A Life of Pi and Poetry
Research Professor of Biostatistics
July 30, 2018, Biostatistics, Faculty, Big Data, Cancer, Chronic disease, Epidemic, Genetics, Global Public Health, Health Disparities, Mentorship, Research, Statistics
Since childhood, I've loved numbers. I don't know exactly where it came from. My father was a professor of English literature, and my mother taught Bengali language. I grew up listening to my father recite Shakespeare, Shelley, Keats, and Byron from memory and spending long summer afternoons reading Tagore and other poets of the Bengali Renaissance period.
A high school teacher introduced me to the beauty of mathematics. Calculus was my favorite subject, and the concept of limit going to infinity fascinated me. The commonality between poetry and mathematics intrigued me most. Both are rooted in the actuality of our world while taking our imaginations far beyond. Both demand intense creativity. Understanding pi or infinity takes a tremendous leap of imagination. As Einstein remarked, mathematics is the "poetry of logical ideas."
Throughout my adolescence, I spent hours doing mathematical puzzles for fun. I took a statistics class and loved it, and was even hailed as the math "genius." I followed my teacher's suggestion to study at the Indian Statistical Institute (ISI) in Kolkata and was the only female in a cohort of 22 math geniuses. In that pool of incredibly talented students, my confidence took a hit. I knew I had some work to do.
Far beyond probability puzzles, the fun of statistics is how data can speak through thoughtful, rigorous analyses applied to real-life problems.
As I was embarking on my graduate student life in the US, I realized my view of statistics as a field had been quite narrow. Far beyond probability puzzles, the fun of statistics—and its profound mathematical underpinnings—is how data can speak through thoughtful, rigorous analyses applied to real-life problems, such as those that exist in clinical and population-wide health care.
At first, I would panic at the thought of having to work with messy real data that did not conform to textbook examples, but with experience came understanding and purpose. Fresh out of graduate school and as a faculty member at Wayne State University, I was the lone statistician working on an NIH-funded prostate cancer project. I sat in the weekly meetings with two dozen medical doctors wondering how I could do statistics for them without understanding what they were talking about. I knew I had two options—quit or learn the language. I began sitting in on biology classes, shadowing colleagues in clinics, and doing extra reading to understand the science.
The Beauty of Team Science
It paid off, and my career in biostatistics began to flourish. I became involved with Detroit's Surveillance, Epidemiology, and End Results (SEER) cancer registry, one of the oldest population-based cancer registries in the US. Utilizing the uniquely large African American population in Detroit's SEER registry, we conducted vital research on racial and ethnic disparities in cancer care and outcomes, developing church-based screening programs for early detection, understanding the role of race in survival, and designing studies to disentangle the effects of cancer biology from the impact of socioeconomic status and access to care on outcomes.
I became increasingly cognizant of the importance of sound statistical design and methodology in addressing a range of health problems and of my ability to contribute to the field. Three years later, while presenting at an American Society of Clinical Oncology meeting, a professional in the audience remarked, "You seem to know a lot about statistics for a clinician."
It was good to be affirmed in my understanding of clinical settings, especially because my statistical research continues to be inspired and motivated by the needs of the underlying scientific questions. With colleagues at Michigan Medicine, I now study population-wide variations in health care delivery and outcomes, primarily in cancer and pediatric cardiac care. We recently looked at the use of imaging tests after primary treatment for thyroid cancer and found a marked rise since the late 1990s in the use of such tests, which has increased treatment for recurrence with no clear improvement in downstream survival. Our study showed that more imaging might not equal better care and that the benefits of imaging likely do not outweigh financial costs, heightened patient anxiety, and risks of patient harm from treatment. Statistics empowered us to argue, with great accuracy, against unnecessary imaging and for weighing treatment plans against patient risk.
This and many other successes are attributable to the beauty of team science at Michigan. My regular collaborations with clinical and health services researchers have led us all to insights that directly improve quality of care. In my role as director for biostatistics in the Center for Healthcare Outcomes and Policy (CHOP), I mentor fellows and junior faculty and direct the design, planning, and analyses of collaborative research studies. Our projects employ advanced statistical methodology, including the use of novel and rigorous statistical methods that strengthen our ability to make inference. Clinical data is complex, and statistics provides fundamental frameworks for uncovering evidence that can advance policy change. As a faculty member in Global Public Health, I am involved in research and educational partnerships in South Asia focused on raising cancer awareness and screening for thalassemia in India and breastfeeding and immunization in Bangladesh.
As beautiful as pure logic is, it's much more rewarding and meaningful to apply that beauty to research.
Outside my work in public health, I wear another hat—I am a singer and a writer. I've released multiple music albums and published poetry, and I perform regularly in the US, India, Bangladesh, and the UK. I am also executive director of the nonprofit art foundation Tagore Beyond Boundaries, which brings music from my culture and country to Ann Arbor public school children. And I'm a biostatistical scientist working to improve cancer care delivery and outcomes for people suffering from a terrible disease.
The sciences and the arts complement each other well. In my own life, one enriches the other every day. Equations and numbers might be my first love, but as beautiful as pure logic is, it's much more rewarding and meaningful to me to apply that beauty to research, with the possibility—the probability even—of helping people live healthier lives.