Policy ForumRESEARCH ETHICS

Bystander risk, social value, and ethics of human research

See allHide authors and affiliations

Science  13 Apr 2018:
Vol. 360, Issue 6385, pp. 158-159
DOI: 10.1126/science.aaq0917

Embedded Image

A lab worker exposes his arm to Aedes aegypti mosquitoes, which spread the Zika virus.

PHOTO: BLOOMBERG/CONTRIBUTOR/GETTY IMAGES

Two critical, recurring questions can arise in many areas of research with human subjects but are poorly addressed in much existing research regulation and ethics oversight: How should research risks to “bystanders” be addressed? And how should research be evaluated when risks are substantial but not offset by direct benefit to participants, and the benefit to society (“social value”) is context-dependent? We encountered these issues while serving on a multidisciplinary, independent expert panel charged with addressing whether human challenge trials (HCTs) in which healthy volunteers would be deliberately infected with Zika virus could be ethically justified (1). Based on our experience on that panel, which concluded that there was insufficient value to justify a Zika HCT at the time of our report, we propose a new review mechanism to preemptively address issues of bystander risk and contingent social value.

Bystander Risks

A Zika HCT would pose risks to bystanders not enrolled in the study because subjects could transmit the virus through sexual activity, pregnancy, mosquito vectors, or other unknown ways (2). Zika HCTs are not the only type of research involving bystander risk. Dual-use research, such as research on avian influenza that could be used to make biological weapons, could harm bystanders. HIV cure trials that withdraw antiretrovirals to study new treatments can place subjects' sexual partners at risk of HIV infection (3). Experimental mitochondrial replacement techniques can affect future children (4).

There are some, albeit incomplete, protections for certain types of bystanders. In dual-use research, bystander risk is addressed by institutional biosafety committees and, in the United States, by the National Safety Advisory Board for Biosecurity (NSABB). Many countries have regulations on research involving pregnancy and reproduction. For research that can harm identifiable communities by generating data that could cause stigma or contradict cultural beliefs, community consultation can help address group harm, although this may be difficult to implement for fragmented communities (5).

However, there is no clear or systematic mechanism for protecting bystanders from research risks arising through sexual or environmental transmission. These types of bystanders are generally not attended to in research regulations, international ethics guidance, and institutional review board (IRB) deliberations because they are not “human subjects”—researchers do not directly intervene or interact with them, or collect their identifiable data (6). IRBs are instructed only to consider risks to subjects, and because “possible long-range effects” should not be considered, it is unclear whether existing regulations permit IRBs to address bystander risk (7). A U.S. Institute of Medicine report merely instructs investigators that they may have ethical obligations to anticipate and plan to address bystander risks that are “foreseeable and significant” (8).

It is important to protect all research bystanders because they may be unable to protect themselves; obtaining their consent might be impossible in some cases and problematic in others. For example, Zika HCT participants might choose sexual partners spontaneously or anonymously, precluding advance consent from all at-risk bystanders. Some studies might expose large numbers of bystanders to risk of environmental transmission, rendering consent infeasible. One approach might be to inform participants about risks to bystanders in consent forms. Yet researchers cannot simply pass their responsibilities on to subjects who lack accountability to inform and protect others.

Unfortunately, researchers or IRBs attempting to address bystander risks will find no consensus on or framework for determining when bystander risks are ethically and legally justifiable (6). For instance, an individual could be infected with Zika virus from an HCT participant, and then become pregnant. This risk could be mitigated but not eliminated. How high of a chance of bystander harm should be tolerated? How should direct conflicts between the interests of participants and bystanders be resolved?

IRBs are not well-suited to consider these issues. IRBs are explicitly charged with protecting participants' welfare and liberty, but lack a mandate to protect bystanders. They may also be reluctant to address such unsettled ethical issues (9). Public trust is critical to biomedical research, and the public often reacts strongly when researchers or institutions impose risks on individuals—even if risks of similar magnitude are already naturally present (10). Taking steps to protect bystanders may help avoid public outcry in the event that bystanders are harmed.

Substantial Risk, Contingent Value

In addition to bystander risk, Zika HCTs pose risk of potentially long-term harms to participants (11), and substantial uncertainty about the level of risk. That HCT researchers deliberately induce pathology in otherwise healthy people may be hard for the public to understand, even if ethically justifiable (12). When direct benefits cannot justify the risks participants face, the risks must be weighed against potential social value. Zika HCTs therefore needed an especially high social value to justify the risks, but measuring this value was challenging because it hinged on several factors external to the research.

The panel deliberated about whether results from Zika HCTs were likely to meaningfully contribute to Zika vaccine research. This required considering evidence about active or planned research trajectories for candidate vaccines from regulators and researchers. HCTs are not part of the standard requirements for regulatory approval to license and market new vaccines, so the extent to which regulators would be willing to rely on them was unclear. We assessed whether the results from proposed research would alter the research trajectory in some critical way and whether the epidemiology would continue to make field testing possible and necessary.

