OBJECTIVE: Clinical evidence is often reported as an average treatment effect across a large population. This is appropriate if all patients experience the same effect from a given treatment. However, more often, different patients experience different outcomes on the same medication. If this is true, then averaging the effects of treatment obscures the outcomes received by most patients. It also makes it difficult for physicians to utilize this evidence to select the most appropriate treatment for individual patients. This interpretation of average outcomes by physicians leads to geographic variation, inappropriate care, and increased health care costs. An essential step towards optimizing therapy is to provide evidence that recognizes inter-individual differences in drug response. METHODS: The PPS is defined as the expected outcome (on control) given the individual's covariates. To calculate the PPS, the outcome of interest is regressed on the covariates for those patients treated with the control(Drug A). Using the coefficients from this model, in conjunction with patient characteristics, the PPS is computed for all patients; as if every patient was a member of the control group. Variations in treatment effect are then identified across subgroups by partitioning patients, according to PPS, into strata and calculating the treatment effect within each stratum. This analysis is repeated using the alternative treatment(Drug B) as the control. By identifying and comparing the stratum that receives the optimal benefit from each treatment, the patient characteristics that are uniquely associated with success on Drug A and Drug B can be determined. RESULTS: To demonstrate the use of the PPS, a convenient sample of California Medicaid beneficiaries diagnosed with schizophrenia will be used. CONCLUSIONS: The outlined approach will allow physicians to more accurately prescribe the most beneficial treatment for each and every patient, by linking patient characteristics to treatment success.