The PsN prospective evaluation tool is pretty cool. It can evaluate the predictive performance of a model on the individual level.
We could probably coax TDMore to do the same thing. This may even allow us to quickly evaluate the prospective predictive power of several different models.
Specifically for Tacrolimus, this exercise may also be interesting: does it matter which model we use? Does it matter if we specify a model with misspecified covariates? Or if we only use the base model?
Prospective evaluation can show the impact on concentration predictions, but TDMore can show the true clinical impact : will we recommend different doses?
The PsN prospective evaluation tool is pretty cool. It can evaluate the predictive performance of a model on the individual level.
We could probably coax TDMore to do the same thing. This may even allow us to quickly evaluate the prospective predictive power of several different models.
Specifically for Tacrolimus, this exercise may also be interesting: does it matter which model we use? Does it matter if we specify a model with misspecified covariates? Or if we only use the base model? Prospective evaluation can show the impact on concentration predictions, but TDMore can show the true clinical impact : will we recommend different doses?