What is Biostatistics?
Biostatistics is a field at the intersection of statistics, public health, biology, and computing. The aim of a biostatistician is to use statistical techniques to solve biomedical problems or problems which impact public health. For someone with a backgroud in both Mathematics and Health and Wellness, it is very exciting to have found an academic trajectory which will allow me to work in both of these fields simultaneously. As part of my involvement in the McNair Scholar's program, I am preparing to submit applications for PhD programs in Biostatistics this fall. I have found that Biostatistics is still niche enough that most folks not involved in the field aren't familiar with what Biostatistics is and what it encompasses. I wanted to share some areas which Biostatisticians study so it will become clearer why I am so excited about it. All of the information to follow is what I learned this summer while participating in the Iowa Summer Institute in Biostatistics at the University of Iowa.
What do Biostatisticians Study?
Clinical Trials and Experimental Design - Clinical trial design and analysis is a major area in which biostatisticians have classically worked in. Clinical trials are meant to test the safety and efficacy of new drugs, medical devices, or other health interventions. Biostatisticians work on developing methodologies to conduct clinical trials efficiently and ethically. Key questions in this area include:
How can we design trials that balance the need for rigorous scientific evidence with ethical considerations, such as minimizing patient risk?
How can adaptive trial designs be implemented to allow modifications based on interim results without compromising the validity of the study?
Survival Analysis - Considered another "classical" area of biostatistics, survival analysis handles time-to-event data. For instance, this encompasses data such as time elapsed until a patient experiences a particular health event. Biostatisticians develop methods to predict this time and identify risk factors which may impact it. Key questions include:
How can we model complex survival data that include competing risks factors?
How can we incorporate time-dependent influencing factors, which change over the study period, into survival models?
Epidemiology and Public Health - In epidemiology, biostatistics is essential for understanding the distribution and determinants of diseases in populations. This area involves designing studies, analyzing public health data, and developing models to predict disease spread and evaluate interventions. Key questions include:
How can we improve the accuracy of models predicting the spread of infectious diseases?
What statistical methods are best for assessing the effectiveness of public health interventions?
How can we address biases in observational studies, where randomized controlled trials are not feasible?
Statistical Genetics - Statistical genetics focuses on the development and application of statistical methods to understand the genetic basis of traits and diseases. The main goal is identifying genetic variants that contribute to complex traits and to understand how these variants interact with environmental factors. Key questions include:
How can we improve the detection of rare genetic variants that may have large effects on disease risk?
How can we model gene-environment interactions to better understand the multifactorial nature of diseases?
Network Analysis - Network analysis is increasingly important in understanding the complex relationships between biological entities, such as genes or individuals within a population. Modeling these interactions as networks turns out to be an effective model for identifying major players and understanding the structure of biological systems. Key questions include:
How can we infer the structure of biological networks from high-dimensional data?
What methods can be used to identify critical nodes or pathways in these networks that are essential for disease progression or treatment response?
Example of a network of drug-drug interactions. Source: https://towardsdatascience.com/network-analysis-and-visualization-of-drug-drug-interactions-1e0b41d0d3df
6. Big Data and Machine Learning - The rise of big data has opened new avenues for research in biostatistics, particularly in the development of machine learning algorithms to analyze complex, high-dimensional datasets. This area explores how advanced computational techniques can be applied to biomedical data. Key questions include:
How can machine learning be used to predict patient outcomes based on electronic health records or imaging data?
What are the best practices for ensuring that machine learning models are interpretable and generalizable in clinical settings?
How can we address the challenges of data privacy and security in the analysis of sensitive health information?
As you might garner from this list of topics, biostatistics is a field that is only expanding with the advancement of technology. Biostatisticians are at the forefront of data, eithics, and health outcomes. Their work may be theoretical in nature through the development of methodologies, or may be applied in the capacity that they collaborate with other researchers in designing and interpreting biomedical experiments. With the increasing number of data publically available, biostatisticians are empowered to ask and answer biomedical questions directly.
Comments