Data Sharing During Pandemics: Reciprocity, Solidarity, and Limits to Obligations

As noted by Pratt and Bull in their recent literature review, maximizing utility is the primary driver behind calls for data sharing in health research and outbreak response (Pratt and Bull, 2021). In the context of an infectious disease outbreak or pandemic, one could argue that there’s an obligation to share data for surveillance and research purposes given (a) the severity of the situation, (b) the time constraints associated with responding to pandemics, and (c) because the knowledge gained usually cannot be ascertained at all or at a very high cost if data isn’t shared between jurisdictions and states. However, solely focusing on utility leaves us vulnerable to problems like the possibility of exploitation and the inequitable distribution of the benefits and burdens associated with data sharing (Pratt and Bull 2021).

Solidarity is another value commonly invoked to justify the obligations of states to share data during outbreaks and pandemics because it acknowledges utility while taking seriously questions of equity and fairness. In a paper we co-wrote with Langat et al. in 2011, we argued that “the health research community, as a community, is and ought to be regarded as united under a goal of combating disease” and that “embracing the value of solidarity highlights our common interests and shared vulnerabilities, providing additional impetus for collaboration in protecting human health and promoting human well-being” (Langat et al. 2011, 9). In short, we used solidarity as an ideal value to help justify an obligation to share data during pandemics. But note that we invoked shared humanity, common interests, and shared vulnerabilities; we wanted to promote data sharing for everyone’s benefit, thereby striving for a balance between utility (i.e., advancing the world’s capacity to understand and respond to infectious diseases) and equity (i.e., ensuring everyone benefits from the world’s capacity to understand and respond to infectious diseases). Solidarity is similarly deployed by the WHO in their ethics guidance documents in relation to both pandemic planning (World Health Organization 2007), as well as in response to COVID-19 (World Health Organization 2020). Meanwhile, McLennan and colleagues use the value of solidarity to argue that the European General Data Protection Regulation (GDPR) exemption should be invoked to facilitate data sharing during COVID-19 (McLennan, Anthony Celi, and Buyx 2020).

In turn, reciprocity is often used narrowly in the context of data sharing as a reason to support and protect individual or teams of researchers in sharing data, since they are in a position of risk when sharing. Research incentivizes secrecy—e.g., intellectual property to monetize products, the notion of “publish or perish,” etc.—so authors (again, including us) have previously deployed reciprocity as a means of articulating the need to think creatively about how to compensate and help those individuals who risk enjoying the full fruit of their research for the shared common interest of global health. However, writing in the context of the Ebola outbreak, the U.S. Presidential Commission for the Study of Bioethical Issues wrote that “reciprocity indicates that individuals and communities who have shared their biospecimens should have access to benefits that result from such research …. [especially] when dealing with samples collected from potentially vulnerable populations, including populations that are potentially vulnerable socioeconomically … ” (The Presidential Commission for the Study of Bioethical Issues, 2015, 47). Thus, the Presidential Commission correctly suggests that the risk of not benefiting from data sharing are not only borne by scientists but by their communities, too.

What public health ethicists (including us) have largely previously failed to imagine is not that those in positions of power would fail to share the benefits of data sharing with marginalized groups or states, i.e., being wronged via an omission, which was foreseen and is rather common. Instead, we failed to imagine the commissive prejudicial actions on the part of HICs and those in positions of power toward LMICs and marginalized communities as a consequence of data sharing with punitive results. It is particularly vile to take data shared by South Africa and use it to hinder it and other southern African countries. What makes the latter particularly morally problematic concerns the intention (or lack thereof) to abide by previously agreed to international rules, e.g., the International Health Regulations (World Health Organization 2016). HICs expect LMICs to abide by previously agreed to rules regarding pandemic response but repeatedly disregard whichever rules are inconvenient for them. Whatever other reasons exist to articulate why rules are morally important, it matters that following rules provides predictability and security that certain actions will follow in future instances should certain situations eventuate. Breaches of rules does away with predictability and the ability of agents to plan.

Given the realities of global health, namely, a backdrop of gross inequality and the fact that even during moments of extreme crises HICs appear incapable of acting fairly toward LMICs and global institutions, the use of solidarity increasingly rings hollow. The invocation of solidarity to morally justify data sharing assumes that all or most actors will acknowledge humanity’s shared vulnerabilities and common interests and then act on that basis. This assumption is false. In a non-ideal world that requires non-ideal norms and reasoning, solidarity is still valuable as lighthouse guiding ships in the fog. We should strive toward solidarity in global health, particularly as it relates to data sharing, but it largely fails to straightforwardly articulate the moral demands for data sharing, especially on the part of LMICs and others who are potentially marginalized.

Instead, data sharing should be guided by the principle of reciprocity. If reciprocity is about providing good for good received, then the onus is on HICs to demonstrate how (a) they will share the benefits of data sharing, and (b) will not take any actions that make LMICs worse off when they do share data. Data and samples are valuable; as such, it is morally permissible for LMICs to guard their dissemination and use unless and until politicians and governments in HICs can prove to them that they’re capable of playing nicely. No reasonable moral theory dictates that people must make themselves worse off when others have demonstrated repeatedly they are unwilling to abide by prearranged rules. Reciprocity has conditionality built into as a principle, whereas solidarity does not necessarily, which in the case of actors acting in bad faith, should be our moral starting point.

One could object that this will make addressing future pandemics more difficult with a quid pro quo attitude about data sharing, thereby reducing the utility associated with the practice. We would hope we could proceed without such arrangements, that solidarity would prevail and HICs would act accordingly. Our point is that there is little reason to believe this will be the case. Consequently, we may need to proceed on a quid pro quo basis.⁵ This is arguably going to make future pandemic responses more difficult for HICs. But it may make things more equitable. And efforts to advance equity may often require increased burdens on those in positions of power. On a simplistic understanding of utility and utilitarianism, we’ll bite the conceptual and normative bullet for the sake of equity. However, we’d argue that a more nuanced articulation of utilitarianism could still potentially make room for our conclusion, i.e., premising data sharing on the basis of reciprocity may have long-term overall better outcomes, while avoiding the potential deterrent effect of other countries hiding information given what happened to South Africa (Mallapaty 2021).