As an anthropologist and STS researcher, a great deal of my academic career has been proudly dedicated to studying and denouncing the bias, inequalities, and prejudice within both scientific and medical practices. Such critique, far from intending to undermine scientific credibility, comes from a place of deep respect, trust, and, I dare say, great optimism regarding what kind of project we have for science in the long term: one where knowledge is comprehensive and accessible, and where expertise is not build upon the concealment of information.
One of the many important topics highlighted by the COVID-19 crisis is the undeniable credibility struggle that scientific knowledge is facing, which can be seen in the widespread controversies around the usage of masks, the efficacy of quarantining, and the safety and benefits of some drug therapies. But such contends do not seem to come from the creation of more democratic and cooperative spaces that we have been trying to build. Instead, they are part of a layered, multifaceted phenomenon deeply connected to the emergence of far-right movements worldwide, the consolidation of strategies we’ve come to dub “fake news” and post-truth, and, of course, the fragile basis on which science has built its authority: the hiding of controversies and erasure of human agency from scientific practices that we call black boxing (Epstein 1997, 29). Hence, part of this growing distrust seems to come from a very honest place: many people feel that science is not transparent in its interests, and its methods are often perceived as too remote, too technical and too abstract (Crease 2019).
In that sense, black boxes have worked well to ensure that people have no option but to take science’s word about things—but bad practices and choices may age poorly away from sunlight exposure. This was exactly the case three decades ago when AZT (Zidovudine) was first presented by Burroughs Wellcome labs as a promising drug with the potential to put an end to the AIDS crisis. AZT wasn’t designed for fighting HIV infection; instead, it had been created decades prior for cancer treatment, but was deemed “too toxic” for such use. The AZT hype began as soon as the lab showed the wonderful results that the compound had on fighting HIV in vitro, the very first step of an extremely long list of procedures designed to test both the efficacy and the safety of new drugs. With the immense pressure of the government, doctors, patients, and activists, the drug was approved in record time by the FDA, supported by trials that were shortly after proved not only to be deeply flawed but also unblinded by doctors and patients alike in the first week of the drug trials (Epstein 1997, 202).
So, in 2020 when it became clear that the novel disease caused by an unknown coronavirus was set to become a major health crisis, and I started seeing pharmaceutical companies combing their shelves for a “miracle drug” drug to fight and eventually cure it, my initial thought was: this is all too familiar. Since chloroquine (and its derivative hydroxychloroquine) had already been considered as a possible drug therapy for other SARS viruses (Vincent et al. 2005), they rapidly became a candidate for treating SARS-CoV-2 infection. In a matter of weeks, dozens of studies began taking place—first in China, and soon, all over the world—and chloroquine monopolized the debate as a possible salvation for the imminent pandemic despite the lack of evidence pointing in that direction. Many scientists, scholars, and doctors started asking for caution and clarity in the pursuit of this hypothesis, and pointed out that the results were incipient, only the beginning of a long chain of required steps. But this is a hard thing to argue in a context of unprecedented social anxiety, especially when the tools for better understanding how science is actually produced and a broader understanding of what “good science” is have been historically denied to the general public.
In the midst of hundreds of peer-reviewed publications, two studies on hydroxychloroquine became particularly prominent in the run for pinpointing the real place it would occupy in the history of COVID-19: “Hydroxychloroquine and azithromycin as a treatment of COVID-19: results of an open-label non-randomized clinical trial” (Gautret et al. 2020) and “Hydroxychloroquine or chloroquine with or without a macrolide for treatment of COVID-19: a multinational registry analysis” (Mehra et al. 2020). Even though they claimed very different things, both studies received a lot of media coverage worldwide, with the first one becoming instantaneously championed by negligent administrations such as the ones led by Donald Trump in the US and Jair Bolsonaro in Brazil. Gautret’s study was published in March 2020, and it seemed to confirm that the use of hydroxychloroquine (in combination with azithromycin) could help control the viral load of infected people. However, not long after being released, the study was heavily criticized for its methodology, because the subjects in the control and in the treatment group must be as similar as possible in a comparative study. They must have about the same age, have the same gender ratio, and be in the same stage of the disease, but these guidelines were not followed in this experiment. This blatant flaw in the design, among other criticism, caused Elsevier, the prominent and reputed publication house that had published the research, to state that “concerns have been raised regarding the content, the ethical approval of the trial and the process that this paper underwent to be published within International Journal of Antimicrobial Agents” (Joint 2020).
A few months after that, in May 2020, the Mehra study was published, and it seemed to prove not only that hydroxychloroquine was not effective in treating COVID-19, but that it actually increased the patient’s risk of death. This time, the study was even more impactful, to the point that the World Health Organization decided to recommend the suspension of every other study with the drug. But it didn’t take long until it was made clear that this research too had serious problems, this time regarding the veracity of its primary data. After a manifesto signed by over 100 scientists was published, the Lancet retracted the study supported by three of the four authors: “The Lancet takes issues of scientific integrity extremely seriously, and there are many outstanding questions about Surgisphere [the company that provided the study’s data] and the data that were allegedly included in this study” (Lancet 2020).
In the middle of an immense credibility crisis, and a global health crisis, two important peer-reviewed studies that could potentially alter the course of the pandemic were discredited by other scholars in a matter of weeks after their publication. This is not at all uncommon; retraction is a well-known practice in scientific journals. But it is possible that these episodes will be used to fuel the current suspicion of science even further. After all, the truth remains that both of these studies were extremely flawed: the first was poorly designed and the second had very questionable data, besides other ethical concerns raised in both cases. In Gautret’s study, for example, there were serious indications that the study started before it was approved by the French National Agency, which is obligatory in studies where people are given drugs initially approved for another disease. In Mehra’s study, on the other hand, the main problem was a conflict of interest. Not only does one of the authors own Surgisphere, the company that provided the primary dataset on which the conclusions for the study were drawn, but they also failed to provide explanations on how the data for the study was gathered, after being questioned by researchers who pointed out concerns regarding the unusual homogeneity of the data from different countries. In this case, there was also the matter of inclusion of data from hospitals that didn’t have digital records of their patients, which raised even greater suspicion on the authenticity of the dataset.
