Data Science x Public Health

How does the United States actually track diseases like COVID-19, measles, or cancer? In this episode, we break down the major U.S. disease surveillance systems—including NDSS, NHANES, BRFSS, SEER, and vital records—and explain how the CDC’s Data Modernization Initiative is trying to rebuild the nation’s public health data infrastructure after the COVID-19 pandemic exposed major weaknesses. If you...

Clinical trials traditionally rely only on data from newly enrolled patients. But Bayesian borrowing allows researchers to incorporate external data from past studies to strengthen new trials. In January 2026, the FDA released draft guidance expanding how sponsors can use external controls and Bayesian methods in clinical trials. This episode explains how Bayesian borrowing works, why it can make...

Artificial intelligence is rapidly entering healthcare, but algorithms trained on biased data can reproduce and amplify health disparities. This episode explores how bias enters the machine learning pipeline, real cases of algorithmic harm, and how fairness-aware machine learning could help build more equitable health systems. 👉 Enjoyed the episode? Follow the show to get new episodes automatical...

Climate change is altering where and how diseases spread across the world. From expanding mosquito ranges to wildfire smoke exposure and extreme heat events, epidemiologists are studying how a warming planet is reshaping public health. This episode explains the emerging field of climate change epidemiology and the methods used to track disease in a changing environment. 👉 Enjoyed the episode? Fol...

Clinical trials have traditionally relied on frequentist statistics, but Bayesian methods are gaining momentum in modern trial design. With new FDA guidance released in 2026, Bayesian approaches are becoming more prominent in evaluating drugs and medical devices. This episode explains the differences between frequentist and Bayesian statistics, how Bayesian adaptive trials work, and why this shift...

Health data is often locked inside hospitals and health systems, making it difficult to build powerful AI models. Federated learning solves this problem by training models across multiple institutions without moving patient data. This episode explains how federated learning works, why it matters for public health, and the challenges of training AI across distributed healthcare systems. 👉 Enjoyed...

Disease surveillance systems are the backbone of epidemiology. In this episode, you’ll learn the major types of surveillance systems used in public health, including passive, active, and sentinel surveillance. We’ll explore real examples like NDSS, NHANES, BRFSS, SEER, and hospital discharge data to understand how epidemiologists track diseases and monitor population health. 👉 Enjoyed the episode...

Missing data is common in health research, but the way it’s handled can dramatically change study results. This episode explains the three missingness mechanisms (MCAR, MAR, MNAR), how multiple imputation works, and why sensitivity analysis is essential for producing reliable and transparent biostatistical conclusions. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If y...

Synthetic health data promises privacy and scalability, but it also raises new risks. This episode explains how synthetic datasets are generated, when they actually help public health research, and where they can fail by leaking patterns, amplifying bias, or misleading models. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, consider leav...

Most emerging infectious diseases originate in animals, yet human, animal, and environmental health have historically been studied separately. This episode explains One Health epidemiology and how integrating these systems improves outbreak detection, antimicrobial resistance monitoring, and global disease surveillance. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If...

Medical research often produces conflicting results. Meta-analysis is the statistical method that combines multiple studies to produce a clearer estimate of evidence. This episode explains how meta-analysis works, how to interpret forest plots, and why heterogeneity and publication bias can shape the conclusions. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you fou...

Supervised learning answers the questions we already know to ask. Unsupervised learning reveals the ones we don’t. This episode explains how unsupervised learning uncovers hidden patient phenotypes, community risk profiles, and early outbreak signals in public health data—and why these methods are still underused. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you fo...

Genomic surveillance helps public health detect emerging variants, map transmission, and adapt response strategies. This episode explains how pathogen sequencing works, what it can and cannot tell us, and why it plays a critical role in modern outbreak response. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, consider leaving a rating or...

Health outcomes are not deterministic. This episode explains how stochastic processes model randomness in disease progression, epidemics, genetics, and clinical trials, and why these tools are essential in modern biostatistics. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, consider leaving a rating or review—it helps support the podcas...

What if public health could detect outbreaks before hospitals fill up? Wastewater-based epidemiology turns sewage into an early-warning system for population health. This episode explains how machine learning helps clean noisy wastewater data, detect trends, and support real public health decisions. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content...

Epidemiology is full of patterns, but public health decisions require causes. This episode explains how causal inference helps move from association to intervention, using tools like DAGs and target trial emulation to design clearer, more trustworthy answers from observational data. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, conside...

Big data doesn’t stop at basic computing. This episode continues the discussion by exploring software development, databases, visualization tools, and large-scale computing systems used in biostatistics. This is Part Two of a series on how big data is computed and scaled. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, consider leaving a...

Public health depends on surveillance to detect outbreaks, monitor trends, and protect populations. This episode breaks down the main types of epidemiologic surveillance and explains how they work together to track disease in real time. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, consider leaving a rating or review—it helps support t...

Big data drives modern biostatistics, but how is it actually computed? This episode breaks down the core ideas behind computing with big data, including data processing, optimization, programming tools, and version control. This is Part One of a deeper series on how large-scale data is handled in biostatistics. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found...

Supervised learning is one of the most powerful tools in modern public health analytics. In this episode, we break down what supervised learning is, how it works, and how it’s used to predict outbreaks, assess risk, and support medical decision-making. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, consider leaving a rating or review—it...

Chatbots are becoming powerful tools in public health—from emergency communication to patient education and triage. This episode introduces how public health chatbots are built, where they are used, and why they are becoming an important part of modern public health systems. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, consider leavin...

This episode is Part Two of our deep dive into the theory and methods of sample design. We move beyond the basics and explore advanced sampling techniques used in real-world public health, epidemiology, and survey research. Learn how multistage sampling, variance estimation, non-response adjustments, and imputation shape the quality of population data—and why these concepts matter more than you th...

Genetic epidemiology and statistical genetics help explain how genes and environment work together to shape health outcomes. This episode introduces the foundations of both fields, their methods, and why they are essential to modern public health and epidemiologic research. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, consider leaving...

Pharmacoepidemiology plays a critical role in understanding how medications are used and how safe and effective they are in real-world populations. This episode introduces the field, explains what pharmacoepidemiologists do, and shows why this discipline is essential to public health and regulatory decision-making. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you f...

Sampling design determines how well data represents real populations. In Part One of this series, we introduce the theory and methods of sample design, explain why sampling matters in biostatistics, and explore how major national health surveys collect reliable data. 👉 Enjoyed the episode? Follow the show to get new episodes automatically. If you found the content helpful, consider leaving a rati...