CHIBE Q&A with Kristin Linn, PhD, MStat

Get to know Kristin Linn, PhD, MStat, Assistant Professor of Biostatistics at the University of Pennsylvania, and a CHIBE-affiliated faculty member, through this CHIBE Q&A below.

What projects are you working on right now?

In my dissertation, I developed statistical methods for optimally tailoring interventions to individuals. Recently, I’ve been applying these methods to data from the STEP UP randomized trial (joint work with Mitesh Patel and Amol Navathe). Our goal is to develop a personalized decision rule that uses demographic, social support, and/or personality traits to assign participants to one of three interventions in a way that maximally increases individuals’ physical activity levels. I’m particularly excited about the potential for impactful methodological advances based on this type of data. For example, developing new statistical methods for precision medicine that appropriately handle missing data could lead to better insights and more effective intervention strategies in the future.

I’m also working with Amol Navathe to evaluate the effects of a new primary care payment system on cost and quality measures, and in a separate project we are investigating whether there is any empirical evidence of racial unfairness in a risk score used by the Veteran’s Association (joint with Ravi Parikh and Helen Yan). I have ongoing work with Atheen Venkataramani, Kevin Volpp, Kristen Underhill, Erica Dixon, Will Ferrell, and Lizzie Bair related to a Section 1115 Medicaid demonstration waiver that was planned but not implemented by the Commonwealth of Kentucky. Outside of CHIBE, I am an Associate Director of PennSIVE, a biomedical imaging group at Penn. A few of my imaging projects include: addressing confounding in high-dimensional data; harmonizing images from multiple scanners; fusing information from multiple imaging modalities; and personalizing transcranial magnetic stimulation treatments for depression and anxiety.

What do you find rewarding about your work?

Power calculations… kidding! One of the most satisfying feelings is when you find a bug in your code or otherwise identify an error in an analysis and the results suddenly make way more sense after you fix it. But the best part of my job is interacting with and learning from others. One reason why I enjoy working with CHIBE investigators is that they are incredibly skilled at communicating technical, scientific work to audiences with different backgrounds. I experienced this first-hand as a new faculty member when I traveled with a team of CHIBE investigators to meet with policy makers in Kentucky. I was inspired by the conversations where Kevin, Atheen, and Kristen beautifully sold the value of randomization, ensuring our team’s evaluation of the state’s new Medicaid program would be rigorous and impactful.

What’s something that people may not know about you?

I discovered statistics while majoring in Music Performance (Euphonium) at the University of Michigan. My roommate at the time, who is now the tuba player in the Philadelphia Orchestra, told me statistics was the easiest way to satisfy our lab science requirement, since the lab portion was a short, weekly computer lab. I loved the content of the class so much that I took all of my remaining elective courses in the statistics department. I also ice skated outside for the first time in Michigan. When I was in middle school I figure skated competitively, and last fall I started practicing again at Penn’s Class of 1923 rink. My favorite jumps are the loop and the flip!