Background

Ownership of research outputs has been identified as an important practical, legal, and ethical issue for many citizen science projects (Dratwa 2015; Fiske, Prainsack, and Buyx 2019; Guerrini et al. 2018; Resnik, Elliott, and Miller 2015; Riesch and Potter 2014; Scassa and Chung 2015a; Vayena and Tasioulas 2015). Whereas project inputs include contributions to a project, such as data and biospecimens, outputs are the tangible and intangible ideas and things that result from research processes. Depending on the design and objectives of a project, its outputs might include data sets, research findings, publications, or discoveries of new methods, products, or technologies (Figure 1).

Although citizen scientists might not have property rights to project outputs, factors internal and external to a project can promote the idea that they have valid claims to access, control, or share in the benefits of the project’s outputs. This second, more expansive, meaning of ownership claims—as claims grounded in ethical principles, whether or not recognized as rights secured by law (Guerrini et al. 2019)—is the focus of this article. More specifically, and in recognition of the fact that some projects are led by or in partnership with citizen scientists, it is concerned with the ownership claims of citizen scientists who are not project leaders. Further, this article is focused on ownership of project outputs, not inputs. Nevertheless, literature and legal precedent considering the propertization and commercialization of personal research inputs (Roberts 2017; Roberts, Pereira, and McGuire 2017; Spector-Bagdady et al. 2018) are relevant to the extent that persistent rights in inputs influence moral claims to ownership of outputs.

Project practices that facilitate citizen scientists’ exclusive or non-exclusive access to, control of, or share in the benefits of research outputs are referred to as ownership practices. Examples of ownership practices include providing citizen scientists access to processed individual-level data and data sets; coauthoring publications with citizen scientists; sharing with citizen scientists the profits from commercial transactions and royalties from published works; inviting citizen scientists to co-present results and sharing honoraria from speaking engagements; recognizing or securing legal rights of citizen scientists to intellectual property (IP); and facilitating citizen scientists’ access to commercialized products or technologies that result from their work. A project might, but need not, document or formalize these practices in public statements, policies, or agreements.

Unlike citizen science projects conducted in other disciplines, such as ecology and space science, biomedical citizen science projects sometimes include features that might be especially likely to stir moral claims of ownership. Although biomedical citizen science projects are diverse (Table 1), a common feature is their invitation for participants to contribute personal health data or biospecimens for analysis (ECSA 2020; Fiske, Prainsack, and Buyx 2019; Wiggins and Wilbanks 2019). Depending on the circumstances, the intimate nature of these inputs can prompt feelings of ownership over them that extend to the outputs that result from analysis and use of those inputs. Genetic and genomic data, including data generated from single nucleotide polymorphism (SNP) microarrays and whole genome and exome sequencing, are arguably some of the most personal kinds of biomedical data that a person can share given that they can be predictive of individual health, behavior, and appearance. Separately, participants might have heightened feelings of entitlement to access or use biomedical citizen science research outputs when projects focus on medical conditions that affect them or their loved ones and they stand to benefit from, or might be harmed by, those outputs.

Empirical studies have described citizen science project practices related to accessing, controlling, and sharing in the benefits of outputs, with a focus on data sets, and probed relevant perspectives of citizen science practitioners, participants, and scholars (Borda, Gray, and Fu 2019; Bowser et al. 2020; Guerrini et al. 2019; Guerrini et al. 2022). Potential legal claims of citizen scientists to project outputs also have been examined (Guerrini and Contreras 2020; Scassa and Chung 2015a,b).

In these and other publications, ethical management of ownership is widely recognized as an important consideration for citizen science projects. For example, the European Citizen Science Association (2020) describes the practice of informing citizen scientists how their data are used (such as by sharing outputs with citizen scientists who contributed to them) as an ethical responsibility of projects. Focusing on IP, Scassa and Chung (2015a, p. 5) explain: “IP management in citizen science has ethical dimensions because IP rights regulate a series of relationships between individuals and in relation to intangible goods.” Especially when collaborators are in a hierarchical relationship, IP management “must be carried out ethically to sustain trust” between them (Scassa and Chung 2015a, p. 5). In the context of ownership of results of participatory health research and development, an advisory body of the President of the European Commission makes an explicit appeal to fairness in asking whether “greater reflection [should] be paid to allowing citizens/patients to share in the advantages of their contributions” when later monetized by others (Dratwa 2015, p. 60).

