Immortal time bias

[malcolmbarrett]

• mentioned in the target trials section
• in data prep section
• in sensitivity analysis
• covered in depth via cloning in causal inference over time

[malcolmbarrett]

After sleeping on this, I think @LucyMcGowan's discussion on simulating a dataset for this might be best. Then we can deal with all the issues involved with this type of data/bias while also having the data be kind of fun or different

[malcolmbarrett]

In retrospect, I think the right place for this is the longitudinal chapter so we can discuss the cloning solution to this, with a shoutout in the time zero discussion earlier in the book