Improved adherence adjustment in the Coronary Drug Project

@NoahHaber @Andrew___Baker the other: https://t.co/W4HPXOXefC

And it's @EpiEllie's research! https://t.co/PvHCo0TRq3

@Jabaluck @ashtroid22 @autoregress Our group also has a number of papers comparing the use of obs epi causal methods to control for adherence to placebo & showing that when you expect no direct effect of placebo it is possible to recover the null throu

@BenMazer @JAMA_current But the effect of adherence to placebo arm is largely removed if you use more appropriate methodology for per protocol analysis: https://t.co/0z5b1L19wY