Primary objective: much research and practice related to the design and implementation of information technology in health care has been atheoretical. It is argued that using extant theory to develop testable models of health information technology (HIT) benefits both research and practice. Methods and procedures: several theories of motivation, decision making, and technology acceptance are reviewed and associated theory-based principles of HIT usage behaviour are produced. Main outcomes and results: the case of medical error reporting technology is used to support the validity of the proposed HIT usage behaviour principles. Further, combining these principles produces a testable, theoretical multilevel model of HIT usage behaviour. The model provides an alternative to atheoretical research and practice related to HIT. Conclusions: developing, testing, and revising models of HIT like the one presented here is suggested to be beneficial to researchers and practitioners alike. (Contains 4 figures.)