We propose a novel applied decision analytics solution in clinical outcomes analysis for deriving outcomes to be used as benchmarks in designing appropriate therapies in personalized medicine and predictive pharmacology.  The efficacy of comparative effectiveness research in clinical medicine and pharmacology is limited by the lack of a defined solution to derive clinical outcomes across diverse patient populations and a variety of disparate data sources that collectively define a clinical profile at particular point in time.  An outcome at time T1 is driven not only by static factors such as race, ethnicity and occupation, that are generally time-independent, but also by the condition profile and resultant outcome of the patient’s condition at T0.  Our solution is an ensemble analytical framework that leverages a temporal rule induction algorithm to create derived outcomes profiles across the time continuum.  It performs analysis on structured and unstructured data from EMR/EHR, clinical, biological, biomarker, behavioral and demographic data sources that are integrated into a composite data warehouse via our propriety semantic resolution and natural language processing algorithms.  The outcomes profiles reflect an index or aggregate score for the amalgamation of all available data for a particular patient at a particular time.   Outcomes profiles from thousands of samples are catalogued and normalized in a registry and are used to establish a baseline matrix for application in higher level statistical and predictive analyses for comparative effectiveness studies in pharmacology.  Using this approach, it is possible to determine based on available data both the appropriate treatment to affect a desired outcome and the predicted outcome based on a given treatment at a given time.