biostatistics

‘All of Our Paths Are Different’

Scarlett Bellamy, new chair of Biostatistics, outlines her approach to mentorship and her hopes of increasing the department’s visibility.

While the thought of more New England winters may have given her a moment’s pause, biostatistician Scarlett Bellamy has returned to Boston for a new challenge as the new chair of the Department of Biostatistics at the School of Public Health.

She spent six chilly years in Boston as a grad student but Bellamy’s lifelong love of mathematics started in the rural North Carolina town where she was raised by her grandparents until she was 10 years old. She grew up riding ponies and exploring the woods near her home and spent summers with an aunt who taught math at North Carolina Central University. In an interview with the American Statistical Association, Bellamy explained that her aunt would ask her to help prepare coursework for summer college classes, but Bellamy soon realized that her aunt was subtly teaching her algebra several years earlier than her peers.

After graduating from Hampton University in Virginia with a bachelor’s degree in mathematics, Bellamy headed north for her first stint in Boston to earn her doctorate in biostatistics from Harvard University. She left Boston for Philadelphia to accept a biostatistics faculty position at the University of Pennsylvania Perelman School of Medicine, and eventually rose to full professor and served as a senior scholar at the Penn Center for Clinical Epidemiology and Biostatistics (CCEB).

At Penn, Bellamy developed and implemented statistical methods for cluster- and group-randomized trials that have been used in longitudinal trials of behavioral modification interventions and multicenter trials that address health disparities. After 15 years at Penn, she moved six blocks across Philadelphia’s University City neighborhood to the newly renovated Dornsife School of Public Health at Drexel University.  

The move to Drexel added additional responsibilities to Bellamy’s expanding portfolio of research, teaching and advising, while also shouldering larger administrative roles as the inaugural director of graduate studies in biostatistics—developing the PhD program in biostatistics— and as the inaugural associate dean for diversity and inclusion. Her Boston connection was also rekindled in 2021 when she accepted a remote adjunct advising post at the Harvard T.H. Chan School of Public Health in response to students’ expressed desire for access to more diverse faculty advisors and mentors.

As someone whose career was shaped by the guidance of mentors, Bellamy understands the critical part that effective mentorship can play in someone else’s career. The importance of the job is not lost on Bellamy, who says she is still advising students at her previous institutions (and colleagues, in some cases) but mostly on an informal, unofficial basis.

“I think the hardest part is to get people, especially students, away from this idea that there’s a single prototypical path to success for them,” Bellamy says. “A lot of my job as an advisor is just reassuring folks that your path is, by definition, going to be exactly that, your path. It’s going to be different, it’s going to be the only one, and it’s that way for a reason. I wanted to give students the confidence and the comfort of understanding that all of our paths are different, and what we want to do with our lives post degrees is very different. And that’s okay.”

BUSPH is one of 10 schools that offer the Summer Institute in Biostatistics and Data Science (SIBS), an NIH-funded program to encourage students to explore careers in Biostatistics. I’m curious about whether early exposure to math and science in your background opened doors for you, or set you on your current career path.

As an undergraduate, I had a similar experience in a program like SIBS. I was an undergrad at Hampton University, an HBCU in Virginia, and I was very fortunate to be on a scholarship program. A requirement for that program was to conduct research during the summer in the hope that we would pursue an advanced degree in a STEM discipline as a math major. And being a native of North Carolina, I applied to and was accepted into a similar research program as SIBS for 10 weeks at the University of North Carolina at Chapel Hill [the Summer Pre-graduate Research Experience – SPGRE].

The “Cliffs Notes” version of the story is that, at that point, if I’m honest, I really didn’t have any interest in going to graduate school because it was outside of my experience and my imagination. I honestly only did those summer programs because it was a requirement for the scholarship. But at UNC, not only did I get turned on to research, but it was my first exposure to this discipline called biostatistics. And I was matched to a mentor who just happened to be in the biostatistics department at UNC who helped me connect my love of math to health. The rest is history.

My two prior home institutions were not fortunate enough to have a SIBS program or even a summer program like that, but we—and frankly, many graduate biostatistics programs—were very grateful for BU Biostatistics being one of the founding institutions to have such a program. So, while we never had our own in-house program, we benefited from the many trainees that came through SIBS programs. It was definitely to an applicant’s advantage to have gone through SIBS or a similar program because we could be assured that students with such experiences had a rigorous exposure to our discipline and a good sense of our profession. I’m very grateful for SIBS at BU and in other places being able to be an asset to many programs, not just the ones in which they were housed.

You’ve said that mentorship plays a really important role as one of your academic functions. What will your approach to mentorship be at BUSPH?

Over the arc of my professional career, I can see both ends of the spectrum. As an undergraduate trainee, I was operating from the perspective to take my mentors and their perspectives very, very seriously—to the point where I just did what they told me to do. That is not something that I would freely advise mentees to do today, but it worked for me and that’s exactly what I needed at the time.

