The COVID-19 pandemic is at the core of a triple crisis facing the U.S. population. The economic impact, both as a direct consequence of the pandemic and from the cost of accompanying mitigation measures, is the second element of the crisis; it has manifested in lingering high levels of unemployment, with some 26.8 million workers, almost 16 percent of the entire U.S. workforce, either unemployed, otherwise prevented from working by COVID-19, or employed but on reduced pay.

Linked to both are the exacerbating effects of the third element of the crisis: the civil unrest and protests linked to systemic racism. While these emerged last year in response to the killing of Black Americans at the hands of police, long-standing racial inequities have also resulted in the overrepresentation of minoritized groups among those exposed to COVID-19, those experiencing severe infections and deaths, and the ranks of the unemployed.

In the ongoing public debate around how best to bring COVID-19 under control, a series of solutions have been proposed and enacted. Many of these are defended, by their proponents, as efforts to “follow the science.” Indeed, phrases such as “follow the science” and “follow the evidence” became something of a rallying cry in the early months of the pandemic, particularly around the imperative for social distancing measures. While there is no question that an understanding of the science needs to be at the heart of any effort to contain a pandemic, we suggest that the adoption of science as categorical imperative overstates the role of science in the nuanced and fundamentally moral and political nature of decision-making, while also alleviating decision-makers from the responsibility for difficult moral and political choices.

We argue there are three key ways in which narratives of “following the science” are both conceptually flawed and counterproductive.

First, science is complex, incomplete and ongoing. During the pandemic, there have been examples of good science, but also poor and opportunistic science. Peer review, designed to catch our mistakes at the best of times, can suffer from in-group bias and in any case has been put aside to an unprecedented degree in the explosion of preprint research during the pandemic. In 2020, between January and June alone, scientific papers on COVID-19 doubled every 14 days, reaching over 100,000 separate articles. An analysis of the first 10,000 COVID-19 articles found that most failed to focus on the questions identified as key for pandemic prevention, and more than 60 percent were opinion pieces rather than original research.

As any experienced researcher working in the field will attest, it is becoming increasingly difficult to critically review this burgeoning literature. Setting these limitations aside, any given scientific paper, no matter how well it was designed, conducted, reported and reviewed, represents but one drop in the ocean that is the relevant evidence base, and so can rarely, if ever, be the tipping point on which policy might be hinged.

Finally, even when following the science seems clear, for example in the development of multiple safe and effective vaccines, science cannot make the value judgments (such as who gets the vaccine first) that are needed with regards to policy implementation. Geoffrey Rose, one of the fathers of modern epidemiology and public health, famously said that while the best science can help inform decisions, “in a democracy the ultimate responsibility for decisions on health policy should lie with the public.”

This is at the core of our second observation. Following the science suggests that decision-making is scientific in nature, but that is not how decision-making works, or indeed should work. Through defining and choosing which branches of science, or streams of evidence, to prioritize, politicians can under the appearance of science justify a wide variety of positions. More fundamentally, to claim to rely on science as the determining influence on policy, even during a pandemic, is to mischaracterize how science is conducted, how it is packaged and operationalized by democratically elected representatives, and the breadth and nuance of policy options available in response to any combination of evidence presented. This is why such decisions are inescapably, and rightly, political in nature, and why responsibility for them should rest with democratically elected leaders, who are answerable to the public. As the aphorism goes, “to govern is to choose.”

Third, and perhaps most centrally, a “follow the science” narrative masks a key alternative role science should play in informing ethical decision-making, namely that of helping to inform and communicate the potential trade-offs accompanying policy choices. Trade-offs are at the core of economic theory but are also well understood in bioethics and in public health, particularly in prioritizing between improvements in health equity and overall health improvements.

In the context of COVID-19, it has been argued that perhaps the most public example of trade-offs, the “health vs wealth” debate, is an example of a false dichotomy, and that the scientific consensus is clear on the ways to act to minimize both costs and health harms. However, this does not change the fact that informing our understanding of trade-offs in policy choices represents a core, yet neglected, part of what science brings to decision-making. Redressing this balance in ways that avoid harmful and false dichotomizing, could serve several critical functions in the context of COVID-19, particularly in terms of avoiding growing inequity.


