The rise of quality improvement research and comparative effectiveness research in health care settings constitutes progress toward the goal of what the Institute of Medicine has called a “learning healthcare system,” in which we are “drawing research closer to clinical practice by building knowledge development and application into each stage of the healthcare delivery process.” As clinical research and clinical practice move closer to a deliberately integrated system, the distinction between the two is increasingly blurred, although the sharp distinction in U.S. regulations and research ethics literature remains in place. In the 1970s and for two decades thereafter, this distinction was helpful: for some forms of research, it sheds light on which activities require ethical oversight. Research that is closely integrated with health care—notably, health delivery research—was then uncommon, however. That is no longer the case, and regulations and research ethics need to change to accommodate the new landscape.
In this paper, we argue that conceptual, moral, and empirical problems surround the received view that we can and should draw sharp distinctions between clinical research and clinical practice. We start with the history of the research-practice distinction in the reports of a U.S. national commission and in U.S. federal regulations, and then offer a critical assessment of five characterizations of research that have been used in policy documents and the scholarly literature to try to make a sharp distinction between research and practice. We challenge the clarity and the tenability of these characterizations as a way of distinguishing research from practice. We argue that the received view of the research-practice distinction leads to overprotection of the rights and interests of patients in some cases and to underprotection in others. We contend that a new ethical foundation needs to be developed that facilitates both care and research likely to benefit patients, and that provides oversight that, rather than being based on a distinction between research and practice, is commensurate with risk and burden in both realms.