Changes to Informed Consent in the Era of Learning Healthcare Systems

In 2007, a seminal report from the Institute of Medicine (IOM) threw a sharp spotlight on a series of problems facing the broader U.S. healthcare system:

Evidence on what is effective, and under what Cover page of IOM report - The Learning Healthcare Systemcircumstances, is often lacking, poorly communicated to decision makers, or inadequately applied, and despite significant expenditures on health care for Americans, these investments have not translated to better health.

—The Learning Healthcare System (IOM Workshop Summary), 2007

The IOM report called for a new approach to closing the existing gaps in patient care, one that would create a system in which patient care, quality improvement efforts, and clinical research would exist as integrated components within a virtuous cycle of feedback—the “learning healthcare system.” (The original report, plus a series of related reports [PDF], can be read for free online at the National Academies Press website.)

In the years since The Learning Healthcare System was first published, researchers, healthcare providers, health systems, and governmental and regulatory agencies have all struggled with the monumental task of building such a system. One aspect that has presented a particularly complex set of challenges centers on the issue of informed consent. The principle of informed consent—the idea that all patients have the right to make a fully informed decision, free from coercion or other undue pressure, about whether or not to participate in research—is a foundational tenet of clinical research ethics. However, some of the tools that are widely considered to be critical to the success of achieving a workable learning healthcare system, such comparative effectiveness studies and cluster-randomized trials, are difficult or even impossible to conduct under existing models of individual informed consent.

Recent efforts from the Ottawa Consensus Statement Group and a series of articles published in an issue of the Hastings Center Report have explored explored informed consent in such circumstances. These are now joined by a pair of articles published in the February 20, 2014 issue of the New England Journal of Medicine. In the first, Faden and colleagues outline a case for streamlining or even dispensing with individual informed consent in certain kinds of randomized comparative-effectiveness or quality improvement studies that present a minimal risk of harm to patients [1]. The authors also describe the larger framework that would provide transparent and accountable oversight of such studies, as well as overseeing the integration of findings from such research into the patient-care process.

The second article, by Kim and Miller, presents a different vision for informed consent,  the “Integrated Consent Model” [2]. Unlike Faden and colleagues, the authors argue for preserving the element of informed consent for all randomized pragmatic research, even in circumstances considered to pose minimal risk to patients. Kim and Miller suggest that such a model, in which the prospect of participating in research is offered as part of the general clinical discussion about treatment options and documented by the physician, will satisfy ethical imperatives for informing patients while remaining sufficiently streamlined to meet the demands of pragmatic clinical research. They also argue that an “integrated consent” approach could be accommodated under existing regulations.

Regardless of whether either or both of these perspectives can be incorporated into the emerging learning healthcare system, the larger questions surrounding informed consent are  garnering significant interest, as evidenced by the response to the Department of Health and Human Service’s 2011 call for public comment on a proposal to modify rules governing clinical research. And with the proliferation of new technologies and trial designs that use data extracted directly from patient’s electronic health records (such as the demonstration projects coordinated through the NIH Collaboratory) and the emergence of innovative networks devoted to pragmatic patient-centered research, the need for a solution both protects patients and enables vitally needed research will only continue to grow.

For additional perspective on the recent publications in the New England Journal of Medicine, see "Research Permissions–Angels on the Head of a Pin, or the Key Issue to Decipher?" by NIH Collaboratory PI Dr. Rob Califf.


1. Faden RR, Beauchamp TL, Kass NE. Informed consent, comparative effectiveness, and learning health care. N Engl J Med 2014;340:766-768. 10.1056/NEJMhle1313674.

2. Kim SYH, Miller FG. Informed consent for pragmatic trials — The integrated consent model. N Engl J Med 2014; 370:769-772. doi: 10.1056/NEJMhle1312508.