The Promise and Reality of Public Engagement in the Governance of Human Genome Editing Research

Abstract

This paper analyses the activities of five organizations shaping the debate over the global governance of genome editing in order to assess current approaches to public engagement (PE). We compare the recommendations of each group with its own practices. All recommend broad engagement with the general public, but their practices vary from expert-driven models dominated by scientists, experts, and civil society groups to citizen deliberation-driven models that feature bidirectional consultation with local citizens, as well as hybrid models that combine elements of both approaches. Only one group practices PE that seeks community perspectives to advance equity. In most cases, PE does little more than record already well-known views held by the most vocal groups, and thus is unlikely to produce more just or equitable processes or policy outcomes. Our exploration of the strengths, weaknesses, and possibilities of current forms of PE suggests a need to rethink both “public” and “engagement.”

Introduction

Genome editing is a vital research tool with broad applications to human disease. Its significance and novelty raise many questions about governance and policy, including who should determine the ethical boundaries of its use. In 1975, confronted with the challenges of writing safety guidelines for recombinant DNA technology, the government and public deferred to the authoritative judgment of scientists (Hurlbut 2015). This expert-driven, technocratic model of regulation (Blasimme 2019) has been gradually undermined by the new CRISPR-Cas9 technology, which has enabled the proliferation of human genome editing possibilities. In what may have been a turning point, the He Jiankui experiments with CRISPR-edited babies were described by Nobel laureate David Baltimore as a “failure in self-regulation by the scientific community” (Normile 2018). As part of a broader critique of top-down approaches, commissions, NGOs, academic commentators, and others considering genome editing research have nearly universally endorsed public engagement (PE) as a desirable, even essential feature of good governance and policymaking.

PE may provide valuable information to policy makers, enhance public understanding of and confidence in science, and be crucial to democratic governance (Iltis et al. 2021, 3; Kuzma and Cummings 2021; Gusmano et al. 2021; Adashi et al. 2020). But how should calls for PE translate into practice? As Nelson and colleagues observe (2021, 384), “most authors are content to call for greater public involvement—usually after research but before deployment—and leave the discussion there.” Relatedly, demands for “broad societal consensus” (WHO 2021b, 15) do not define how such consensus could manifest itself or be discerned.

One branch of PE that does address such questions is the deliberative democracy movement. It seeks to involve small, representative, and inclusive groups of citizens in deliberation, which goes far beyond discussion. It uses multiple modes of communication to promote pluralism—the development and expression of multiple perspectives--which may or may not equate with consensus, meaning acceptance of a single perspective (Curato et al. 2017). Deliberative democracy employs complex sampling methods to seek both demographic and discursive representation, the latter referring to representing “discourses”—viewpoints, discussions, and narratives—as distinct from individuals who are articulating them (Dryzek and Niemeyer 2008). Advocates see deliberative democracy as an answer to group polarization in deeply divided societies (Curato et al. 2017).

Despite logistical challenges, many deliberative democracy events (sometimes called citizen assemblies or juries) have occurred on a variety of bioethics topics, some closely related to genome editing. Several others are in the planning stages in the United Kingdom, Brazil, and elsewhere. The Australian Global Citizens Assembly example is discussed below. Others include—all at a relatively small scale—the 2019–2020 Dutch DNA Dialogue (Houtman et al. 2022), a series of citizen assemblies in Ireland in 2016–17, preparatory to a referendum legalizing abortion (Council of Europe 2019), a European network of climate change assemblies (KNOCA 2022), and an online public DNA debate in Belgium (Mayeur et al. 2021). While the Dutch dialogue, which has been extensively studied, did not result in participants changing their views in any consistent way, they reportedly appreciated the opportunity to improve their understanding of the issues (Houtman et al. 2022). The Irish assemblies were praised for helping to overcome social divisions on a polarizing issue (Council of Europe 2019). At a larger scale, during the mandatory seven-year review of France’s bioethics law in 2018, 21,000 citizens participated in 271 regional gatherings (Institut Pasteur 2019). This effort was praised for the scope of participation but criticized for lack of representativeness (Council of Europe 2019) and (according to interviews we conducted) the government’s ultimate rejection of most recommendations.

PE can be seen as promoting equity, raising the intertwined questions of what equity means and who should define it. Some define an equitable policy as one that is the product of a democratic and inclusive process, rejecting as inequitable a policy that emerges from a process dominated by influential groups (Dryzek et al. 2020). This approach puts a heavy burden on achieving and defining inclusiveness. Another approach posits an a priori definition of equity and asks the public’s advice on how to achieve it. This approach diminishes the public’s role, since the organizers of the PE control the definition of equity. Nonetheless, as we demonstrate in this paper, this seems closest to most human genome editing PE efforts to date.

