The DILEMMA project has addressed clinical, technical and methodological questions in the design and implementation of aids for medical decision making and patient management. Technically, the project has focused on the use of logic programming and knowledge-based methods for constructing a generic technology to support clinical decision making, and for assisting in the scheduling and execution of clinical tasks in protocol-based care in 3 main clinical settings: general medical practice, to provide support for a range of clinical tasks, including prescribing and guideline-based management of common conditions like asthma and otalgia; hospital-based cancer care, paying particular attention to the use of oncology protocols; shared care of cancer and cardiology patients outside the hospital by primary care physicians, nurses, and paramedics, in consultation with hospital-based specialists. DILEMMA has succeeded in integrating its generic decision support technology and electronic health care record with existing clinical information systems in a number of clinical domains. DILEMMA has built on conventional primary and secondary sector clinical informatics systems by providing clinicians with protocol-based decision support. DILEMMA technology has aimed to extend the capabilities of existing conventional clinical information systems by adding value in such areas as patient data entry and retrieval, hypermedia database access, diagnosis, care planning, prescribing and follow-up. The development of clinical applications in DILEMMA was underpinned by a methodology for establishing clinical and functional requirements for decision support systems, and by a generic business model which formalises protocol-based care. DILEMMA technology included: an electronic patient record, developed to accommodate protocol-based care plans; a protocol-oriented system for supporting clinical decision making and the management of care plans; specific clinical applications in hospital oncology, primary care and cardiology shared care. These applications were linked to the decision and protocol support functions which access patient data stored in the electronic patient record.