Safeguarding patients' private information is one of the most challenging issues in the design and implementation of modern e-Health systems. A patient's consent should usually be obtained before his/her private information can be disclosed. Recent advances in Hippocratic Databases (HDB) show a promising direction towards the enforcement of privacy policies in e-Health systems. With HDB, patients need to specify their privacy preferences about what data to be disclosed to which recipients for what purposes. However, this may become a daunting task in a complicated application that involves potentially a large number of combinations of data recipients, purposes, and granularities of data. This thesis tackles issues in applying the HDB design to e-Health systems. More specifically, I design an architecture for integrating APPEL preferences with HDB, extend the original HDB design to support fine-grained privacy authorizations demanded by patients and adapt the design to a multidimensional model; I propose a series of methods for patients to more conveniently specify their privacy preferences based on the hierarchies naturally existing in each dimension of a privacy preference, define meta-policies to resolve potential conflicts between preferences specified over time, and discuss how to represent such preferences with a snow-flake schema in backend databases. Finally, I illustrate the implementation issues and justify my designs with experimental results.