This paper explores significant gaps between the conceptual framework and practice of PE in human gene editing research and governance. We compare public engagement-related recommendations of various groups focused on gene editing governance with the actual practices of those groups. Using data from our interviews and conference observations, as well as the public statements and reports of these groups, we compare PE’s promise and reality, including its capacity to enhance equity.

Project Overview

We are conducting a longitudinal empirical study of the emerging governance of human genome editing research. Governance refers to both formal regulation by national governments and international authorities and activities by non-state actors such as scientific and medical organizations, patient groups, funders, universities and other nonprofit research centers, journal editors, civil society participants, and private industry (Conley et al. 2020). By developing a deeper understanding of governance, we aim to propose policy that is both conceptually sound and grounded in reality.

PE has emerged has a central topic in our study because the people we are interviewing and the organizations we are studying have focused our attention on it. Calls for PE, as well as actual initiatives, have come from stakeholders of every sort who are intent on creating more just and responsive governance. This paper reflects our effort to understand these calls and the actions that they have produced.

Our research team is multidisciplinary, including specialists in bioethics, public policy, philosophy, law, sociology, anthropology, and genetics. Our methods reflect this diversity. We have surveyed genome scientists and other stakeholders (Waltz et al. 2021; Cadigan et al. Forthcoming); and are interviewing an internationally representative group of scientists, policy makers, and advocates; observing conferences (including the 2018 International Summit on Human Genome Editing in Hong Kong where He Jiankui announced his experiments); and tracking, observing, and in some cases participating in the activities of groups engaged in governance advocacy, as well as analyzing their reports. We draw on these data sources in this analysis of the conceptualization and practice of PE. We begin by reviewing data gathered on five groups that have been notably active in discussions of human gene editing governance, including PE.

Five Governance Groups

In this section, we describe and characterize both the calls for PE that the five governance groups have made and the processes by which they have developed their respective goals. We pay particular attention to whether the groups have themselves employed PE methods, and if so, which kinds.

The descriptions below are derived from the respective organizations’ meetings, websites, and publications and our interviews with multiple members of each group. We compare the PE policies that each group recommends with its own practices in gathering information and soliciting views. We pay particular attention to how the group structures its outreach and engagement efforts, in terms of who it treats as the relevant public[s], how they are allowed to participate, and what form that participation takes. The groups’ internal governance and PE practices reflect a variety of approaches, from expert-driven, top-down models dominated by scientists, policy experts, and civil society groups to citizen deliberation-driven, bottom-up models, featuring bidirectional consultation with local citizen groups, as well as hybrid models that combine elements of the other approaches.

ARRIGE (Association for Responsible Research and Innovation in Genome Editing).

Founded in 2018 at a meeting in Paris (which was itself an offshoot of an international initiative by France’s National Institute for Health and Medical Research [INSERM] Ethics Committee), ARRIGE is a private, voluntary organization that seeks to provide “a comprehensive setting for all stakeholders—academics, researchers, clinicians, public institutions, private companies, patient organizations, and other nongovernmental organizations, regulators, citizens, media, governmental agencies, and decision makers from all continents” (Montoliou et al. 2018). Its major objectives include “developing a robust and specific reflection on the role of the lay public in this debate and the necessity for improved public engagement,” with a particular focus on the Global South and lower-income countries generally.

ARRIGE’s statements indicate a commitment to wide-reaching PE, which would suggest a commitment to citizen-driven practices. The first publication by its founders—distinguished members of the French science and bioethics establishment—noted their intent to foster a global debate that would achieve unique breadth and depth. ARRIGE’s 2019 and 2020 (virtual) annual meetings included a breakout group on PE, chaired by a bioethicist, and sessions with representatives of patient groups. A 2021 seminar on gene drives featured a roundtable on social and ethical implications with several representatives of major policy groups. Since its founding, ARRIGE has issued two statements, one on gene drives and one on gene editing for crop breeding. The gene drive statement strongly advocates for community and stakeholder engagement. While ARRIGE’s activities have been limited by the pandemic, its practices—in contrast to its aspirations--have thus far been expert-driven, with some participation by civil society. The participants seem limited to experts and civil society representatives who are invited or self-identify and volunteer to take part in the organization’s committees and events.

