Background: Despite over 40 years' work on general systems theory, informed by critical social science, there is a mismatch between the theories used to explain and influence clinical practice in nursing and the way in which transferring new knowledge into practice is articulated.
Data sources: The analysis and emerging propositions were based on a critique of seminal texts published in English up to 2008 covering critical social science, action science, diffusion of innovations, practice development and the management of innovations.
Discussion: There is an implicit adherence to the world view that healthcare systems operate like machines, and much of the science generated around knowledge translation research tends to be logico-deductive. This is in direct contrast to the prevailing arguments of general systems theorists, who view the system more as an organism. Five propositions are posited: knowledge translation is a necessary but not sufficient mechanism to transform systems; the 'system-as-machine' metaphor is profoundly unhelpful to knowledge translation; the healthcare system is best viewed as a complex entity; successful innovation is a function of the level of local autonomy experienced by individuals, teams and the unit involved; innovation is most effective when it involves key stakeholders.
Conclusion: The purposeful integration of systems theory with knowledge translation theories and models may enable the application of research and new knowledge to practice to be speeded up.