Hi, I'm trying to figure out if a Bayesian multilevel parametric survival model can fit for the follow study:
n patients received a lifelong regimen of treatment which include a combination of 5 drugs that could start at different time points and can be overlapped. The regimen of treatment start after organ transplantation Blood level of drugs and times were collected as repeated measures (long format), were (id) identifies patients, (drug) the type of drug administered, (levels) the blood levels of administered drug, (days) the time from the transplantation to the first blood level determination and so on (baseline levels are 0 for all treatments)
The aim is assess if the curves that describe blood levels of drugs over time during the first year are associated with survival
I would like to use multiple strata for id and drugs and then modelling the curves of blood levels of drugs over time but I'm not sure it could be feasible as drugs could be overlapped and start at different time point.
Any suggestion?