Consent Approach and Patient Notification in Pragmatic Comparative Effectiveness Trials in Critical Care

Advances in clinical informatics have facilitated the emergence of the learning healthcare system (LHS) model, which is defined by the National Academy of Medicine as one in which knowledge generation is so embedded into the core of the practice of medicine that it is a natural outgrowth and product of the healthcare delivery process and leads to continual improvement in care (1). Pragmatic comparative effectiveness research plays a vital role in an LHS by rigorously evaluating two or more standard-of-care interventions under real-world conditions. Indeed, pragmatic clinical trials are increasingly used in critical care research (2), and at least one large survey study suggests widespread acceptance by the general public (3). However, conducting such real-world research often requires rethinking traditional research regulations and ethics (4–6).

Ethical and regulatory requirements for individual informed consent in human subjects research serve as important safeguards. Indeed, in response to atrocious historic research abuses, the 1979 Belmont Report established the foundational ethical principles for human research in the United States as “respect for persons,” “beneficence,” and “justice.” In 1991, federal regulation 45 CFR §46, better known as the Common Rule, was developed based on these principles to create a unified standard for human subject protections in research. However, this regulation was not developed with comparative effectiveness research or the LHS model in mind, and its interpretation has not been uniform across institutional review boards. To address some of these challenges, a Revised Common Rule was implemented in 2019 that sought to improve the informed consent process for participants and to reduce administrative burdens for minimal risk research studies.

The Isotonic Solutions and Major Adverse Renal Events (SMART) trial was conducted at an LHS (7) and demonstrates the potential for pragmatic comparative effectiveness research to improve intensive care unit (ICU) care and outcomes. In this cluster-randomized trial of more than 15,000 critically ill adults, there was a small but significant reduction in major adverse kidney events with the use of balanced crystalloids compared with saline. Because of the ubiquitous use of intravenous fluids in ICU care, translation of these results to routine ICU care could yield tremendous population benefit. However, obtaining individual informed consent would have been impractical, likely taken decades to complete (8), and resulted in substantial selection bias. The SMART trial was approved with a waiver of informed consent because of the minimal risk posed by participation, the time-sensitive nature of the treatment decision, and the standard-of-care treatments being compared. Yet, little is known about critically ill patients’ perspectives on comparative effectiveness research in the ICU and the use of such alterations of informed consent.

In this issue of AnnalsATS, Palakshappa and colleagues (pp. 1027–1034) describe the perspectives of 20 survivors of critical illness from two medical centers regarding the conduct and postparticipation notification of pragmatic comparative effectiveness research (9). The authors performed a reflexive thematic analysis of semistructured interviews conducted with patients 2–15 weeks after a hospitalization that included at least 24 hours in the ICU. During the interviews, participants were asked to consider two hypothetical comparative effectiveness trials with interventions embedded in routine care delivery: one trial compared two standardly used sedative medications for endotracheal intubation, and the other compared a nurse transitional care team after ICU discharge versus usual ward transfer.

The authors report that patients with lived experience generally support pragmatic comparative effectiveness studies in the ICU. Factors that positively influenced participants’ perspective of pragmatic comparative effectiveness research included having trust in their healthcare providers, a history of positive experiences in the healthcare system, and being familiar with medical research. Participants were also motivated by altruism and a desire to contribute to new knowledge. Interestingly, trust in healthcare providers could also negatively influence participants’ receptivity to research participation because of concerns about restricting their decision-making autonomy with random treatment assignment and a research protocol. Some patients expressed the view that all research was experimental, regardless of if the treatments or care delivery processes under study were already considered standard of care. Most patients shared the view that postparticipation notification of enrollment in a pragmatic comparative effectiveness research study was desirable and important for respecting autonomy, but they did not necessarily agree on when or how this debriefing should take place.

Palakshappa and colleagues’ findings are largely consistent with prior studies that have shown wide acceptance for alterations of informed consent in pragmatic research studies but less enthusiasm for no notification (10, 11). They are also aligned with findings from a qualitative study of ICU surrogates’ decision making on behalf of a loved one regarding research participation (12). Although qualitative studies often have limited representativeness because of their small sample size and purposive sampling methods, the validity and generalizability of Palakshappa’s results are further strengthened by inclusion of patients with broad eligibility criteria from two different medical centers, a large academic center in the Southeast and an integrated healthcare system in the Midwest. The authors applied their findings to propose a framework for investigators to ensure respect for patient autonomy throughout the research planning, design, conduct, and dissemination phases.

Applying this framework to the SMART trial and others like it suggests that postparticipation notification should be strongly considered even when not required. Much guidance is still needed to understand how investigators and health systems can operationalize individual patient notification of participation in large pragmatic comparative effectiveness trials. For example, in-person or even verbal notification would not be plausible for many pragmatic trials that enroll thousands of patients. Moreover, trials enrolling seriously or critically ill patients may be unable to notify a substantial portion of participants because of death or prolonged incapacity, and it is unknown if grieving surrogates would similarly want to receive such notification. In addition, because postparticipation notification typically includes an option for withdrawing one’s data, it is not clear that this option would extend to deceased patients who are not generally considered to fall under the same human subjects protections, which raises concerns of fairness and data biases. We hope that these and many other unanswered questions about patient notification in pragmatic clinical trials, including optimal approaches for dissemination of results to participants, will be the targets of future research.

Aligning the research process with the priorities and preferences of patients and their families is central to patient-oriented research. Pragmatic comparative effectiveness research and LHS share this goal from the perspective of the population and healthcare system, which at times can seem or truly be at odds with the preferences of individuals. The tension between population health and individual health care is neither new nor unique to pragmatic trials (13) and underscores the importance of engaging patient and caregiver partners throughout the research process to help guide investigators where traditional research ethics and regulations currently fall short.