A recent article on the risks of treating COVID-19 with (hydroxy)chloroquine and a related article on cardiovascular risk and COVID-19 had a snowball effect involving the scientific community, the media (TheScientist, The Guardian and ScienceMag), and regulators.
Whether we trust or doubt these papers, the case calls for us to reflect on how (public health) decisions are made, how the underlying ‘evidence’ is generated and disseminated, and how exposed and vulnerable to manipulation the system we rely upon is.
The process revolves around three fundamental steps:
- how information is obtained, analysed and interpreted, and by whom;
- how it is communicated and shared;
- how it is used –what does one do with it.
When any of these building blocks of evidence-based policy making falters, the whole process risks collapsing, including people’s trust in science and policy making. For the process to arrive at the best possible decisions and be trustworthy, it requires singular public-health purpose and transparency and accountability at each step.
As the hydroxychloroquine example illustrates, we have an increasingly opaque system of data generation and ownership. In this case, a for-profit organization enters into agreements with “governments, countries and [ostensibly public and private] hospitals” – through which the latter give away information about their citizens and patients on exclusive and secretive terms, as raw material for business. At the same time, the for-profit organization has agreements with highly-respected academia for analysis and publication – raising the question as to whether they were fully informed of the underlying third-party agreements, and knowingly and unreservedly signed off on their duty of transparency and integrity.
We also have very influential publishers, which carry a lot of weight and are expected to be guarantors of the robustness of the presented evidence, but which tolerate or fail to see fallacies and shortcomings in the material submitted, and forego their own very commitment to transparency and data sharing (see also). Public scrutiny proved way more efficient in pinpointing deficiencies in the publications that were missed by the journals’ peer-review system. With mounting pressure, some authors have now withdrawn, and journals retracted the papers (see also).
All this occurs within a profit-driven, extractive business model that monetizes medical information and academic research, which has long been condoned and is here being exposed. Publishers of scientific journals are fundamentally themselves money-generating enterprises, often owned by large corporations, driven by the need to be market leaders. Increasingly, also academia is subjected to commercial pressures. COVID-19 is just amplifying these anomalies. It is absolutely critical to question the legitimacy of such a system, where vital evidence to inform decisions about our health is controlled by opaque alliances between data users and providers: data-mining companies, publishers, hospitals and researchers.
And finally we have the international and country policy makers and regulators (the World Health Organization and national ministries of health and regulatory agencies) under close scrutiny and immense societal and political pressure, which, without close examination of the facts, are swayed by conflicting information to first halting (the lowest-risk option) and then within days reopening enrolment into trials. They will now be compelled by hard evidence to stop (hydroxy)chloroquine treatment arms altogether.
Public health decision making has direct impact on people’s lives, including the difference between life and death. It deserves the utmost level of scrutiny, transparency and accountability in terms of methodology and potential conflicts of interest. The secrecy and lack of verifiability associated with commercial data-mining business models, and the commercialization of medical information more broadly, are incompatible with those principles.
Alternatives exist, and prove to work. ISARIC (the International Severe Acute Respiratory and Emerging Infection Consortium) collects, analyses and reports on clinical information of hospitalized COVID-19 cases. Currently this has ~67,000 individual patient’s data from 407 sites across 37 countries; contributors sign terms of submission which stipulate what we can and cannot do with the data; the data belong to the contributing institutions which act on behalf of their patients; analyses are agreed upon with contributors, and all contributors are acknowledged in the reports; all this is publicly available; we are now moving towards terms of submission whereby contributors may agree to have their primary data made available for third-party use and analyses through data access terms and conditions. All that is done in full transparency, through voluntary contributions in the public interest and without financial transactions – as one should expect to be the case for public health.
Blog by Piero Olliaro, Professor of Poverty-Related Infectious Diseases and Director of Science, ISARIC Global Support Centre (International Severe Acute Respiratory and emerging Infection Consortium)
Centre for Tropical Medicine and Global Health, University of Oxford, Old Road Campus, Roosevelt Drive, Oxford OX3 7BN, UK
Competing interests: I am Director of Science of ISARIC (the International Severe Acute Respiratory and Emerging Infection Consortium) Global Support Centre, hosted by the Centre for tropical Medicine and Global Health, Nuffield Department of Medicine, University of Oxford. I am also one of the signatories of the open letter to the Lancet, and an investigator of a trial of prevention of COVID-19 infection with (hydroxyl)chloroquine.