OBJECTIVE: To investigate the conformance of antipsychotic drug therapy used to treat Medi-Cal patients with schizophrenia to the Schizophrenia Patient Outcomes Research Team (PORT) Treatment Recommendations and to measure the impact of conformance on direct health care costs. METHODS: Minimum effective doses and completed therapy were defined based on the PORT Recommendations. We identified 2,567 patients with schizophrenia who were treated with antipsychotic medications during their first post-treatment year. Data were drawn from January 1987 to July 1996. Allowances were made for titration and changes in therapy. Multiple logistic regression models were used to identify factors affecting titration success and conformance. OLS regression models were used to estimate the impact of the PORT conformance on health care costs. RESULTS: Only 16.9% of the 2,567 patients treated with antipsychotic drugs achieved conformance with PORT recommendations. Fewer than 50% patients achieved titration success on their initial drugs, and the likelihood of PORT conformance increased with initial dose. Patients who received at least the minimum therapeutic dose on their first prescription were 2.9 times more likely to achieve conformance than patients who were prescribed less than 50% of the minimum dose (p<0.0001). This odds ratio increased to 8.2 for patients treated initially at > 3 times minimum dose (p<0.0001). Patients treated initially with fluphenazine were more likely to achieve conformance (odds ratio = 1.834, p<.0001), while thioridazine patients were less likely to conform (0.436, p=0.0228). PORT conformance was not associated with lower total costs (-$2,543, p=0.2453). CONCLUSION: In clinical practice, the rate PORT conformance was low and conformance did not significantly reduce total direct health care costs.