Literature

Curbside Consults Podcast


Published November 26, 2019

Missing data are common in randomized, controlled trials. Why does this happen? How do we interpret these missing data? In this episode of Curbside Consults, Dr. David Harrington, statistical editor at the NEJM, joins us for a discussion on missing data and how to interpret missing data in studies.

00:00 – Introduction
00:54 – Summary of the ARCH trial – Romosozumab vs. alendronate
1:56 – What are missing data?
2:25 – Why do we have missing data?
3:40 – Problems associated with missing data
7:25 – How do we interpret missing data?
13:59 – What is an appropriate amount of missing data?
14:59 – What is a sensitivity analysis?
15:50 – Summary

Resources and articles discussed in this episode:
1. Saag, K et al. Romosozumab or Alendronate for Fracture Prevention in Women with Osteoporosis. NEJM 2017.
2. Ware et al. Missing Data. NEJM 2012
3. Little et al. The Prevention and Treatment of Missing Data. NEJM 2012.

The Curbside Consults series complements the foundational information in Rotation Prep by taking a deep dive into key clinical topics with expert clinicians and educators. These podcasts explore and critique the evidence behind clinical practice and break down statistical concepts for the busy clinical trainee.

res360
David Harrington is Emeritus Professor of Biostatistics and Statistics at Harvard T.H. Chan School of Public Health
res360
Dr. Amanda Fernandes is an Assistant Professor of Medicine, Endocrinology at The Warren Alpert Medical School of Brown University. She was a 2018-2019 NEJM editorial fellow. She completed her fellowship in Endocrinology at the University of Vermont.
NEJM Knowledge+