To assist in management of research outputs, scholars have identified ethical issues to consider when making decisions related to access to data and data sets (Cooper, Rasmussen, and Jones 2021) and sharing benefits of commercial exploitation, including royalties in inventions, with participant communities and individual citizen scientists (Scassa and Chung 2015a). They have also provided useful suggestions for management of ownership that include negotiating instruments “that recognize the interests of all stakeholders” and discussing ownership and IP with citizen scientists at the beginning of and as needed during the project period to promote understanding (Resnik, Elliott, and Miller 2015, p. 478).

Building on this literature, we propose a framework to help biomedical citizen science project leaders systematically evaluate the ethical soundness of ownership practices they are considering or have implemented. We begin by describing the traditional approach to ethical review of scientific research, including its ownership-relevant plans. We then explain why the traditional approach does not always translate well to scientific activities designed as citizen science, necessitating consideration of a different approach to ethical review of their ownership practices. In an effort to meet that challenge, we propose a framework for evaluating those practices that centers on four considerations: reciprocal treatment, relative treatment, risk-benefit assessment, and reasonable expectations. Although this framework might be useful to projects unrelated to health or medicine, we focus its application on biomedical projects given the special issues they raise related to ownership, as described above. Accordingly, we anchor the discussion with a hypothetical project relevant to human genetics and genomics, a popular focus of self-inquiry (White 2019). Ultimately, the intention of the proposed “4R” framework is to help citizen science projects evaluate the ethical defensibility of their ownership practices.

A Hypothetical

Consider a hypothetical citizen science project called PersonalPRS. The PRS in the project’s title refers to polygenic risk score, which is an estimate of a person’s risk for a disease based on the combination of their genetic variants (Torkamani, Wineinger, and Topol 2018). The aims of the project, run by a small group of United States (US)–based Ph.D. students during their free time, are twofold: (1) most immediately, to generate PRS for common diseases for individuals from their SNP files, and (2), for specific diseases, to identify environmental exposures common among individuals who have the disease but low PRS for it.

Individuals can contribute to the project in several ways. First, they are invited to upload their SNP files, previously obtained from direct-to-consumer (DTC) genetic testing companies, and complete a brief questionnaire about their health history. The crowdsourced data are used to improve PRS construction methods. In return, data contributors are provided their PRS for common diseases. Second, in an online forum, individuals (including but not limited to data contributors) are invited to suggest modifications to the project’s PRS construction methods and calculations, which are published on the platform. Elsewhere in that forum, third, they are invited to discuss their PRS results and suggest improvements to the platform and its communication of results, some of which the project leaders have implemented. The leaders have invited the most active participants—called SuPRStars—to serve as moderators of the forum. SuPRStars are in regular, direct contact (by email and video conference) with the project leaders.

With respect to future plans, the project leaders hope to roll out disease-specific challenges in which self-selecting participants will work to identify environmental exposures common among individuals who have a specific disease but low PRS for it. Specifically, participants will partner with project leaders to develop supplemental questionnaires that will be sent to these individuals and analyze deidentified SNP files linked to deidentified questionnaire responses contributed by participants who have specifically opted into this research use of their data.

The PersonalPRS website describes the project as “citizen science” and invites members of the public to “join the team” as “partners.” The website also includes a disclaimer that the platform is provided for educational purposes only; is not intended to be used for any diagnostic, treatment, or preventative purpose; and is not a substitute for professional medical advice.

Approximately 75,000 individuals around the world have uploaded their SNP files to PersonalPRS and completed health questionnaires. Recently, the leaders were approached by a DTC genetic testing company interested in purchasing the PersonalPRS platform and participants’ deidentified SNP files linked to their deidentified questionnaire responses.