National Academies of Science, Engineering and Medicine (US)-Royal Society (UK), International Commission on the Clinical Use of Human Germline Genome Editing (International Commission).

The International Commission was established following—and in reaction to--He Jiankui’s announcement at the Second International Summit on Human Genome Editing in 2018. It produced the 2020 report, Heritable Human Genome Editing (International Commission 2020). The process leading to the report included two in-person Commission meetings supplemented by virtual meetings and a “Call for Evidence” from the public in the form of written responses to 12 questions posed by the Commission. Answers were gathered in a public access file available upon request. In its report, the International Commission recommended that heritable human genome editing (HHGE) should not proceed in any country without public engagement concerning its acceptability (International Commission 2020, 28). While disclaiming any mandate concerning how PE should proceed, the report set out five “societal considerations to inform future discussions of HHGE” (International Commission 2020, 147). These include engagement with “people who have conditions that might be considered for HHGE,” recognition of “historical experiences with stigmatization and eugenic practices concerning disease and disability,” engagement with “other communities whose voices have not always been considered in medical decisions, including minority and indigenous communities,” and “discourse among civil society” (International Commission 2020, 147–48).

The International Commission’s work reflects a hybrid approach, expert-driven with some opportunity for civil society influence. Despite calling for a range of PE, its own information-gathering process was limited in practice to people and groups with the interest, scientific knowledge, and capacity to participate. Participation required voluntary effort by those people and groups and consisted of the opportunity to submit written comments.

The World Health Organization (WHO) Expert Advisory Committee on Developing Global Standards for Governance and Oversight of Human Genome Editing.

After a three-year process---also triggered by the He Jiankiu disclosure--this WHO Committee produced two publications: a Framework for Governance (2021a) and a set of Recommendations (2021b) (the Committee’s Position Paper (2021c) summarizes both publications). The Committee describes the Framework as “develop[ing] a governance framework that draws from good practices in the governance of emerging technologies and appl[ying] them specifically to human genome editing” (WHO 2021c). The Framework is general: it “can be implemented in different contexts [and] is intended to help those tasked with strengthening oversight measures, regardless of whether this is at the institutional, national, regional or international level” (WHO 2021a). It also argues (that those responsible for genome editing governance “must make efforts to educate, engage and empower many publics” and offers specific “best practices” for doing so (Ibid., 15, emphasis added).

The Committee’s process was largely dependent on meetings and consultations with experts and civil society representatives. It also conducted two month-long online consultations open to all. These consultations yielded 394 unique responses from individuals and organizations, mostly from those in the U.S., though 32 countries were represented.

The WHO Committee’s own practices reflect a hybrid approach. In design and execution, the process leading to the Recommendations was top-down, directed by the Committee members and staff. Scientific experts and civil society leaders dominated the consultation process. Yet broader engagement was solicited and, up to a point, achieved. Uniquely among the five groups, members of the Committee whom we interviewed were divided about the inclusiveness and efficacy of this process, with views ranging from active approval to acceptance of an unavoidably flawed process to criticism. The critical views may reflect the inherent limits of a process run by a large, bureaucratic, quasi-governmental body with a broad global mandate and hard time limits, all exacerbated by the pandemic.

European Group on Ethics in Science and New Technologies (EGE).

The EGE is an expert board providing independent advice to the European Commission (EC), the executive branch of the EU, and EC’s president. With representatives from law, philosophy, ethics, and the natural and social sciences, the EGE advises the EC on “all aspects of EU legislation and policies, where ethical, societal and fundamental rights issues intersect with the development of science and new technologies” (EGE 2022). The EGE’s Opinion on the Ethics of Gene Editing (2021), requested by the EC in 2018, offers recommendations similar to those of the WHO Framework. It stresses the value of PE as part of an “inclusive societal debate . . . constantly striving for global consensus” (EGE 2021, 87). It advocates “innovative formats for public deliberation” that include education and citizen assemblies. Such deliberation “should be open to everyone, involve a wide variety of stakeholders and forms of expertise and be inclusive.” (ibid.)

In developing the Opinion, the EGE conducted hearings and consultations with a range of experts and stakeholders, an Open Round Table, and an International Dialogue on Bioethics and Ethics of Science and New Technologies. While “these events emphasized awareness raising and the importance of structural public engagement on genome editing and on all its cognate ethical questions” (ibid., 106), there is no evidence that they involved the general public. The EGE’s process was thus similar to the International Commission with respect to PE, and perhaps even more expert-driven.

