Some may never have heard about the occupation of biostatistician or the field of biostatistics. When I started graduate school in 1973, there were only a handful of biostatistics programs around the country. Most programs in statistics grew out of mathematics and evolved around theorem proving. That has definitely changed, and the related field of data science is now considered one of the hottest areas to pursue as a career.
I think of myself as a problem solver. I like to work with scientists and use my skill in statistics to solve problems. John Tukey, a famous statistician at Princeton, once told a colleague, “The best thing about being a statistician is that you get to play in everyone’s backyard.” I think Tukey captured my passion about what I do.
Let me tell you a little about myself.
East New York, New York
My parents had survived the Holocaust. My mom spent 24 months in Auschwitz and was the only one in her immediate family to survive. Both of my parents lost many relatives in Europe. My dad didn’t discover that his brother had survived until years after the war ended.
Because both of them had relatives in the United States, two years before I was born, my parents decided to move here with my older brother. They moved to a three-story walk-up in a Brooklyn neighborhood called East New York just before I was born because the apartment had heat.
Sea Gate, New York
When I was six, we moved to an area of Coney Island called Sea Gate. My younger brother was born soon after we moved. I went to public school for only four hours a day; the school was so crowded that some grades had classes from 8 AM to noon and others from noon to 4 PM. I think each class had about 35 students.
Because the Holocaust had disrupted theirs, my parents always valued education. Although my mom never reached what we would call middle school, she could read and write in several languages. My dad was a plumber who complained that English was much harder than the multiple languages he knew from Europe. Still, he was appreciated for his ability to innovate when needed. They made me understand that people could be very smart without having advanced degrees. My parents had to work hard just to get by, but after the horrors of the Holocaust, they were so grateful to be in the United States.
Because I was good at standardized tests, I jumped from seventh to ninth grade at Mark Twain Junior High School. However, the neighborhood was so rough that staying after school for any extracurricular activity was dangerous. When I got to Lafayette High School I was fortunate to have a few amazing teachers. I completed high school equally in love with mathematics and biology, and I wanted to find a way to use them both. It wasn’t until college that I realized that girls were not expected to like math!
Stony Brook, New York
I was admitted to the State University of New York at Stony Brook. Even though I had a small scholarship, it was a reach for my parents to send me. I was a math major until an advisor asked if I planned to teach high school. When I said “No,” he replied, without knowing anything about me, “Then what could you do?”
As a result, I decided to investigate the requirements for a biology degree. Because many biology majors were not terribly quantitative, my biology advisor encouraged me to pursue the interface of math and biology. I had friends who were pre-med students studying math, but I ended up being the only math–bio double major I knew in college. The summer after my junior year, I received a National Science Foundation undergrad research award and spent the summer doing simulations in population genetics. Learning how to program was a great experience. However, I felt that population genetics wasn’t applied enough for me. With no idea what biostatistics really entailed, I considered studying it in graduate school.
Seattle, Washington
After a year of working as a computer programmer for New York Life Insurance in Manhattan, I started graduate school in the biomathematics program at the University of Washington and was offered a research assistantship in the Department of Biostatistics. Having programming skills in the 1970s helped me get the position.
The PhD program offered training in the interface of mathematics/statistics and biology, but my assistantship was in biostatistics. On my first day in the biostatistics department, the male faculty were introduced to me as “Doctor so and so,” while the female faculty were introduced by their first names only. I probably should have been upset by what today would surely be considered sexist, but at that time, I had never met a woman with a PhD. I was pleased to see women faculty, which I took to mean that I stood a chance of finishing.
I loved my grad school program, made lots of friends, worked really hard, and completed my PhD in five years. I never felt that the statement “women can’t do math” applied to me, nor did I ever feel as if I should be limited in my career goals because I was female. I was lucky to have a strong group of women graduate students and several women faculty at the University of Washington.
I spent my fifth and final year of graduate school as an instructor in the College of Fisheries’ Center for Quantitative Science. The person who hired me noted that I had more biology and ecology courses than many of the quantitative ecology students in the center. I taught biostatistics to forestry, fisheries, and biology students. My boss was a prominent statistician who happened to be the dean of fisheries.
My PhD thesis was on the use of high dimensional classifiers, a topic that became much more popular with advanced computation. Such methods are now used to analyze medical images to spot cancers. I was really interested in working in basic biology rather than clinical trials, and my family was back east, so I left Seattle for the University of Maryland in the Department of Dairy Science, where I worked as a biostatistician within the College of Agriculture and Life Sciences.
College Park, Maryland
When I joined the faculty at UM, I was pretty naïve. I didn’t know how to negotiate for resources or salary. I was the first female in my department and discovered several women in the college of agriculture who were also “the first.” Some of us were treated better than others.
Most appointments in the college were for 12 months, with a mix of research, teaching, and industry outreach. Mine was a nine-month teaching appointment with basically no support for a small research program. Try teaching statistics or doing research without access to computing! Although I had several friends among senior faculty who gave great advice, the chair had total control over funding.