The Traditional Research Ethics Framework and Application to Ownership Practices

There are several ways that PersonalPRS participants can engage with the platform, with project leaders, and with each other. Each of these interactions generates distinct research outputs. What criteria should be used to evaluate the ethical acceptability of approaches to manage claims to access, control, and share in the benefits of those outputs?

According to a traditional research ethics analysis, this question would be answered primarily by reference to the principles identified in the Belmont Report, which is the basis of ethical oversight of biomedical research provided by Institutional Review Boards (IRBs) in the US (National Commission 1979). Briefly, there are two sets of federal regulations that require IRB review of US-based biomedical research. First, the Federal Policy for the Protection of Human Research Subjects, otherwise known as the Common Rule, applies to all non-exempt research involving human subjects conducted or funded by the US federal government (45 C.F.R. § 46). Like many citizen science projects, PersonalPRS is probably not subject to the Common Rule’s IRB requirements because its activities are not conducted or funded by the federal government and, to the extent that any of the leaders’ institutions apply the Common Rule to all human subjects research in which they are engaged as a matter of institutional policy (Meyer 2020), the institutions do not appear to have sufficient nexus to project activities to be within their IRBs’ jurisdiction.

The US Food and Drug Administration (FDA) requires IRB review of certain clinical investigations involving FDA-regulated medical devices, including software that functions as a medical device, and these rules largely mirror the Common Rule’s procedural and substantive details (21 C.F.R. §§ 50, 56). There is a question whether any of PersonalPRS’s activities qualify as a clinical investigation requiring IRB review under FDA rules, but that question is moot if PersonalPRS does not qualify as an FDA-regulated medical device. The FDA has generally ignored third-party genetic interpretation platforms (Greely 2020), many of which include disclaimers similar to those published on the PersonalPRS platform, suggesting that the agency might view them as the kind of low-risk device over which it has stated it will exercise enforcement discretion even if subject to FDA regulations (FDA 2019a,b).

Still, some US states extend federal research subject protections to research over which it has jurisdiction, and many journals require IRB approval of research that they publish (Meyer 2020; Resnik 2019). For these or other reasons, some otherwise unregulated research, including citizen science projects, might seek out IRB review by commercial boards when they can afford it.

If an IRB reviewed the activities of PersonalPRS, whether such review is required or requested, the activities would be scrutinized using a lens focused on the three ethical principles outlined in the Belmont Report. First, respect for persons encompasses respect for their autonomy, or ability to make considered decisions for themselves (National Commission 1979). As operationalized by the federal rules (45 C.F.R. §§ 46.111(a)(4), 116; 21 C.F.R. §§ 50.20, 50.25, 56.111(a)(4)), demonstrating respect for research participants means obtaining their voluntary and informed consent to participate in research based on a clear articulation of the research aims and procedures, its anticipated risks and benefits, and alternatives to participation. Second, beneficence is an obligation to secure the well-being of research participants through attention to the risks and benefits of research activities (National Commission 1979). Risks and benefits can be physical, psychological, social, economic, or legal in nature. According to the federal rules, and representing a departure from the Belmont Report (Meyer 2020), risks must be minimized and reasonable in relation to benefits to participants and knowledge to be gained, but benefits need not be maximized (45 C.F.R. § 46.111(a)(1)-(2); 21 C.F.R. § 56.111(a)(1)-(2)). Third, justice in traditional research ethics is focused on fair distribution of the benefits and burdens of research, which can be evaluated according to formulations such as “to each person an equal share” or “to each person according to individual effort” (National Commission 1979). Both the Belmont Report and federal rules operationalize justice as fair selection of participants, including consideration of whether they belong to groups who are likely to benefit from research outputs, such as improved treatments (45 C.F.R. § 46.111(a)(3); 21 C.F.R. § 56.111(a)(3); National Commission 1979).

