Our Careers in Statistics: A Most Enjoyable Ride
Jonas Ellenberg is a professor of biostatistics at the University of Pennsylvania. For 23 years, he served in progressive leadership positions in the biometry branch of the National Institute of Neurological Diseases and Stroke at the National Institutes of Health. Following his service there, he spent 10 years as vice president and head of biostatistics at Westat. He is an ASA Fellow and past ASA president.
Susan Ellenberg is a professor of biostatistics at the University of Pennsylvania. Prior to joining in 2004, she held leadership positions at the National Institutes of Health and U.S. Food and Drug Administration. She is an ASA Fellow and recipient of the ASA Founders Award.
Susan: Both of us slipped into statistics more accidentally than deliberately.
Jonas: My family business background, steeped in the New York City 7th Avenue textile trade, led me to the Wharton School at the University of Pennsylvania as an undergraduate. I chose statistics as a major, as I had always enjoyed math. Wanting to earn extra money while at Penn, I signed on to the University of Pennsylvania Periodic Health Examination project under the tutelage of Stanley Schor (famous for the “Mystic Statistic,” a regular JAMA piece aimed at teaching clinicians some basic statistics). This experience introduced me to the application of statistics to medicine.
Wharton, in those days (1959–1963), did not have the mathematically sophisticated curriculum it has today and I found myself inadequately prepared for the rigors of the graduate program in mathematical statistics at Harvard, to which I had been accepted. To fill the gaps, I took summer courses prior to my first year and enrolled in the undergraduate advanced calculus course during my first Harvard semester. I remember vividly the Harvard undergrads writing up their homework sets on the back of computer output during class while I spent many long nights on such difficult tasks. Meanwhile, my classmates were surging ahead in math stats. I eventually caught up and was privileged to work with W. G. Cochran as my thesis advisor.
Susan: Growing up, I never thought about any career other than teaching and, by high school, had selected math as my subject. I pursued a Master of Arts in teaching degree following college graduation and happily taught high-school math for three years, first in Cambridge and then in Maryland after Jonas completed his PhD and took a position at the NIH. My plan at that time—have two children, be home for 7–8 years until the youngest was in pre-school, return to teaching—was perturbed by Janet Wittes, a graduate school classmate of Jonas’, who was also in Bethesda working as a research associate for the legendary Jerry Cornfield while her MD husband completed his military commitment as a public health service physician at the NIH. Janet and Jerry needed some programming support, and I agreed to help out for a few months before our first child was due. A few months extended into many years, during which I entered the PhD program in statistics at The George Washington University and moved away from pure programming to more statistical tasks at the GWU Biostatistics Center.
Jonas: My first position out of graduate school was at the neurology institute (NINDS) at the NIH. I worked with medical colleagues on evaluating prenatal, perinatal, and early developmental risk factors for cerebral palsy and convulsive disorders from the NINDS Collaborative Perinatal Project (CPP) database (recruitment 1959–1965). This observational database was extraordinary for its time and even today. Sixty-thousand pregnant women were followed through pregnancy and their offspring followed through eight years of life. This was in the early ’70s and my first encounter with very large data sets and the unique statistical issues they presented. Lugging around 11’x14′ continuous computer printouts is a charm of yesteryear.
One element that highlights the meticulous planning of the CPP was the inclusion of a CPP nurse in the delivery room who was responsible only for capturing the critical data during delivery and birth (no responsibility for patient care). My collaborations with Dr. Karin B. Nelson (pediatric neurologist at NIH and now emerita) forced the rethinking of many established medical paradigms. For example, our work showed that a febrile seizure, a fairly common occurrence in infancy, was only an important risk factor for epilepsy or seizure disorders in a very small and well-defined subset of children. This result effectively discouraged the then-common practice of treating all children with febrile seizures with neuroactive drugs for extended periods. We also showed that cerebral palsy was due largely to prenatal factors and not problems in labor and delivery (thereby enraging the cadre of lawyers whose incomes sprang from litigation against obstetricians.)
My ‘class’ at NIH included (among others) David Byar, Dave DeMets, Mitchell Gail, Barry Margolin, Richard Simon, James Ware, and Janet Wittes. This group continued the exciting and critical contributions to statistical science begun by the early giants of NIH biostatistics: Jerry Cornfield, Harold Dorn, Sam Greenhouse, Max Halperin, Nathan Mantel, and Marvin Schneiderman.
Susan: Jerry Cornfield died in 1979 of pancreatic cancer; during his illness, efforts to obtain new projects slowed and I found myself in need of a new position. I joined the then-fledgling (now thriving) EMMES Corporation in 1979 to take responsibility for some National Cancer Institute (NCI)–funded cancer studies. I loved this work and, after three years, had the opportunity to learn more about the methodology of cancer research when I took a position with Richard Simon at the NCI. That was the beginning of an exciting and rewarding 11 years at NIH, six at NCI, and five at NIAID, where I became the first biostatistics branch chief in the division of AIDS at the very beginning of AIDS drug development and at the height of the AIDS activism movement. It was an extraordinary time (well depicted in the 2012 Oscar-nominated documentary, “How to Survive a Plague”), when statisticians reached out to the activist community and initiated a highly productive period of mutual education about trial design and community needs that led to better accepted studies and advances in AIDS therapeutics.
Jonas: After 26 years at NIH, I moved to Westat as vice president and head of biostatistics. Westat is an employee-owned research organization providing services primarily to federal government agencies, including NIH. There, I was provided a crash course in grant writing and jumped into HIV/AIDS research and issues of reliability of patient reporting. Although Westat is a profit-making employee-owned organization, its spirit of collaboration, intellectual curiosity, innovation, and integrity made it feel much more like an academic center than a business.
Susan: In 1993, I had a new opportunity—to head the Office of Biostatistics and Epidemiology at the U.S. Food and Drug Administration’s (FDA’s) Center for Biologics Evaluation and Research. I had had many interactions with FDA through my work on AIDS drugs and decided to accept this offer. I knew FDA would be a very different environment, and it was, but there was no shortage of interesting and important problems to address—appropriate trial designs, use of placebos, vaccine safety, operation of data monitoring committees—all highly motivating. I stayed at FDA for 12 years, far longer than I would have guessed when I started.
At the end of 2004, I began yet another new career, as an academic. In this role, I’ve had the opportunity to learn many new things and to bring my federal experience to bear on research programs at Penn. I’m pretty sure this will have been my last career change!
Jonas: Although I had been quite happy at Westat, Penn provided Susan and me with exciting new opportunities. We both took on administrative positions in addition to research. My funding continued to be primarily from NIH, albeit with new disease areas (prematurity and cardiovascular disease) raising difficult but exciting design issues in multicenter activities. I was hesitant about my ability to teach students, having received a “Stay away from Ellenberg” review in the Harvard Crimson Confidential Guide as a Teaching Assistant some five decades ago. I did not feel that my experience in lecturing outside of the classroom would give me the background for working with students effectively. I was wrong; the teaching experience has been the greatest joy of my years at Penn.
Both: Our work in statistics has been enormously satisfying, as we have both been involved in addressing important public health issues. In addition, throughout our careers, we have both enjoyed, and have been grateful for, the opportunities to serve our professional communities: Jonas as treasurer, then president, of the International Biometric Society and then president of the ASA; Susan as president of ENAR and the Society for Clinical Trials, chair of AAAS Section U, and (currently) chair of the NISS Board of Trustees. We feel fortunate to have chosen careers in statistics!