BERD Core Brings Essential Biostatistics to Researchers’ Studies
Biostatistics involves the development and application of statistical methods to design, analyze, and interpret data in medical and public health studies. While crucial for biomedical research, the discipline can be intimidating to many researchers.
Alleviating this concern and making biostatistics resources more accessible to the Einstein community are goals of the Biostatistics, Epidemiology, and Research Design (BERD) Core, which is part of the Harold and Muriel Block Institute for Clinical and Translational Research (ICTR) at Einstein.
“Our priority is to provide high-quality statistical expertise and support to promote rigor and reproducibility of research,” said Mimi Kim, Sc.D., professor of epidemiology & population health, head of the division of biostatistics, director of the ICTR BERD Core and of the Quantitative Sciences in Biomedical Research Center, and associate director of the ICTR.
A Helpful, Productive Flock
The BERD Core comprises 10 statisticians who offer their expertise through more than 4,500 consultations per year on average, while also contributing to hundreds of publications and grants, and running hands-on workshops and biostatistics courses.
“In addition to our numerous collaborations and teaching activities across the institution, the biostatistics faculty have their own independent research programs to develop innovative statistical approaches to address new methodological challenges that arise in our increasingly data-rich biomedical research environment; we’re a very robust, active, and productive core,” said Dr. Kim, who also holds the Harold and Muriel Block Chair in Epidemiology & Population Health.
Melissa Fazzari, Ph.D., associate professor of epidemiology & population health, emphasized the importance of biostatistics at all stages of research. “Sound study design minimizes bias and maximizes precision, while appropriate analytical methods ensure that the results are valid so they can be interpreted and reported properly.”
Dr. Fazzari is spearheading development of an online biostatistics toolbox known as the “BERD House,” which is available to investigators throughout the Einstein Montefiore community.
When asked when in one’s project is best to seek guidance from the BERD Core, Dr. Kim noted, “Coming in early for support is paramount. There is no analysis that can make up for bad design and insufficient power.”
The Wind Beneath Research Projects’ Wings
Core staff members work closely with many researchers. “No project is too small for a biostatistician, and many often lead to productive and fulfilling collaborations that can last an entire career,” said Dr. Kim.
Ari Moskowitz, M.D., associate professor of medicine, has two ongoing projects with the core. “Working with them has been a true pleasure. They’re all so helpful and responsive and are able to apply complicated statistical concepts to real-world patient-oriented studies,” he said.
One project uses machine learning to try to predict which patients are at risk for infection after central line placement. “Dr. Fazzari was involved early on and has been tremendously helpful in adapting and troubleshooting our analysis plan,” he noted.
The other project aims to assess differences between two methods used in the management of patients’ airways when in the hospital.
“The trial has a complicated study design, and we’re redefining an outcome for a population, so we need the statistics to back it up,” said Dr. Moskowitz. “Mimi has been integral as a collaborator since the planning stage.”
When asked what advice he’d offer to other investigators about reaching out to the BERD Core, he said, “The most important thing is to come in already knowing what your goals are. Biostatisticians can’t read minds, but they can help you if you know what questions you’re trying to ask.”
Helping Researchers Take Flight
The core offers a weekly virtual walk-in consulting service that is open to the entire Einstein Montefiore research community.
“We help upwards of 100 people a year with all sorts of questions,” said Dr. Kim. “That includes graduate student and postdocs, as well as junior and senior investigators.”
Michael Rogers, a seventh-year M.D./Ph.D. student in the genetics department, used the walk-in service every year of his doctoral work. “It was reassuring to have done my due diligence before starting new experiments, instead of just trying to guess or rely on historical precedent,” he said.
He added, “Even when the help I needed wasn’t straightforward, I always walked away less confused than when I came in. The BERD Core statisticians are very patient, so we were able to explore the statistics as far as we needed to.”
A BERD in Hand Is Worth the Data Provided
The statistical help that BERD Core statisticians provide is becoming increasingly valuable. “Biomedical research, in general, is becoming more data-intensive, so we need good statistical practices to extract knowledge from increasingly complex data,” said Dr. Fazzari. “And, in a practical sense, almost all journals and grant funding agencies have statistical reviewers now.”
Dr. Moskowitz noted, “It’s important not to overestimate our abilities as researchers when it comes to biostatistics. When it’s not your expertise, it can be easy to make a small mistake in the planning or analysis that can have serious implications.”
“To paraphrase one of my favorite statistical quotes by R.A. Fisher,” said Dr. Kim, “‘To consult a statistician after an experiment is like asking them to conduct a post-mortem: they can perhaps say what a bad experiment died of.’”
Using the expertise of those in the BERD Core, BERD House, and walk-in service offers some helpful measures for avoiding such a prognosis with one’s research. To learn more, visit the BERD Core website!
Posted on: Monday, January 08, 2024