To use the analogy of an individual physician advising a patient, the ability to predict and communicate potential side effects of a course of treatment is an important part of good care. It helps the patient understand the severity of each course of action, and, in consultation, allows for strategies to be discussed that might mitigate those side effects. Both the benefits of the treatment, and the side effects of that treatment, are equally part of “the science.” In extremis, identifying a minority of individuals who might suffer disproportionately from those side effects, through for example a medical allergy to certain medications, or poorer underlying health, a physician can identify alternatives better suited to those groups. This approach is empowering, transparent, just and arguably is a fuller expression of “following the science.”

At the opposite end of the scale, the need to cut carbon emissions is an example of a clear scientific consensus and global priority. There are however, potentially negative consequences of cutting emissions, for example through lost employment from changes in the nature of an economy that disproportionately affect those without assets or high levels of education (as sectors such tourism, agriculture or extractive industries adapt to emissions targets), or the necessary energy costs associated with governments “raising the floor” in terms of living standards and infrastructure for the most disadvantaged individuals, communities and countries. Understanding, communicating and ameliorating these effects is therefore well understood as a critical component of any such plan, be it the Green New Deal’s proposed investments in new jobs and training through an economy focused on renewable energy, or the Paris Agreement’s inclusion of technology transfer, capacity building and financial assistance schemes to assist countries in achieving their targets, and avoid their being unfairly disadvantaged.


Although scholars have written measured deliberations on the ethical pitfalls of the mitigating measures governments have taken during the pandemic and issued nuanced recommendations on how to manage the potential consequences of reopening, these arguments have, by and large, taken a back seat to far more emphatically stated simplifications around a suggested clear-cut scientific “right answer” that has dominated public discussion and largely informed decision-making around COVID-19.

We suggest that it is more important now than ever to pause and reflect on the contributions science can—and cannot—make to the pandemic moment. In spite of the advent of the first COVID-19 vaccines, it is clear that further difficult decisions must soon be made in the U.S. and around the world regarding social distancing and school closures. It is therefore imperative that a greater emphasis be placed on informing policy makers of the trade-offs involved in COVID-19 policies through evidence curation, collation and predictive modelling, so that such policy choices can be best informed, and tailored, in ways that reduce the disproportionate economic and health impacts on vulnerable populations. This is an important role for science, and one it is well-placed to play for several reasons.

First, the consequences of such measures are well known and predictable, and in that sense, are as much the “science” as the predictions regarding pandemic spread. There is evidence with predictive utility regarding the effects of wage reductions and job losses on physical and mental health and how they are mediated by prior inequality. Equally, we understand the health consequences of prolonged social isolation and on whom they might fall most heavily.

Second, these consequences map onto inequity, which is at the core of all three crises facing the U.S. We know that income is a powerful determinant of health, and that during the pandemic, lower neighborhood income has been associated with a lower ability to engage in social distancing, a condition that government physical distancing policies have not ameliorated. We know that Black Americans are less able to work remotely, are overrepresented among essential workers, have been overrepresented among the recently unemployed and are less likely to possess savings with which to weather gaps in employment.

We know that students from minority backgrounds are more likely to live in single-parent households, that one in three Black Americans and Latino Americans still do not have access to computers in their homes, or access to broadband and that gaps in educational attainment are widening. Yet, methods to predict, model or communicate these potential pitfalls remain woefully underdeveloped, even though the evidence underpinning these inequities and their effects on health is in some ways far more developed than the evidence on COVID-19-related policies.

Third, the likely impact on these groups is such that tailored relief, informed by the science, to ameliorate those consequences is therefore likely both cost-effective, highly desirable and politically expedient, yet it hasn’t been the focus of policy during the pandemic thus far. This has short- and long-term consequences that are closely related to some of the foundational causes for our vulnerability to COVID to begin with. Many of these solutions could aid social distancing by increasing the number of citizens able to do so and decreasing the disproportionately large and in many cases long-lasting sacrifices they must make.

The Biden administration’s plan to control COVID-19 has been lauded, perhaps correctly. However, we argue that honestly communicating on the nature of policy trade-offs, and what’s done to ameliorate them, should be at the heart of decision-making in the post-COVID era, both for the U.S. and the world. This would serve to improve trust and accountability in leadership and expand the role of evidence, while ensuring we place equity and health at the heart of our definitions of science and decision-making in the years to come.