Global Citizens’ Assembly on Genome Editing (GCA).

This distinct group, founded and currently based in Australia, offers a model of deliberative consultation with representative groups of citizens. According to our interviews, the organization has two bases of inspiration, one practical and one theoretical. On a practical level, immediately following the 2018 He Jiankui announcement, the Australian documentary filmmaker Sonya Pemberton called for a citizen-driven initiative to prevent the kind of regulatory failure that allowed the He experiment. She obtained an Australian government grant to create a documentary about that initiative, currently titled Mutant. She then met with John Dryzek of the Centre for Deliberative Democracy and Global Governance, a longtime leader in the theory of deliberative democracy. Dryzek was first author of a Science article in which he, Pemberton, and others prominent in the genome editing governance debate laid out the theoretical case for a program of local, national, and ultimately international citizens’ assemblies (Dryzek et al. 2020). Carefully selected citizens would meet, hear from experts, and then—guided by facilitators—discuss and debate recommendations to governments and other authorities.

Like ARRIGE, the GCA was founded by prominent academics and policy experts—but with much of the impetus provided by a unique stakeholder, filmmaker Pemberton. Also like ARRIGE, GCA espouses a radical purpose: to promote bottom-up, citizen-driven PE. From its inception, GCA has decried “the pattern of self-appointed representation from global civil society (dominated by NGOs from wealthy countries)” (Dryzek et al. 2020, 1437) on the grounds that “civil society is not the same as the citizenry” (ibid., 1435). Although some authors of the 2020 Science article are also ARRIGE leaders, the article criticizes ARRIGE (ibid.):

ARRIGE has only got as far as a website, open meetings, and international expert meetings. Others have called for monitoring and learning that would involve experts, scholars, policy-makers, and organizations. However important they may be, such initiatives do not fully address the practicalities and specifics that would enable meaningful participation of citizens from around the world.

The GCA has delivered on its critique, at least up to the limits of its funding. Seeking to engage the public at large, it conducted a 23-member pilot assembly of Australian citizens in June 2021. Moving to the international stage is contingent on securing additional funding, and, according to GCA leaders we interviewed, that effort has been unsuccessful to date.

Defining and recruiting the relevant public is complex. The GCA’s interim report (Nicol et al. 2022) describes the citizen recruitment process:

The recruitment of participants involved the development of a screening survey capable of identifying the position of potential recruits (Niemeyer 2010) and randomly selecting from a pool of 132 volunteers to achieve “discursive representation” in the distribution of participants across the position (i.e. “discourse”) map.

Thus, although the GCA is bottom-up in its exclusive focus on deliberation by representative citizens, deciding who is representative relies on a selection process designed and executed by experts.

The GCA interim report includes principles and recommendations developed by the citizens’ assembly participants. They are strikingly similar to those arrived at by the very different WHO and EGE processes. The report also provides information about what the participants took away from the process. The organizers identify six policy positions held by the 23 citizens. The report identifies “significant shifts in position among individuals,” but “no overwhelmingly strong trend” in the direction of these shifts, although participants tended to move toward a more favorable view of genome editing (Nicol et al. 2022, 31–32). A post-deliberation survey found that participants had a favorable reaction to the experience, the format, the experts and facilitators, and their fellow participants. However, half said that the information they were given was not fair and balanced between different viewpoints, and slightly more than half thought that the small discussion groups were dominated by one or more people, making it difficult for others to express their views. These are potentially serious flaws in the small-group deliberation model.

Analysis

In this section, we discuss how the five different groups define “the public,” which types of PE each recommends, and what each sees as the goals and potential benefits of PE. We also consider the extent to which each of the five groups practices what it preaches, based on each group’s use of PE in developing its recommendations. These five groups reflect the immense practical challenges any governance group faces in defining the public and designing its methods of engagement. Contrary to their aspirational recommendations for the future, four of the five groups, in their own processes, have limited the publics they engage. Although it is reasonable for a group to identify ideals for others to follow in the future, the fact that, in their own processes, only one of the groups--the GCA--even approached its stated ideals is evidence of how elusive those ideals may be in practice.

A group’s definition of the relevant public[s] and its engagement methods seem inextricably intertwined. If the WHO, for example, had started by defining its relevant publics as primarily experts and civil society representatives, then its methods of engagement would have seemed obvious and adequate. Alternatively, had it begun by acknowledging that its engagement methods were logistically limited, then the publics that could be engaged would be equally limited and probably dominated by experts and representatives of organized groups. Either way, its PE would have looked the same, consisting largely of formal sessions with experts and civil society. Meanwhile, the philosophical foundations of the GCA entail mutually reinforcing commitments to defining the public in the broadest possible terms and engaging its representativeness in an intensive way.