In conclusion, it is clear that, ideally, such studies should never have been published. But these are exceptional times. Governments and people from all over the world, the World Health Organization, big pharma companies, and the main business conglomerates worldwide are all demanding that science comes up with a solution for this pandemic. People are scared, tired, isolated, and want a rapid return to normalcy. So, yes, these studies did get published. And since this happened, it is a good thing that such discussions are taking place in broad daylight.
Then, once more, I reflect on the AIDS epidemic and how part of the scientific community settled for poorly collected results. I think about how quickly the US administration jumped on the boat of a drug with potentially deadly collateral damages. AZT completely dominated the market and the research surrounding HIV treatment for an entire decade, but this had nothing to with it being indisputably beneficial—it happened because it became extremely difficult to shift attention to any alternative solutions. Many patients, feeling isolated and neglected, risked their lives testing AZT and other drugs on themselves without any medical monitoring whatsoever, having been denied the participation in such spaces of knowledge production. In the end, the great game changer for positive patients was the Highly Active Antiretroviral Therapy (HAART), a treatment composed of AZT and two or more drugs from different classes, which completely transformed the AIDS epidemic and the survivability rate of people living with HIV. This combined drug therapy only exists due to the efforts of several independent groups and patients who kept asking for better drugs, who pursued better pharmacological alternatives themselves, and who kept pushing for less inert research practices.
But this is where the parallels between the two drugs end: the World Health Organization has officially halted any further trials with chloroquine and its derivatives based on several studies from all over the world that have offered definitive proof of its inefficacy in fighting COVID-19. Unlike AZT, chloroquine will be regarded as no more than a transitional drug like so many others tested in the initial stages of an epidemic. So, to a certain extent, I do see the chloroquine dispute as a sign that black boxes might be becoming more translucent. After all, it is a positive thing that we could see this controversy unwoven in real time. It is good that scholars and researchers are keeping each other in check and holding each other’s works to the highest standards of quality and ethics, but it is also extremely important for us to remind ourselves that, contrary to what our instincts might indicate, good science is made possible by failures and mistakes. At the same time, we cannot downplay the impact that pernicious studies and experimentations have had on the general perception of scientific practices: there are far too many Tuskegees, Dalkon Shields, and other despicable cases of malpractice and abuse in science’s account, especially against Black people, Indigenous people, and women.
There are still many ethical challenges ahead of us when it comes to this pandemic, not only because chloroquine is still being largely used in Brazil against all recommendations, but also because vaccine testing and distribution are rapidly becoming a reality and this should raise our concerns on how such practices will be handled. Also, it is clear that actors from separate spheres, such as politicians, policymakers, and major companies, will continue to play a central part in the way many of these implementations will be conducted, adding many layers of complexity to such ethical decisions.
Finally, let’s not forget that epidemics pose some impossible choices for researchers and medical practitioners: should we be purists and inflexible with scientific standards, risking the loss of many lives by denying them access to potentially beneficial therapies? Should we risk people’s health by administering them drugs whose collateral damages are still unknown? Should we delegate the choice to the patients, knowing that many of them do not have the appropriate tools to evaluate the risks? Forty years after AIDS, we still don’t have the answers, and chances are that we will never have them. Nonetheless, it is in our power to push for transparency, knowledge accessibility, collaborative production, and better science education and literacy. If we want scientific practices that are less exclusionary and truly democratic, we better start turning our limitations into strengths, and disengaging our credibility from a supposed infallibility while we are still part of the collective debate. If we don’t, other actors will fill in the gap with far more palatable answers and we will be left talking to ourselves.
Crease, R. P. 2019. The Workshop and the World: What Ten Thinkers Can Teach Us about Science and Authority. WW Norton & Company.
Epstein, S. 1996. Impure Science: AIDS, Activism, and the Politics of Knowledge. University of California Press.
Gautret, P., J. C. Lagier, P. Parola, L. Meddeb, M. Mailhe, B. Doudier, … and S. Honoré. 2020. “Hydroxychloroquine and Azithromycin as a Treatment of COVID-19: Results of an Open-label Non-randomized Clinical Trial.” International Journal of Antimicrobial Agents, 105949.
Joint I. S. A. C. Elsevier statement on Gautret et al. paper [PMID 32205204].
Lancet. 2020. “Retraction: Study on Chloroquine and Hydroxychloroquine in COVID-19 Patients.” ScienceDaily. Retrieved June 12, 2020 from www.sciencedaily.com/releases/2020/05/200522113712.htm.
Mehra, M. R., S. S. Desai, F. Ruschitzka, and A. N. Patel. 2020. “Hydroxychloroquine or Chloroquine with or without a Macrolide for Treatment of COVID-19: A Multinational Registry Analysis.” Lancet (published online May 22). 10.1016/S0140-6736(20)31180-6.
Vincent, M. J., E. Bergeron, S. Benjannet, B. R. Erickson, P. E. Rollin, T. G. Ksiazek, N. G. Seidah, and S. T. Nichol. 2005. “Chloroquine is a Potent Inhibitor of SARS Coronavirus Infection and Spread.” Virology Journal (2)69. https://doi.org/10.1186/1743-422X-2-69.