After two years, I attended a meeting where I met a few University of Washington grads. One was Rita Colwell, who knew my undergrad research advisor at Stony Brook and my boss when I was an instructor at the University of Washington. We clicked. Although Rita was high-powered and successful, she helped many of the women faculty. Together, we wrote grants and papers and worked to bring computing support to campus, including cutting-edge statistical software. From Rita, I learned that mentorship is critical. I swore then that I would mentor those who came after me.
I met my husband, a statistician at the Bureau of Labor Statistics, three days before I was to hear about tenure. When I got tenure and was promoted to associate professor, I told him he was my lucky charm. Some have asked what it is like to be married to someone in your own field. In my case, with the exception of a trip to the main American Statistical Meetings, which we combined with a family vacation, we never saw each other in a work-related setting. Nevertheless, my husband has always been supportive of my career. (BTW, my daughters described that meeting as 5000+ dorks in one city, although we did get to see a lot of the United States and Canada.)
At the University of Maryland, I worked across many science departments, helping graduate students and faculty solve problems. Sometimes it meant spending long hours in the library learning new statistical methods. Some of the papers I coauthored were written with graduate students but were not part of a thesis. I required data projects in my graduate classes, and a few managed to publish papers that included their contributions. I believe that there is no point in teaching applied statistics unless the students can actually use the methods to solve problems.
Students working on their theses often came by with questions. Some days I would see six PhD students from five different departments. While it could be taxing, given that the terminology often varied by department, I really enjoyed working with them. A few of my biology graduate students switched to statistics; one of my undergraduates went to Hopkins for a PhD and is now a fellow of the American Statistical Association.
I served on a number of committees at the University of Maryland, including those related to computer support. With few females in agriculture at the time, women faculty tended to be tasked with more unrewarded service roles than did the men. My colleague and I wanted to use computers to teach, but there was no way standard departmental resources would have been there to pay for them. Luckily, I was able to use the knowledge from my committee work and pleaded my case to our dean to gain two computer classrooms, one for my department and one for biology.
Because I was passionate about working on infectious diseases and public health, I loved collaborating with Rita Colwell and other colleagues from her department as well as from the veterinary medicine unit on campus. I traveled to meetings in Bangladesh, Ecuador, Jamaica, and England to give talks or interact with various research groups. I was equally comfortable working in nutrition, ecology, and evolution and in animal and dairy science. I think I used material from every biology course I ever took in college and graduate school and was not afraid to read journal articles in these fields.
In 1989–1990, I took a year-long sabbatical leave at the National Cancer Institute — a fantastic opportunity. The 50+ statisticians there formed a very prestigious group, on par with the best departments in the country. With Richard Simon, I published methodological papers that included some theoretical work in statistics. Rich took on one statistician a year for a number of years, and we found topics of mutual interest to work on. I returned to the National Cancer Institute for a few summers and completed several papers. I also made many friends whose interests were similar to mine. I was promoted to full professor in 1995.
In 2001–2003, I chaired the promotion and tenure committee for my department at the University of Maryland. Most faculty thought the committee was important but weren’t crazy about being in charge. Within two years, we went from two tenured women faculty to five. When I told my department chair I was pleased with my accomplishment, he said he hadn’t noticed.
In 2003, our department underwent an external review with no statisticians represented on the review committee. The decision was made to hire biologists to teach statistics. I would be the only statistician in the biometrics group in the Department of Animal Sciences, because out of the three of us, one retired and one left for an academic position in Florida.
I decided I needed a change as well. When a friend suggested I consider working for the FDA, I decided to apply. It was local, and I wouldn’t have to move. (My daughters were in seventh and twelfth grade by then, and I was concerned about switching jobs because academic life comes with a lot of flexibility.)
Montgomery County, Maryland
I was hired by the FDA in 2004 and spent the past 15 years of my career working there. I began at the Center for Devices in its Division of Biostatistics. I asked to join the Diagnostics Statistics group and found myself using skills from graduate school and all the consulting I’d done at the University of Maryland.
I became a team leader in that group soon after I arrived, often asked by statisticians to explain the science and by scientists to explain the statistics. The Office of In Vitro Diagnostics asked me to speak at relevant meetings on statistical issues. I had found my niche.
After two years at the FDA, I had collaborated on projects in every review division in the Center for Devices, including the first microarray as a medical device there. I had worked with similar technologies at UM, but now had to think about what it took to approve its use as a medical device.
I believe my team of six or seven PhD-level statisticians did more reviews and published more papers than any other team in the division. I wish I could take credit, but I worked with some amazingly talented people. I helped edit the papers and presentations, including those presented at FDA advisory committees. My own publication rate definitely slowed down as I spent a lot of time reviewing submissions to the FDA, but I thought team leaders should help with professional development. I guess it wasn’t held against me, because in 2010, I became a fellow of the American Statistical Association.
That year I moved to the Center for Biologics Evaluation and Research (CBER), the smallest of the three medical centers within the FDA. I was originally appointed deputy division director for the Division of Biostatistics and became division director a year later. I signed off on statistical reviews for vaccines and many products to treat rare diseases. While the Center had some diagnostic device submissions to consider, I had much more experience than most and helped where I could. The division grew during my tenure; we made some outstanding hires.
