This paper addresses a process in which we combined educational guidelines (EG) from heterogeneous sources in one set of coherent computable statements to support dynamically generated and precisely tailored patient education material. The Guideline Interchange Format (GLIF), predicate logic and decision tables were assessed. An extended formalism of GLIF was applied to break up composite sentences of the educational material in atomic sentences. The differentiation of atomic sentences and combinations of atomic sentences from heterogeneous sources lead to a simplified overall content and model, and a significant reduction of conditional sentences in the EG. The resulting streamlined and personalized guidelines are expected to provide an improved user experience.