Neither the Common Rule nor FDA regulations require that researchers adopt practices to facilitate research participants’ access to, control of, or share in the benefits of research outputs, including data, findings, treatments, or technologies. Rather, the rules focus on making relevant plans transparent to research participants. Thus, in certain circumstances, the regulations require that prospective participants be informed if “clinically relevant research results, including individual research results, will be disclosed” to them (45 C.F.R. § 46.116(c)(8)). (This is distinct from the statement, required in certain circumstances, that “significant new findings” will be provided to participants that might relate to their willingness to continue participation (45 C.F.R. § 46.116(c)(5); 21 C.F.R. § 50.25(b)(5)).) Further, prospective participants must be informed if their biospecimens “may be used for commercial profit” and whether they will share in those profits (45 C.F.R. § 46.116(c)(7)). According to the federal agencies that have adopted these rules, informing prospective participants how their inputs will be used promotes their autonomy (HHS 2017).

Yet, in the absence of regulatory obligations to do so, some researchers have voluntarily adopted practices that recognize participants’ interests in the outputs of the work they support. For example, some have adopted plans to return individual-level data, such as raw genetic data files, to participants who indicated they would like this information (Thorogood et al. 2018). (This practice is distinct from the federal obligation of covered entities, including some researchers, to provide patients their protected health information on request (45 C.F.R. § 164.524).) These plans are justified on grounds that they promote autonomy and beneficence with respect to participants who are interested in knowing this information and potentially using it for their own benefit or sharing it to benefit others. However, as one scholar has noted, the Common Rule “clearly perceives return of individual results to be a source of potential harm to participants” (Meyer 2020, p. 68). For this reason, IRBs might scrutinize plans to return individual results to ensure that their perceived risks— for example, that participants could misunderstand the results and change their behaviors in potentially harmful ways based on the misunderstanding—are minimized. Ultimately, IRBs can decline to approve such plans.

But citizen scientists have a different role in the research process than participants in traditional biomedical research, prompting some scholars to examine whether conventional ethical frameworks are appropriate and sufficient for biomedical citizen science (Kasperowski, Hagen, and Rohden 2021; Rasmussen 2021; Vayena and Tasioulas 2013). Specifically, citizen scientists make research contributions other than or in addition to serving as a source of specimens and data for analysis by others and, as demonstrated by PersonalPRS, include potential contributions at many stages of the research process (ECSA 2015; Fiske et al. 2019; Wiggins and Wilbanks 2019). Consistent with these diverse and potentially robust participatory opportunities, some citizen science projects are structured to share power and decision-making with participants (Rasmussen 2019). Such projects might describe their objectives and the role of citizen scientists in relevant activities using a rhetoric of engagement and democratization (Fiske, Prainsack, and Buyx 2019) that could be inappropriate in communications about traditional biomedical research (Woolley et al. 2016).

Analysis of the ethical defensibility of ownership practices of biomedical citizen science projects requires an approach that takes into account these departures from traditional research. For example, depending on the project, respect for persons might require ongoing transparency about elements of projects that are not usually shared with traditional research participants, such as reporting data and findings in ways that will allow citizen scientists to “make meaning” of them (Cooper, Rasmussen, and Jones 2021, p. 2). Further, citizen scientists are sometimes in need of protection against different kinds of harms than traditional research participants (Rasmussen 2021). Some protections, such as prohibitions on the return of interpretations of contributed data, might be viewed by citizen scientists as paternalistic. They might also be confusing and disappointing given that, as noted by Wolf (2020, p. 163), return of results is a “core element[]” of participant-driven forms of citizen science. As a final example, justice concerns shift to providing fair opportunity to participate and minimizing barriers to participation, consistent with citizen science’s focus on inclusivity (Wiggins and Wilbanks 2019), and also encompasses fair distribution of opportunities to access, control, or share in the benefits of research outputs.

The 4R Framework

For all of these reasons, citizen science projects are advised to give careful attention to ownership practices. Yet, there does not exist practical guidance for project leaders or external reviewers, such as IRBs, community ethics committees, or ethics consultants, regarding how to systematically evaluate the ethical soundness of practices that citizen science projects are considering or have adopted.

