In this paper we describe a model of clinical information designed to make health information systems properly interoperable and safely computable. The model is a response to a number of categories of requirements, ranging from the semantic to the performance of software at runtime. We argue that the starting point of a successful model must be an ontological analysis of the process of clinical care delivery, seen as a scientific problem-solving process. From this approach we develop a classification of types of clinical information called the Clinical Investigator Record (CIR) ontology.