. The other two—Ron Nachum and William Ding—found Shah independently.

As Ding says, “I first saw Professor Shah in one of Stanford Medicine’s virtual town hall videos about COVID-19. I then found his various projects on Stanford’s website and reached out to him to see what I could help with.”

Not long after that, Omiye, who holds the equivalent of an MD degree from the University of Ibadan in Nigeria, approached Shah and several other Stanford faculty to ask about serving as a paid research assistant. Omiye noted that Shah’s focus coincided with Omiye’s interests in big data and medicine.

When Shah asked if Omiye was willing to serve as an unpaid mentor to four high school students on a project he was thinking about, the master’s candidate jumped at the opportunity.

The Assignment

Shah gave his “assignment” to Omiye and the four high school students during June 2020, when COVID-19 cases were decreasing and some states were beginning to loosen shelter-in-place orders.

“I posed a research question to them, asking if we could use public data to identify the effect of various states’ reopening orders. That is: ‘Using public data, can we identify which reopening orders are good and which are bad?’” Shah says.

Shah gave a few hints at how to approach the problem and then left the mentor and the four students alone.

“I pointed them to a few data sets and told them that the best way to figure out whether a policy like masks, distancing, curfew, or whatever is working or not working is to look at how many people are going into hospital beds, because an increasing number of inpatients puts a burden on the entire system. So we figured we would count hospitalizations and ask questions like ‘If people are ordered to wear masks, what happens to hospitalizations?’ and ‘If you allow restaurants to reopen, what happens to hospitalizations?’” he says.

Omiye, who was working full-time as a medical intern in Lagos, Nigeria, created a framework with milestones for the project, and he arranged weekly meetings by Zoom so that he and the teenagers could discuss accomplishments and challenges in real time from their locations in California, Virginia, and Nigeria.

The students devised and maintained schedules for completing the work, they created a WhatsApp group where they would post daily progress notes, and they used Google Docs to keep meeting minutes that they would share with Shah.

Working Quickly

Less than two months later, the team had developed a presentation for the members of Shah’s lab.

Using their own computers and publicly accessible data, they learned how to figure out how long it takes for a policy decision made today to affect the rate of hospitalizations down the road.

Specifically, they investigated the effect of reopening orders on COVID-19 hospitalizations in the U.S. They discovered that reopening restaurants/bars and houses of worship correlated with the most significant spikes in hospitalization cases. “In the end, they determined that if you open up restaurants, that’s bad; and that’s exactly what all the famous public health scientists concluded four to six weeks later! So some kids who knew nothing about epidemiology used their computers and some data sets to match wits with the best in the field,” Shah boasts.

Furthermore, Omiye taught his mentees the basics of writing a scientific research paper and led the team into expanding the presentation into a paper worthy of submission to a peer-reviewed journal.

“These kids had never written a paper before in their lives,” Shah says, “but just a few weeks after the presentation to my lab, they completed a paper that was submitted to AMIA, and it was accepted.”

“In the end, they determined that if you open up restaurants,

that’s bad; and that’s exactly what all the famous public health

scientists concluded four to six weeks later! So some kids who

knew nothing about epidemiology used their computers and some

data sets to match wits with the best in the field”

“In the end, they determined that if you open up restaurants,

that’s bad; and that’s exactly what all the famous public health

scientists concluded four to six weeks later! So some kids who

knew nothing about epidemiology used their computers and some

data sets to match wits with the best in the field”

William Ding ponders comments made during a Zoom teleconference call with his fellow high school research colleagues and mentor Tofunmi Omiye

Camaraderie

Among the benefits for the high school students was the experience of teamwork.

“I really enjoyed working together with the team, with the virtual setting allowing us to work together from different parts of the country and share our knowledge in computer science, data analysis, epidemiology, and more,” Nachum says.

“I was so impressed at how hardworking these students were,” Omiye adds. “They were willing to respond to every comment from me, Dr. Shah, and the AMIA reviewers.

”What’s more, says Omiye, the reviewers were “really impressed by the quality of the project, and the feedback was overwhelming.”

Among the comments:

“I noticed that the authors are in high school. Well done! And kudos on a great effort on a very interesting study.”

“Analysis in this paper is clearly conducted and easy to digest as well as the limitations of the study.”

“The submission clearly presents how the methods are structured. The submission follows standard scientific writing.”

Summing up the experience, Omiye says that “we were five strangers from different parts of the world, and we just connected to build an impactful project.”

 Save as PDF