Abstract: A system for processing patient health information (PHI) protects the confidentiality of PHI to achieve regulatory compliance. The PHI contains patient medical data and associated patient identification data. A de-identification agent extracts patient medical data and separates from all identification data to create de-identified patient data. A key is generated that allows subsequent reassociation of the patient medical data and the patient identification data. The de-identified patient data base may be queried for patient screening purposes. Patient queries are processed only if the study or patient screening has been authorized by appropriate authorities, such as an internal review board. Patients whose medical characteristics conform with the patient query are selected for possible use in a study. If re-identification of the selected patients is necessary, and authorized, the key may be used to provide the necessary reassociation. A data log records all access to patient data.