So my mentoring approach has evolved to let the mentees do the driving; recognizing that some will be like me, needing and welcoming lots of guidance. And others, you just need to point them in the right direction and they’re happy to go and figure things out. With those being the extremes, there are also lots of people who are right in the middle. In terms of a mentoring approach, I hope to provide lots of good information and let the mentees do the driving in terms of defining what works, and what would be helpful to each of them. I also fully recognize that my style of mentoring may not be the best fit for all, so I also understand if after some exploration, if people would like to seek mentorship elsewhere. ‘Fit’ is critical.

What other aspects of what you experienced at your previous institutions would you like to incorporate here?

Truthfully, many of the things that I loved about Drexel—and Penn, although it was obviously a little bit farther back in terms of timing—I saw already at BU.

When I joined the faculty at Penn, my first year was only the second year that we had an incoming cohort of PhD students. So it was very early in the development of the program—I was the 13th faculty member in the department. When you think about economies of scale, the only way we were able to make that work was to partner in a very intentional way with the Department of Statistics. At Penn, we also had the benefit of geography with the Biostats and the Statistics department, you could walk between one and the other in under 10 minutes. I only recently learned, in terms of how Biostats works at BU, that there’s a similar partnership here, and I think that in many places that’s unusual. I think that’s a nice model in terms of sharing the load and using resources across campuses, across schools, and across units within an institution.

In terms of Drexel, it was really nice to return to a school of public health setting. And what I love about both Drexel and BU is this orientation around social justice and a focus on practice and being good community partners. That is very much aligned with how I see myself professionally as a public health practitioner. Those are all things that I was drawn to.

Almost 20 years ago, you worked on a project looking at respiratory illness data from children in East Boston who were suffering from asthma at a much higher rate than elsewhere in the city. Are there any lessons that public health researchers can glean from similar long-term community-centric projects?

One of the things that I love about being a statistician is, that to the extent that I have time and bandwidth, I get to play in a bunch of sandboxes. A common thread through my more recent work is to think about working on problems that are, broadly speaking, health disparities problems. In both at my time at Penn and at Drexel, I continued to be very active in looking at disparities as they relate to HIV infection and prevention work. As a statistician, my training and expertise are in clinical trials. I take my skills and experience in designing effective clinical trials and apply it to problems, looking at reducing the risk of HIV infection in a number of populations, but usually ones that are disproportionately affected, such as African Americans and people who inject drugs.

What public health folks can glean from that work is that we have to evolve. What I like about working with clinical trials is that they allow us to take various innovations and see if they work in a prevention context. For example, one of the projects that I’m still involved with, with collaborators at Drexel, is women who inject drugs. In Philadelphia, like many big cities, we have a very bad, very significant opioid injection problem. Many times, there is some previous trauma that these women have experienced. Our trial is looking at whether we can develop interventions that address the trauma, and if that would, along the way, reduce their HIV risks. To our knowledge, this is one of the first studies that are looking at this in a clinical trials context.

Can you speak a little about your current research projects?

I view myself as a collaborative statistician. That’s my comfort zone. That’s my happy place. Unlike others who might be more theoretically oriented, my work is less about developing new statistical methods and more about applying existing methods in rigorous ways to problems that I find interesting and/or to work with people that I think have interesting problems.

Rather than bringing along many of the existing research collaborations that I had been involved with at Drexel as I transitioned to a new leadership role at BU, I was very thoughtful about using this time to transition projects to folks who were at Drexel. This will allow me to have opportunities to explore making new collaborations here.

In terms of current and ongoing work, there are one or two projects that I have brought to BU—including women who inject drugs and HIV risk—but I was very deliberate in leaving open time to make some new collaborations at BU SPH.

What would you like the department to be doing in six months or so? Is there anything that you have identified as a priority?

What I think we should be doing is getting clarity on how we define ourselves. I think there may be some internal collaborative reflection to help define how we want people outside the School, and outside that bigger, university public health landscape, to see our department here at BU Biostatistics. There will definitely be some deliberate thinking about who we are now, and envisioning where we want to be in terms of the identity we want our BU brand to have. It defines how we’re going to recruit new faculty, and how we might refine our student recruitment strategy.

What ideas do you have to increase the visibility of Biostatistics?

A large, “pie-in-the-sky” goal is to become the place where people organically go for all things analytical across the university. If someone has any questions or is looking for expertise around any aspect of data analysis, BUSPH Biostatistics is known as the hub where that expertise lives on campus. It is also my goal for BUSPH Biostatistics to be a leader in reimagining and expanding our thinking around who ultimately becomes a biostatistician. We remain committed to diversifying our field because we know it makes our research and our science better.