Background

There is historical and ongoing controversy regarding the ethical conduct of research involving human subjects. Throughout the 20th century, several studies designed to gain medical knowledge were carried out at the cost of harm to marginalized or vulnerable populations [1]. Recognition of these abuses led to the articulation of ethical principles for clinical research and development of regulations to ensure the protection of human subjects. The Belmont Report outlines three fundamental ethical principles for human subjects research: respect for persons, beneficence, and justice [2]. These principles serve as the foundation for federal research regulations, which are codified as 21 CFR 50 (FDA regulations on the protection of human subjects) and 45 CFR 46 (the Common Rule).

Though these policies set in place important protections, they viewed clinical research as a highly controlled system somewhat separate from medical practice. This divide has resulted in a failure to provide decision-makers with high-quality evidence to make the best choices in medical practice [3-5]. As the field of clinical research has evolved to address this gap, it has become challenging to apply the current regulatory and ethical paradigms. This is evident among an increasing number of pragmatic clinical trials, which study clinically relevant alternatives (often approved or accepted treatments) in representative populations at the point of clinical care [5,6].

In contrast to traditional clinical trials that study new therapies in highly controlled settings, pragmatic clinical trials rely on streamlined processes to measure outcomes in heterogeneous populations in real-world clinical settings [6]. Some pragmatic trials use cluster randomization, in which groups (e.g., clinics, hospitals, cities) are randomized rather than individuals. Because of these differences, the application of ethical principles and regulations to pragmatic clinical trials is complex [7,8]. Questions include, what constitutes research versus a quality improvement initiative under current regulatory guidelines; how should the criteria for determining what is minimal risk research be appropriately applied; and when is a waiver or alteration of informed consent ethical and justified. As observed in the Surfactant, Positive Pressure, and Pulse Oximetry Randomized Trial (SUPPORT), there can be risks and ethical controversy even in studies comparing standard or accepted treatment options.

There have been some attempts to respond to some of the ethical and regulatory challenges faced in the conduct of pragmatic clinical trials, and more broadly, in a healthcare system that integrates research with clinical care. For example, the Ottawa Ethics of Cluster Randomized Trials Consensus Group issued a statement with recommendations for the ethical design and conduct of cluster randomized trials [9]. Issues specific to pragmatic cluster randomized trials were discussed in a subsequent workshop convened by the NIH Health Care Systems Research Collaboratory [7]. Kass and colleagues have argued that the distinction between clinical practice and research is increasingly blurred as our healthcare system advances toward a learning healthcare system [10] and current regulations and ethical principles do not account for this changing landscape. They proposed a new ethics framework for the learning healthcare system [11,12], which has been discussed and debated [13,14].

Part of the changing research environment involves a shift toward patient-centeredness [15]. Patient-centered outcomes research engages patients throughout the research process in determining what research questions are important, providing input during study design and conduct, and planning the dissemination and implementation of results [16,17]. In keeping with patient-centered research principles, engaging patients in shaping ethical and regulatory guidelines for clinical research would promote autonomy and show respect for persons [18].

NIH Collaboratory and PCORnet Writing Project

Recognizing that there are unresolved ethical and regulatory issues critical to the conduct of pragmatic clinical trials [7,8], the NIH Health Care Systems Research Collaboratory and the National Patient-Centered Clinical Research Network (PCORnet) invited a group of stakeholders, including ethicists, clinical trialists, institutional review board professionals, and patient representatives, to address a set of 11 priority issues. The goal was to produce academic manuscripts that would thoroughly review the topics; identify best practices and provide guidance for future pragmatic clinical trials where possible; and describe issues that remain to be resolved. Work groups initiated discussions over teleconferences and then convened at 1.5-day face-to-face meetings with time for cross-group discussions; 78 attendees from 50 organizations were in attendance. The culmination of this effort is a special issue in Clinical Trials, as outlined below.