There is evidence that health information technology can improve quality, safety and reduce costs but that health care providers needed more information about how to implement these technologies to realise its potential. This paper summarises the research and proposes a theory of implementation based on the research evidence. The second part describes two implementations of electronic medical record systems and compares the theory against the findings of these two case studies. The paper provides implementers with research-informed guidance about effective implementation, contributes to developing implementation theory and notes policy implications for current national strategies for IT in health.