As a division director, I led several initiatives, including one associated with increasing the use of innovative clinical trial design and analysis strategies at our center. I promoted efforts to advance training on these and other topics such as correlates of protection, which is important in vaccine development. Additionally, I led an initiative in the post-market safety of vaccines.
I especially enjoyed helping develop the FDA and international guidances (documents published by drug regulators). One of the projects I am most proud to have been part of is the first major statistical amendment to critical international guidance in 20 years: the topic of estimands. The term estimands was new to many in clinical trials. It meant stating the objectives for the study, including what to do when subjects failed to follow the treatment to which they were assigned.
When companies came to the FDA to negotiate how to design a clinical trial to support a drug’s approval, they were often vague about issues such as how rescue medication use would be handled. Curiously, industry thought this was a good idea, because after spending $100 million on a trial, they wanted to be sure the FDA would agree with the interpretations of the trial’s results. I am still asked to give talks and write commentary on this topic.
In terms of staff development, I told senior leadership that it was important for them to learn to speak at professional meetings, so I made sure to give them those opportunities.
In 2016, I retired from full-time employment but returned in May 2017 for a three-year part-time arrangement at the Center for Drug Evaluation and Research (CDER), the largest center at FDA. I was brought on to help with guidances, something I had a lot of experience with in CBER. I assisted junior staff and short-handed senior management. I even wrote a book chapter on estimands with a junior statistician. While I liked the topics and the statisticians with whom I worked, I found this center to have the most bureaucratic environment I’d experienced. Thus, when my three years were up in May 2020, I finally left the FDA for good.
Post-FDA, Maryland
Although I am now retired, I try to stay busy. I regularly review articles for scientific journals. I work with a few research groups led by people active in the American Statistical Association (ASA) and the Drug Information Association and am still involved in publishing papers with a statistical emphasis. Several of those with whom I collaborate are friends I knew from my years at the FDA; a few are new.
A friend gave my name to someone at the Gates Foundation, and I now work on a panel that evaluates projects for potential funding by the organization. This is incredibly rewarding, and both my University of Maryland and FDA experiences are helpful. At the University of Maryland, I collaborated on nutrition and in infectious diseases, and, of course, I was interested in international public health. Novel vaccines funded by the Gates Foundation may wind up at the FDA for review. So I feel very much at home on this panel. Contributing to public health is exactly what I hoped to do in retirement. It should be obvious that I like to keep learning and I also like mentoring.
Staying Active in the Profession
The entire time I was at the University of Maryland and the FDA, I was active in the ASA. Since I was in a small unit of two to three faculty at the University of Maryland, I thought it was especially important to have other statisticians to talk with. If I got stuck on a problem that came up in consulting, I had friends to contact for ideas. At the FDA, being part of ASA was not required, but all of the senior statistical leadership was active, and connections with universities were great when I had to hire new statisticians.
When I chaired the continuing education program for ASA, enrollments went up, and I was congratulated by senior leadership in the organization. I was on the program committee three times for a meeting that had 5000 to 7000 attendees. The fact that I was eclectic and had varied interests was incredibly helpful in setting up a continuing education program and serving on a program committee for activities attended by master’s and PhD level statisticians.
People need to understand that in academia, as opposed to industry or organizations such as the FDA, being eclectic isn’t always a good thing. Being a recognized expert in a narrower niche in your field is more likely to be rewarded there.
Although I do not have the exact numbers, I am willing to bet the percentage of members in ASA who are women is higher than in societies for theoretical mathematicians. I think women like the problem-solving and interdisciplinary aspects of statistics. I was lucky to find my niche.
Balancing Work and Family
Fortunately, my two daughters were born at the end of April and early May, respectively. I was still on a nine-month contract and tenured, so I took the summers off. In the 1970s, women were afraid to have children before tenure. Technically, we could claim six weeks of sick leave but it was impossible to get daycare for a six-week-old.
My chair wasn’t happy when my first daughter was born, but didn’t complain with the second. I think at that point, he realized I was going to remain productive. As to daycare, we had good options once we got to three months, but we had to book way in advance.
As my daughters grew up, I made a point of not telling them that they had to pursue a particular career path. My older daughter has a passion for politics and went to Georgetown University so she could stay in the Washington, DC, area. She is now communications director for Senator Warner from Virginia. Previously, she had internships in the House, Senate, and White House. I have learned that if you are going to major in political science, internships really help you develop that interest into a career.
My younger daughter has a BS from the University of Maryland in biochemistry and was a Banneker Key scholar. She worked in a few research labs while at the University of Maryland, which really helped her when applying for graduate school. She has a PhD in immunology from Johns Hopkins but decided “publish or perish” was not for her. She is doing well as a medical writer and uses her science background when helping companies communicate their achievements. I am very proud of both of them.
Bottom Line: I think each decade has become better than the previous one for women in science. But knowing what counts and how to navigate a career path is important, and having good mentorship matters.
Excerpted from Lessons Learned: Stories from Women Leaders in STEM, edited by Deborah M. Shlian.