Pilot program gives underrepresented undergraduates hands-on experience in collaborative biostatistics
When Alexandra Hanlon, professor of practice in the Department of Statistics, spoke virtually to a group of Virginia Tech and Virginia State University students in February 2021, she had no idea it would lead to the creation of an innovative new summer program.
After describing the work being done at the Center for Biostatistics and Health Data Science, where she serves as the director, Hanlon was asked about internship opportunities for Virginia State University (VSU) students interested in learning more about collaborative statistics.
In light of a long-term partnership between the Virginia Tech statistics department and Virginia State’s math and economics departments, a pair of internships for VSU students was created. In the summer of 2021, one student worked with Hanlon and her team at the Center for Biostatistics in Roanoke, while another was housed with the Statistical Applications and Innovations Group in Blacksburg.
The summer internship planted the seed in Hanlon’s mind to create a more extensive program for underrepresented undergraduate students — and CUBE, which stands for Collaborative Undergraduate Biostatistics Experience, was born.
Hanlon envisioned CUBE as an initiative for the integrated Translational Health Research Institute of Virginia (iTHRIV) — a National Institutes of Health-funded Clinical and Translational Science Award program hub and collaboration between the University of Virginia, Virginia Tech, Carilion Clinic, and the Inova Health System.
An eight-week training program, CUBE was designed to give motivated, underrepresented minority undergraduate students the opportunity to engage in a full-time collaborative data science experience.
“Our goal with this program is to raise awareness of our profession and, in the long term, really diversify our profession,” said Hanlon. “We know there are many people out there — and many people right here in this building — who don’t know what we do as a collaborative biostatistics group.”
CUBE offers students an experience working with a real-world data set under the mentorship of experienced collaborative data scientists as well as clinical experts. This exposure can help students determine if they are interested in pursuing a career in collaborative applied data science while developing skills for the workforce or graduate school.
CUBE is built around four pillars: introductory biostatistics, training in coding in R, professional development, and a collaborative project.
The first five to six weeks of the program are focused heavily on training, particularly in learning the basic concepts of biostatistics and coding methods. They then transition into utilizing those methods to dive into their collaborative projects in the final weeks of the program.
Meanwhile, professional development opportunities — such as learning about career pathways, proper interview etiquette, and how to network effectively — and social events are scattered throughout the summer.
This summer, the career panel in particular was a big hit, as it gave students a chance to see the possibilities of a future in biostatistics.
“I really liked when we had a week of people in different careers in biostats,” said Kayla Williams, a rising senior from Ohio State University who participated in CUBE. “We had someone talk from the FDA and someone who talked from pharma and someone who's in academia. I got to see the different routes that I can go into and learn their paths, which are all different. I thought that was really interesting learning from them.”
Four students took part in the 2022 CUBE pilot program: Two were in residence at the Center for Biostatistics and Health Data Science in Roanoke, while two others were at the University of Virginia. All students received the same training and professional development opportunities, but the approach to the final collaborative projects varied between the two sites.
The students at Virginia Tech — Williams and rising junior Kinara Gasper from Harvard — worked together on a project examining predictors of ACL injury. Meanwhile, University of Virginia students Nhiya David and Jackie Morales worked separately on their final projects — one looking at how the COVID-19 vaccine affects antibody levels in dialysis patients, while the other focused on understanding pain and pain management for cancer patients.
Using information from the Healthcare Cost and Utilization Project database, Williams and Gasper examined whether certain demographic categories — such as race, sex, income, location, and insurance — could predict the occurrence of ACL injuries. With millions of records at their fingertips, they created data visualizations, proposed a statistical analysis plan, and compiled a final report and poster presentation.
Throughout the process of the collaborative project, the pair was mentored by Charlotte Baker, an assistant professor of epidemiology in public health in the Department of Population Health Sciences in the Virginia-Maryland College of Veterinary Medicine.
At the end of the program, Williams and Gasper had the opportunity to present their findings at the Summer Undergraduate Research Symposium in Blacksburg and the Summer Undergraduate Research Symposium at the Fralin Biomedical Research Institute at VTC in Roanoke.
“Even though [the program is] only eight weeks, I feel like I learned so much,” said Gasper. “I think it really allowed me to explore a field that I never thought I would want to go into. Since this world is so competitive now, having this program and being able to work with a small group was really helpful — doing it with the people who are from a similar background or have gone through similar things just made it even more worthwhile.”
The support of funding partners iTHRIV, Virginia Tech’s College of Science, Department of Statistics, the Fralin Life Sciences Institute, the Center for Biostatistics and Health Data Science, and the Institute for Society, Culture and Environment all contributed to the success of the CUBE pilot program. Hanlon is now seeking additional funding in the hopes of expanding the program next year and into the future.
Eventually, she would love to see CUBE turned into a multi-Clinical and Translational Science Award hub program with sites located across the country. With training modules already created for introductory biostatistics and coding methods, the entire group could share responsibility for the professional development series, while the collaborative project would be unique to each site.
“We want it to stay alive. We want to grow it,” said Hanlon. “We know we’ve been successful in exposing [our students to biostatistics] and we already know two of the four students have actually decided to switch gears and move into either biostats or coding as a result of the program. That’s pretty cool.”