To address this gap, we propose a framework for evaluating ownership practices in biomedical citizen science comprising four considerations: reciprocal treatment, relative treatment, risk-benefit assessment, and reasonable expectations (Table 2). As described below, the considerations are rooted in traditional research ethics principles of respect for persons, beneficence, and justice, but consistent with the literature, their application pays special attention to the emphasis in citizen science on participatory experience, inclusivity, balance of power, and trust (Chesser, Porter, and Tuckett 2020; Fiske, Prainsack, and Buyx 2019; Groot and Abma 2022; Rasmussen 2019, 2021; Wiggins and Wilbanks 2019).

Importantly, the framework describes a standardized process for evaluating the ethical soundness of ownership practices. It does not describe a formula that will generate an absolute truth as to the rightness (or not) of any particular practice. However, by providing a guide for conducting these evaluations, the framework can help ensure that they are comprehensive and illuminate ways that a project might modify its practices to improve their ethical defensibility.

The intended users of the framework are project leaders and external reviewers, although our description focuses on evaluation by project leaders given that external review can be difficult to secure. For some projects, leaders might be difficult to identify; governance might be distributed among all or many participants. Usually, however, the leaders are a subset of the project’s institution-based scientists and/or citizen scientists, or if the project does not involve institution-based scientists, they are a subset of the project’s citizen scientists. We return to questions of responsibility and accountability for evaluation in the Discussion.

Discussion

Our proposed framework is intended to be a guide for ethical evaluation of biomedical citizen science ownership practices. It was constructed at a level of generality to potentially be useful to projects regardless of size, setting, objective, or design. In its application, however, context matters. As observed by Cooper and colleagues in their helpful explication of ethical issues relevant to data governance decisions in citizen science, “what arises as an ethical issue and appropriate solution in one project might not in another almost-identical project.” (Cooper, Rasmussen, and Jones 2021, p. 4). Here, also, careful attention must be paid to the details not only of the ownership practice under consideration but how it has or will be operationalized by a project, taking into account its specific features and scientific, organizational, and sociocultural circumstances.

The framework is systematic because it requires examination of each of the four considerations. It is comprehensive because it calls for examination of the various features and implications of an ownership practice from multiple perspectives. But the framework does not purport to be exhaustive. Rather, it leaves open the possibility that, in a particular case, other ethical considerations might require attention. As one example, it may be necessary to consider whether the ownership practice creates, exacerbates, or perpetuates inequities beyond the project that is considering adopting it.

One limitation of the framework’s utility in the biomedical citizen science space is that it will not be relevant to self-experimentation and other n-of-1 activities given that they do not usually present situations where there are multiple interests in or claims to outputs to resolve. Of course, this limitation assumes that such activities qualify as citizen science in the first place. We believe that they can, even if those engaged in n-of-1 activities rarely describe their work as citizen science (Wiggins and Wilbanks 2019).

A different question, raised earlier, is procedural: who is responsible for the analysis and its outcome? Asked differently, whose judgments count? External reviewers knowledgeable about citizen science are well suited for evaluating ownership practices because they are not personally invested in the outcome, but as explained above, external review is not always available to projects. In such cases, project leaders should conduct or oversee the evaluation. Consistent with the citizen science emphasis on fair treatment of participants, it seems reasonable to assume that project leaders will usually handle the evaluation with care. But to promote accountability in the event they do not, leaders should commit in advance to making their analysis available to non-leaders. Of course, some citizen science projects are conducted collaboratively by communities and might not have designated leaders (ECSA 2020). In such instances, all willing participants should be involved or, when this is not practical, representatives should be identified according to processes consistent with project values.

Even when leaders are recognized, they still might want to invite the input of non-leaders on analysis of the four considerations—and potentially others. Indeed, project leaders might feel compelled to do so if, for example, the project’s ownership practices were originally co-created with non-leaders. While soliciting broad input can be helpful and in some cases might be obligatory, leaders should resist any temptation to adopt analytical shortcuts or defaults that would judge a project’s ownership practice as ethically permissible so long as it is endorsed by the project’s non-leader citizen scientists. Our framework describes one way to help guard against unfair outcomes and otherwise increase confidence that judgments about the ethical soundness of ownership practices in biomedical citizen science are systematic, comprehensive, and take into account its singular features.