Jessica Young, PhD is a biostatistician in the Department of Population Medicine at Harvard Medical School who joins the show to discuss the ins and outs of her interesting and important work.

Tune in to learn the following:

• How confounding factors in a study can influence the findings of the study, and how/why the gold standard of randomized trials can address this
• What is meant by the "fundamental challenge of causal inference" and how this explains why assumptions are always necessary in order to claim that a statistical analysis is unbiased
• Why large subject numbers or data points can't overwhelm biases; why bias is a function of the thing being studied

Dr. Young's job is two-fold: she works on both the applications of statistical methods for public health and clinical medicine, and also on the development of methods in these areas. She focuses on causal inference, which is the formal process of understanding how to estimate causal effect from data collected in real-world studies.

Through examples including a longitudinal study on nurses starting in the 1970s to present day studies revolving around the coronavirus pandemic, Dr. Young discusses confounding factors in studies and the effect they have on interpretations of findings, the importance of randomization, the presence of bias regardless of how statistically significant a finding is, meta-analyses, where she sees the field of biostatistics heading in the near future, and more.  

To learn about the basics of causal inference, Dr. Young recommends reading The Book of Why: The New Science of Cause and Effect. Visit https://www.populationmedicine.org/JYoung to learn more about her work and publications.