James Cochran

Nancy L. Geller has been the director of the Office of Biostatistics Research at the National Heart, Lung, and Blood Institute of the National Institutes of Health since 1990. She directs a group of 12 statisticians who collaborate in the design, implementation, monitoring, and analysis of multicenter clinical trials in heart, lung, and blood diseases and sleep disorders. She also administers all statistical activities of the National Heart, Lung, and Blood Institute. She has been and continues to be involved in the design and analysis of a number of large cardiovascular and hematopoietic stem cell transplantation trials, including the Women’s Health Initiative clinical trial of hormone replacement therapy, the Action to Control Cardiovascular Risk in Diabetes or ACCORD trial, and several multiple myeloma trials.

Geller has published more than 200 papers in the statistical and medical literature. She is an associate editor of Biometrics and a fellow of both the American Statistical Association and the International Statistics Institute. She was the winner of the 2009 Janet L. Norwood Award for Outstanding Achievement by a Woman in the Statistical Sciences and 2011 president of the American Statistical Association.

All your degrees are in mathematics. How then did you become a biostatistician?

I majored in math at The City College of New York (CCNY), where I took and really liked probability theory. I applied to graduate schools with the intent of doing a PhD in probability theory, but almost all the activity at that time was in math departments. In graduate school, I had been told that if I wanted to study probability theory, I also had to study statistics, and so I did, and my thesis was on the distribution of test statistics of the Kolmogorov-Smirnov type.

After my PhD, my first faculty position was at the University of Rochester’s Department of Statistics, which cemented my love for statistics. I then became an assistant professor in the statistics department at the University of Pennsylvania. That department is in the Wharton School, and business and economics was never my interest. After a few years at Penn, I moved across town to be the biostatistician at the Medical College of Pennsylvania. If they knew how little biostatistics I knew, they probably wouldn’t have hired me! But certainly my PhD trained me to educate myself, so I was able to study biostatistics—specifically survival analysis—on my own. The position was a lonely one and, after one year, I took the opportunity to move to Memorial Sloan-Kettering Cancer Center to have statistical, as well as, medical colleagues. That’s when I became a real biostatistician!

What did you initially find most attractive about the opportunity to join the National Heart, Lung, and Blood Institute, or NHLBI, at the National Institutes of Health?

After 10 years at Sloan-Kettering, I was interested in trying something new. The biostatistics group at the National Heart, Lung, and Blood Institute had a long and illustrious history of leadership in the field. The opportunity to lead it was an honor I found compelling and still do after many years.

What do you find most rewarding in your current position as director of the Office of Biostatistics Research with the NHLBI?

I find participating in medical research quite compelling. At NHLBI, there are always new challenges and opportunities to learn new things, both about statistics and medicine. And it is amazing and thrilling how smart everyone around me is! I also enjoy mentoring my junior colleagues.

Mentoring of statisticians early in their careers is critical to their professional success, especially in a nonacademic setting. How does the NHLBI handle mentoring of its early career statisticians? As director of the Office of Biostatistics Research with the NHLBI, have you established mentoring programs for the statisticians you bring into your office?

I mentor many people, but particularly my junior colleagues. When someone joins the Office of Biostatistics Research, they are mentored by more than one more experienced member of our group. Often I do a project with them, but since different people have different styles and personalities, it is important that new members of the group do a project with others, as well. Mentoring is thus quite informal, and everyone knows they can ask questions of others in the group. We also have a research seminar that meets regularly with presentations both by outsiders and members of our group. This fosters an open learning environment and the sharing of research ideas.

In your presidential address at the 2011 Joint Statistical Meetings in Miami Beach, you stated, “A unique feature of being a statistician is that you may collaborate in any discipline whatsoever, and you may change fields of application as occasions arise.” You then emphasized that this results in statisticians being unique in seeking out collaborators and interdisciplinary opportunities, which in turn requires us to develop skills that will enable our colleagues in other disciplines to understand and appreciate our contributions. Do you feel that practitioners of our discipline are improving with respect to these skills?

I’ve already mentioned that I switched from business to biostatistics and, within biostatistics, from cancer to heart, lung, and blood diseases. That was quite deliberate, but sometimes you initiate collaborations by serendipity. After getting to NHLBI, I remember thinking that before I could talk to the clinicians, I had better learn the new vocabulary. I found myself unable to speak in sentences, never mind paragraphs. The first challenge I set for myself was to say atherosclerosis without stuttering! Then I progressed to sentences, and then to paragraphs. It’s very hard to initiate collaboration if you don’t know the terminology that the potential collaborator uses every day. I just presented a paper at a journal club that concerned assessing accrual in clinical trials. Those authors could have improved their paper if only they had a biostatistical collaborator! While I could initiate collaboration with these authors, it is more likely that I collaborate with colleagues at NHLBI about this topic. As for whether our colleagues are improving with respect to these skills, graduate programs in biostatistics today are teaching soft skills as well as technical skills, and I think that should help statisticians appreciate what it takes to be a successful collaborator.

You worked on cancer research early in your career, and then you moved into heart, lung, and blood research. What motivated this transition?

A statistician can work in any field of application. Although I never considered going to medical school, I always found medicine interesting. My transition from cancer to heart was not motivated by the medical area, but by the opportunity to do something different in medicine and the opportunity to find new statistical problems.

What was the most surprising or interesting thing that happened to you while you were on the ASA Board of Directors?

First, I was pleased to learn that the ASA works very smoothly, often because Ron Wasserstein and other senior staff will always pick up the slack if someone else (like the ASA president) misses something or cannot do it. Those who work for the government have certain restrictions on their participation in financial matters and in lobbying—such as visits to members of Congress—and these were activities others had to undertake. I was pleased and perhaps a bit surprised that the board worked so cooperatively and that we were never short of volunteers, no matter what the task.

The prestige that comes with the office was a bit of a surprise, especially when people who never talked to me before were suddenly so kind! These things aside, the travel as ASA president and meeting statisticians from all over the world was really interesting. My international travel was especially gratifying, with trips to Korea, Israel, Ireland, and Sri Lanka. The hospitality at the professional meetings I attended and by the faculty and students at these places was amazing! I also had the opportunity to meet a number of eminent statisticians, such as Sir David Cox and C.R. Rao.