In this paper we present a novel approach to generate tailored patient education materials by using an evidence-based clinical practice guideline to model patients and direct the selection of relevant education content. We converted a Canadian guideline for managing dyslipidemia into the XML-based GEM formalism, and then modified it to support the tailoring of patient education material. We created the information content suitable for tailored messages, along with a document structure needed to present them. A rules engine was developed to process the newly converted GEM-based guideline and create the tailored documents based on patient data entered.