A major goal of precision medicine is to use patient information to determine an optimal treatment for the patient.  Patient information comes in the form of data and data modalities have become more complex over time.  Also, many current ITRs do not make the most efficient use of available data.  In this talk, a review of some recent work on developing ITRs using functional data analysis approaches will be given.  Newer results will be presented for developing ITRs using scalar predictors that incorporate longitudinal outcomes instead of the more traditional scalar outcome approaches.  These methods will be illustrated with simulations as well as data from a depression clinical trial.