The GCA’s initial deliberation event prompts questions that are simultaneously theoretical and pragmatic. Although only 23 people participated, this number is on the order of magnitude recommended by the citizen deliberation literature, for reasons of logistical manageability and group functionality (Thaldar et al. 2021; Dryzek et al. 2020). It is also the scale that the GCA has been able to afford. But how representative can 23 people be of a country as large and diverse as Australia? Perhaps very representative, but that is dependent on the validity of the expert-driven sampling model, which seeks both demographic and discursive representation (Niemeyer 2010). What would happen if an organization like the GCA were to obtain funding to scale up to an international assembly? Would it keep the groups small? That would exacerbate the problem of representativeness, since a few people would have to represent an entire society, further challenging the selection model. Alternatively, if the organizers expanded the groups significantly, would they reach a size where meaningful deliberation was impossible? Finally, the organizers might choose to do multiple small-group sessions, but that plan would face financial and logistical constraints.

A second category of questions emerging from the GCA experience involves the potential benefits of public deliberations. Despite the different method, the recommendations that the initial 23 Australian citizens arrived at mirror those of the WHO and EGE. There could be several explanations for this, which are not mutually exclusive: It may be that the big, expert-driven groups, even with their flawed PE, still get a good sense of public concerns and respect them. It may also be that there are some obvious principles at work and anyone, expert or lay, who is exposed to the issues is likely to discover and gravitate toward them. Or perhaps any PE process will reach consensus, in the sense of gravitating to the middle of the range of views. It is also possible that the PE process—especially the citizen assembly version—shapes the participants’ thinking—however unintentionally--through the preliminary education of the deliberating group. That preliminary education is, of course, conducted by the sorts of experts who dominate big-group PE efforts. While belief in the desirability and benefits of educating the public seems universal—illustrated by the WHO’s “educate, engage, and empower” mantra—a deeper question lingers: when you educate, do you inevitably indoctrinate? The GCA participants reported this problem, and also expressed concerns about dominating voices in their deliberations. A related if currently unanswerable question is what, if anything, the participants do afterwards, when they return to their regular worlds. For example, do they become advocates or expert commentators?

Conclusion

Public engagement for research is an established and well-studied practice in many public health and environmental science arenas. Where local communities have been empowered to act on behalf of a relevant public, PE has been able to influence the design, scope, and pace of research (Gusmano et al. 2021). In such cases, inclusive PE may advance equity by asserting community perspectives. On matters of science policy as anticipatory as the governance of human gene editing research, however, the evidence for what real-world PE can accomplish is uncertain. One focus of our investigation has been whether groups recommending PE as part of human genome editing governance have also employed PE themselves—and what it means that they generally have not. Beyond this, we have found little evidence that any of the PE initiatives we are following (or PE more generally) has affected genome editing policy making, especially in the sense of improving equity.

Finding the preferred public[s] is difficult. The GCA took a sophisticated sampling approach, with selection depending on how the organizers defined the relevant diversity axes and then distributed the sample. Money and logistics, including challenges of the pandemic, also imposed significant limits. Yet absent such efforts, the public will end up being those with the means, motive, and opportunity to hear about what is being done, and then come forward to express views. Scale seems to be an overwhelming issue. The citizen assembly literature calls for small groups (as in Australia) to promote true deliberation. While that puts enormous pressure on sampling decisions to ensure inclusive representation, significantly expanding the groups jeopardizes the quality of deliberation.

Equally vexing is how to engage with the identified public[s]. Small-scale citizen assemblies have happened. Furthermore, as noted above, small, relatively homogeneous countries have had some success directly engaging the broader public on bioethics issues. But in a large-country context, the French bioethics review shows mixed results. Other approaches to PE seem to be dominated by organized voices and interests, and activists of various sorts. Even if the definitional and logistical problems inherent to PE can be managed, it is unclear whether the debate about PE, and governance more generally, makes any difference. In fact, the most interesting piece of data may be that citizen assembly participants have developed recommendations similar to major policy groups like the WHO.

Science tends to get done. Governments at national levels do regulate, but they play catchup, often not doing anything until science has moved on to the next thing. PE may have many benefits, but can advancing equity or influencing policy be legitimately claimed among them?