Episode 18, “A Statistician Reads JAMA"
On today's episode of "In the Interim...", Dr. Scott Berry discusses picking up the current JAMA and reading the article on the FAIR-HF2 trial on IV iron supplementation for patients with heart failure and iron deficiency and wow, what an interesting study and publication of the results!
The trial narrowly missed demonstrating significance due to using the Hochberg procedure for testing the “high” deficiency subgroup and separately to the heart failure endpoint. The result for the primary outcome would have been statistically significant without the adjustments. A subjective decision before the trial to allocate a small amount of alpha to a subgroup of the patients completely changes the conclusion of a 6-year, 1,100-patient trial in a medical journal.
And frequentists complain about Bayesian prior distributions…
For further insights to this intriguing study, listen to today's full episode: https://www.berryconsultants.com/resource/a-statistician-reads-jama