I'm planning a study and am considering going with a sequential analysis design (as described here: https://psyarxiv.com/9yegd/download?format=pdf). This design allows one to take multiple looks at the data, while still controlling the type I error rate.
One of my analysis options is mixed-effects modelling, and I'm curious about how that would fit into the sequential analysis design. Should an interim-analysis p-value threshold be applied to each parameter estimate, or is further threshold correction needed? Also, should the same threshold be applied to each model (e.g. intercept-only model, level-1 predictors model etc.) in an interim analysis or are further corrections needed?
I'd be thankful if someone could provide references or general guidelines for conducting mixed-effects modelling (or something similar, like multiple regression) in a sequential analysis framework.
