[Context] Interdisciplinary healthcare teams (IHTs) include practitioners from different disciplines who collaborate for providing care to patients. IHTs often follow clinical workflows composed of tasks that must be executed by practitioners with specific capabilities. The membership in an IHT can however evolve over time for a given patient. [Problem] Existing Business Process Management (BPM) suites and their workflow execution engines are designed for supporting and monitoring general workflows, but they are insufficient in supporting the allocation of tasks to the most suitable practitioners during the execution of healthcare workflows in a dynamic context. [Methodology] Using Design Science Research, this thesis builds on top of an existing semantic layer, which includes an ontology defining IHT team concepts and relationships that are used to reason automatically about team dynamics, in order to add dynamic team management to BPM suites. It does so by proposing and designing middleware (including a generic interface) that enables the semantic layer to command the BPM suite to allocate suitable practitioners to tasks during the execution of clinical workflows. The design and implementation of this middleware are discussed, and the latter is tested on a commercial BPM suite for two realistic clinical processes. [Results] The proof-of-concept implementation demonstrates the feasibility of using middleware with a generic interface to add support for IHT executing BPM suite when managing a patient. In addition, the thesis also demonstrates that the ontology used in the semantic layer is minimal, that is, all of its concepts and relationships are necessary for the required team functionalities (usually absent from BPM tools) to work properly.