Kenneth M. Lee, PhD
Kenneth M. Lee, PhD
Kenneth M. Lee, PhD, is a postdoctoral researcher in the Department of Biostatistics, Epidemiology and Informatics, whose work is shaped by both personal experience and deep methodological curiosity. His research focuses on palliative care interventions and the development of statistical methods for cluster-randomized trials, with an emphasis on causal inference. He recently spoke with us about the motivations behind his work, his experience in the DBEI community, and the ways he finds balance outside of academia.
Read on to learn more about Kenneth in this Q&A.
Can you tell us about your current research and what inspired you to pursue this area?
I had a terminally ill family member pass away in 2021. Bitterly, due to COVID-19 travel restrictions, I was unable to see her before she passed. As a then first-year PhD student in biostatistics, this spurred a deeply rooted interest in palliative care research, which I still contribute to at Penn.
My statistical methods research focuses on the design and analysis of clinical trials, with a particular focus on applying causal inference theory and methods to improve cluster-randomized trials (CRT). The CRT is an essential alternative to the individually randomized trial and refers to the collection of study designs where randomization is carried out at the cluster level (such as a hospital, clinic, or school level), oftentimes for pragmatic reasons. In the past couple of decades, the growing interest in the standard parallel CRT has spawned multiple novel CRT designs, including the stepped-wedge CRT, parallel-with-baseline CRT, and the cluster-randomized crossover trial. Importantly, the multilevel sampling, longitudinal data structure, treatment heterogeneity, typically small sample sizes, and other specific properties of these CRT designs present unique analytic challenges. These and other issues that commonly arise in trial implementation have driven significant methodological development in recent years, revealing different strengths and weaknesses for these CRT designs and their analytic options.
How does your research intersect with real-world challenges?
I began my PhD in biostatistics, researching statistical considerations that emerged during a cluster-randomized trial of a palliative care co-rounding intervention on patient outcomes in hospital wards. As a current biostatistics postdoctoral researcher working out of the Palliative and Advanced Illness Research (PAIR) Center, I have continued to collaborate on multicenter trials examining the effect of palliative care interventions on patient-centered outcomes. Overall, my formal biostatistical methods research is directly driven and supported by these collaborations.
What do you find most rewarding about working in the Department of Biostatistics, Epidemiology and Informatics?
I love how interdisciplinary DBEI is. This interdisciplinary nature isn’t just at the departmental level, but also the individual level. Everyone is so knowledgeable and so curious to learn, discuss, and contribute regardless of their specific expertise!
Can you share a significant recent project, publication, or professional recognition that you are particularly proud of contributing to or achieving?
During my time as a post-doc at Penn, I have been working on the intersection of causal inference and cluster-randomized trials. This is EXACTLY the type of statistical methods research that I was hoping to do during my post-doc. Here, I have been able to learn from and collaborate with so many amazing researchers at Penn and other institutions around the world! So far, this has led to a published paper examining “How Should Parallel Cluster Randomized Trials With a Baseline Period be Analyzed?—A Survey of Estimands and Common Estimators,” another manuscript expanding the estimands framework for cluster-randomized crossover trial designs, and more ongoing causal inference/statistical methods projects.
What advice would you give to students or early-career professionals in your field?
A lot of researchers hear the advice, “It’s a marathon, not a sprint”. Others recall the tortoise and the hare. Yes, these are cliches. Still, I have found this to be the singular most important advice to sustaining my own research in the long run. Many endurance runners build up their stamina and avoid muscle burnout by training in “zone 2”, where they’re maintaining around 60-70% of their maximum heart rate. I think the same principles apply to research to build up your knowledge while avoiding mental burnout.
I’m still an early-career professional. That said, I consistently had a great time working on my research during my PhD and the good times are still continuing now during my post-doc. I owe so much of this joy to finding my “zone 2”.
What is an interest or pastime that you enjoy outside of academia?
I love tennis and boxing. I’ve been playing tennis and boxing for a while and find a lot of similarities between the two. Both are one-on-one competitions that require great focus and mental strength. They can also be incredibly elegant to watch, with many of the best athletes in both having dancer-like footwork and musician-like timing.
While I turn to these sports to enjoy my time away from work, statistical questions still sometimes pop up. Two of these statistical questions: “whether the jab is truly the most important punch in boxing” and “how did Roger Federer win 82% of his matches while only winning 54% of his points played” have led to two separate articles in Significance magazine that I’m very proud of!