Student Spotlight

FIU Stempel College biostatistics students are using data to make an impact on population health

In public health, translating data into clearcut messages can make an impact on the health of millions.

Behind the scenes of major public health decisions are data analysts, like biostatisticians, who develop new methods and model the data to identify trends that help improve approaches to health promotion, disease prevention and clinical care.

“Our job is to use statistical, mathematical and computing methods to analyze the data appropriately and extract value from it to drive population science, medical or clinical decisions. When the existing methods are not adequate, we make modifications or develop new ones to overcome the limitations,” said Zoran Bursac, professor and chair of the Department of Biostatistics , which is ranked No. 43 among public universities by U.S. News & World Report.

At the Robert Stempel College of Public Health & Social Work , a diverse group of faculty members is training the next generation of biostatisticians. Through active research programs, students learn the latest in biostatistical and computing methods and work on projects that impact communities near and far. They also get the opportunity to present their research at national conferences.

This fall, three doctoral students, along with Associate Professor Boubakari Ibrahimou and Assistant Professor Gabriel Odom , from the Department of Biostatistics, presented their research at the Joint Statistical Meetings conference, one of the largest statistical events hosted by the American Statistical Association. Ibrahimou and PhD student Ning Sun on how air pollution may cause preeclampsia - a serious blood pressure condition that develops during pregnancy - in pregnant women regardless of other preexisting conditions. Produced by: Daniel Chavez, FIU DC

Explore their impactful work below.

Ning Sun

Hometown: China

Tell us about the research you presented:

I presented my project on the spatiotemporal pattern evolution of preeclampsia in Florida between 2006 and 2015. We found a striking surge in preeclampsia risk in Central North Florida around 2012, which cannot be explained by well-known risk factors. Thus, it would be promising to investigate it further to understand the hidden reasons.

I hope this study can help us identify unknown risk factors for preeclampsia and other maternal diseases that can, in turn, guide public health resource allocation.

What do you hope to accomplish in your career in Biostatistics?For my career, I want to be a principal biostatistician in the pharmaceutical industry to help develop new medicines.

Sergio Perez-Melo

Hometown Originally from Havana, Cuba. However, I’ve lived in Miami since 2007.
Tell us about the research you presented.

When trying to predict an outcome, we use input variables (e.g., in trying to predict college GPAs (output), we may use inputs such as demographic variables, high school GPA, etc.) Sometimes the input variables are correlated, and that creates an issue called multicollinearity. When multicollinearity is present, one of the ways we deal with it is through a method called shrinkage estimation. My research is focused on developing efficient tests of hypotheses for shrinkage estimation. This is an ongoing field of research in statistical methodology. Some preliminary results were published in 2020.

I hope my research findings will expand the available statistical toolbox for applied practitioners that deal with correlated data in regression models.

What do you hope to accomplish in your career in Biostatistics?

​I would like to continue teaching as I do now, but also be able to teach at the graduate level, as well as try my hand at statistical consulting in the future.

Kazi Tanvir Hasan

HometownDhaka, Bangladesh

Tell us about the research you presented.

Cardiovascular disease (CVD) is a major global health concern. It has impacted 31% of the worldwide population by contributing to the deaths of 17.7 million individuals. Additionally, it is estimated that by 2030, CVDs will contribute to a high mortality and morbidity rate, impacting 23.6 million people. It is a condition with a lot of independent risk factors. Heavy metal is one of them. Humans absorb heavy metals through multiple routes. For example, a study found a link between higher amounts of barium in drinking water and increased cardiovascular mortality. Blood and Urine metals are linked to an increased risk of CVDs, but the effect of these heavy metal mixtures on heart attack remains unknown. In this study, we used Bayesian kernel machine regression (BKMR) as a novel way to study mixtures to identify blood and urine heavy metals as well as heavy metal mixtures that may be associated with heart attacks.

The 'BKMR' model presented here can be used to represent new types of exposure in future studies, allowing us to gain a better knowledge of how the environment as a whole influences health and disease in a particular human population.

What do you hope to accomplish in your career in Biostatistics?

My plan after graduation is to dedicate myself to research and teaching because I feel that I could help make the world a better place, and that is my main goal.