
Sign up to save your podcasts
Or


Epidurals are widely used and widely trusted for pain relief during labor. So when a 2020 study reported that they might be linked to autism, it raised a troubling question: could a routine medical decision have long-term consequences? We follow that claim from headline to evidence—and watch what happens when other scientists take a closer look. We dig into the original study, a wave of replication studies from around the world, and a meta-analysis that tries to make sense of it all. Along the way, we unpack hazard ratios, Cox regression, inverse probability weighting, and sibling analyses—and why even sophisticated statistical adjustment can’t eliminate confounding. Plus: why bigger datasets don’t solve everything, what happens when results shrink after adjustment, and how a controversial study turned into a case study in science working as it should. Bonus: our first guest journalist interview!
Statistical topics
Methodological morals
References
Kristin and Regina’s online courses:
Demystifying Data: A Modern Approach to Statistical Understanding
Clinical Trials: Design, Strategy, and Analysis
Medical Statistics Certificate Program
Writing in the Sciences
Epidemiology and Clinical Research Graduate Certificate Program
Programs that we teach in:
Epidemiology and Clinical Research Graduate Certificate Program
Find us on:
Kristin - LinkedIn & Twitter/X
Regina - LinkedIn & ReginaNuzzo.com
By Regina Nuzzo and Kristin Sainani4.9
3232 ratings
Epidurals are widely used and widely trusted for pain relief during labor. So when a 2020 study reported that they might be linked to autism, it raised a troubling question: could a routine medical decision have long-term consequences? We follow that claim from headline to evidence—and watch what happens when other scientists take a closer look. We dig into the original study, a wave of replication studies from around the world, and a meta-analysis that tries to make sense of it all. Along the way, we unpack hazard ratios, Cox regression, inverse probability weighting, and sibling analyses—and why even sophisticated statistical adjustment can’t eliminate confounding. Plus: why bigger datasets don’t solve everything, what happens when results shrink after adjustment, and how a controversial study turned into a case study in science working as it should. Bonus: our first guest journalist interview!
Statistical topics
Methodological morals
References
Kristin and Regina’s online courses:
Demystifying Data: A Modern Approach to Statistical Understanding
Clinical Trials: Design, Strategy, and Analysis
Medical Statistics Certificate Program
Writing in the Sciences
Epidemiology and Clinical Research Graduate Certificate Program
Programs that we teach in:
Epidemiology and Clinical Research Graduate Certificate Program
Find us on:
Kristin - LinkedIn & Twitter/X
Regina - LinkedIn & ReginaNuzzo.com

882 Listeners

411 Listeners

697 Listeners

997 Listeners

14,435 Listeners

5,121 Listeners

3,594 Listeners

656 Listeners

3,333 Listeners

192 Listeners

2,049 Listeners

1,190 Listeners

819 Listeners

553 Listeners

770 Listeners