Addressing highly contingent social value is a persistent ethical challenge, with some clear (if not exhaustive) examples. When the Ebola epidemic was ongoing and different studies sought to test many interventions in the same population, the value of particular trials was relative to the alternatives and depended on outcomes of others in a way that was difficult to predict and stage (13). Bioterrorism countermeasure studies, such as anthrax vaccine studies, need to be relevant to current threats in order to have social value, but the probability of an attack is hard to assess. Some have argued that trials with highly uncertain risk, innovative designs, and broader policy implications (e.g., repeated transplantation of neural stem cells) require a new approach to ethics review (14).

Our current system is ill-equipped to handle this challenge. Many studies may not require exacting judgments about social value, but if the risks are contentious, precision becomes more important. IRBs review one protocol at a time and lack both the capacity and mandate to gather evidence to contextualize each individual protocol. Decisions made by individual IRBs may also be inconsistent with each other, causing unfairness and confusion. The type of programmatic assessment done by the Zika HCT ethics panel, in contrast, required forecasting the trajectory of the Zika vaccine research program and evolution of the epidemic. Such information may be necessary to determine whether a study contributes to a broader research program, and thus has adequate social value, rather than duplicating effort or generating results that will not be integrated into a path toward development or application.

A More Comprehensive Approach

The type of review provided by the Zika HCT ethics panel was an important response to an evolving research ethics landscape where the status quo was insufficient. IRBs tasked primarily with protecting individual subjects in individual protocols are not the right bodies to address programmatic issues we identified.

We therefore propose that agencies funding biomedical research establish databases of reviewers qualified to serve on ad hoc Comprehensive Ethics Review Committees (CERCs). CERCs should involve ethicists, policy experts, clinicians, patient representatives, and scientists, supplementing with subject matter expertise and/or community input as needed. They should conduct proactive review of research programs (e.g., all proposed projects involving a new study design in a particular disease area) and have at least two clear triggers for activation: (i) substantial risk to bystanders who cannot protect themselves, and (ii) research with contentious risk and highly contingent social value. CERCs should be activated by members of funding agencies and their reviewers, and have jurisdiction primarily over projects submitted to a particular agency, but also made available to IRBs or researchers.

To illustrate how CERCs could fill important gaps in the current research oversight system, consider how the Zika HCT ethics panel engaged with the bioethics literature, ensured its members had expertise in relevant ethical and scientific areas, consulted with other experts, and gathered additional evidence. This went far beyond what a typical IRB could reasonably accomplish, and would be duplicative and taxing if marshalled at every institution facing similar questions.

We assessed the acceptability of bystander risks by comparing their likelihood and magnitude to data on adverse events from similar trials that are generally viewed as ethically acceptable (e.g., phase 1 drug trials with healthy volunteers, malaria HCTs). We noted that key uncertainties about Zika virus transmission might resolve, and protections to avoid the most serious consequences for bystanders could then be developed. We therefore proposed high-priority research questions identifying the duration of infectivity and modes of transmission associated with Zika virus to be addressed in advance of a Zika HCT.

To assess highly contingent social value, the panel was empowered to gather evidence about active or planned research trajectories for various candidate vaccines from regulators and researchers, and were not confident that results from proposed Zika HCTs were likely to be harnessed in a useful way. To underscore the contingent nature of this judgment, circumstances have changed in the past year. Field trials of Zika vaccines have become difficult as ongoing outbreaks are unpredictable or short-lived, yet the threat of future (and larger) outbreaks has not gone away (15). Zika HCTs may become the only way to prioritize vaccine candidates, giving them clear and considerable social value.

Some may object that generalizing and institutionalizing this approach could slow valuable research by adding an additional layer for review. However, embedding this process within funding agencies could preempt ethical problems that might otherwise stymie research. Concerns that CERCs might suffer from “mission creep” could be countered by establishing clear charters and triggers for deploying CERCs. Unlike IRBs, their opinions should be publicly available to provide precedent for future research programs or for IRBs evaluating particular protocols at a later date. CERCs should work with existing review bodies, such as by referring dual-use issues to the NSABB or sending protocols not requiring heightened scrutiny back to IRBs for standard research review. Using CERCs to proactively address these critical ethical issues may signal a commitment to ethical research to policy-makers and the public, which could redound to the benefit of researchers through future funding and support.

References and Notes

Acknowledgments: This manuscript does not necessarily reflect the positions or policies of the U.S. Department of Health and Human Services, the National Institutes of Health, the National Institute of Allergy and Infectious Diseases, and/or the Department of Defense Walter Reed Army Institute of Research. The authors thank B. Eisinger, H. Marston, C. Paules, C. Dieffenbach, Z. Emanuel, N. Michael, C. Grady, D. Opel, N. Garrison, E. Weiss, J. Yu, J. Krick, E. Berkman, J. Jeavons, S. Kraft, L. Dawson, and an anonymous reviewer for helpful comments. The authors particularly thank F. McCutchan, who also served on the Zika HCT ethics panel, for valuable insights.
View Abstract

Stay Connected to Science

Navigate This Article