COVID-19: Medical Research in a Rapidly Developing Pandemic
Posted on 20th April 2020 by Robert Wight
In December 2019, a novel coronavirus strain (now known as SARS-CoV-2) emerged in the Hubei province of China. As of April 2020, there is an ongoing pandemic of COVID-19, the respiratory disease caused by this virus.
The rate at which this pandemic is developing poses substantial challenges to all, from front-line healthcare workers treating patients, to governments making decisions on national policies. Clinical health researchers face the delicate balancing act of producing rigorous, reliable research, but doing so in as short a timeframe as possible.
The Race to Find Effective Treatments
Clinical research is necessarily a careful endeavour, checked by regulation that is designed to minimise the chance of harm to participants and to future patients. Well-designed trials can take years to reach their conclusion, whereas the current situation demands answers as soon as possible.
The clamour for useful interventions has led to calls for the early adoption of treatments that have not yet been tested in randomised controlled trials. One prominent example of this is hydroxychloroquine, a drug that is used to treat malaria, as well as autoimmune conditions such as rheumatoid arthritis and lupus. This drug, possibly in combination with the antibiotic azithromycin, has been promoted by public figures and has already been incorporated into some treatment guidelines. There have been reports of doctors stockpiling it for themselves and of people with lupus and rheumatoid arthritis being unable to pick up their prescriptions due to shortages. All this has come on the back of a non-randomised trial of 36 participants, which has serious questions to answer regarding both its methodology and the authors’ publication ethics (1).
A commonly heard rationale for the use of hydroxychloroquine is that its safety profile has been studied previously and so, in this desperate situation, there is no harm in trying it. There are, however, a number of problems with this line of reasoning. Firstly, while they are relatively safe, drugs such as hydroxychloroquine and azithromycin are not without risk. Both are potentially cardiotoxic, for example, and there have already been reports of deaths due to chloroquine poisoning (hydroxychloroquine is generally safer than chloroquine, but also not risk free) (2). Secondly, we have very little data on the safety of these drugs in this particular patient group, and the history of medicine is littered with examples of unanticipated harms that only emerged when drugs were studied properly. Thirdly, for every patient treated with an unproven drug outside of a randomised controlled trial there is an opportunity cost: they could have been enrolled in a trial to study the effect of that drug (or one of several other candidates), providing vital knowledge about their efficacy and safety. There is another aspect to this last point. As well as eschewing the opportunity to gain useful knowledge about treatments, the widespread adoption of unproven therapies can give the impression that they are already known to be effective. This, in turn, may make it more difficult to recruit participants for randomised trials (3,4).
Countless trials for COVID-19 are now being run across the world, studying hydroxychloroquine as well as a number of other candidate treatments. These trials should be able to produce meaningful results in unprecedented timeframes, as there are a great number of potential participants presenting every day. Additionally, as this is an acute disease, the follow up required to show efficacy is relatively short. Some aspects of regulation, such as the need to ensure control groups receive the best available standard of care, are also less relevant in the context of this previously unknown disease. Given the potential to produce good quality research at an unprecedented rate, there really is no excuse to cut corners by accepting lower quality evidence than we usually would.
Making Sense of a Deluge of Evidence: Some Useful Resources
As alluded to above, a large number of clinical trials have now been initiated to study treatments that may be of benefit. There are also many more studies underway to evaluate other aspects of care, and a vast body of existing evidence that may or may not have some relevance to COVID-19. As a result, researchers, clinicians and the public must grapple with an ever-increasing body of relevant literature.
In efforts to help with this newfound information overload, researchers with expertise in evidence synthesis have been pooling resources to produce summaries of relevant research. Cochrane has established a working group to fast-track the updating of relevant systematic reviews, produce new rapid reviews and collate useful resources. Access to the Cochrane library is also temporarily unrestricted worldwide.
The Centre for Evidence Based Medicine (CEBM) at the University of Oxford are conducting rapid reviews of evidence for a broad range of questions, designed to be useful for clinicians and members of the public. Readers can submit their own questions by email and the reviews are regularly updated with new information.
While it will be necessary at times to make decisions based on little evidence, or using lower quality evidence than we would like, this does not mean that weaker forms of evidence are the new gold standard. We must continue to be aware of the limitations of the evidence we use and, wherever possible, treatments must be tested with the same rigour that we would usually expect.
There is a daunting quantity of evidence being produced on COVID-19, and resources such as Cochrane’s information pages and the CEBM Evidence Service are valuable tools to help in finding useful, reliable information.
- PubPeer. Hydroxychloroquine and Azithromycin as a treatment of COVID-19: preliminary results of an open-label non-randomized clinical trial [Internet]. [cited 2020 Mar 30]. Available from: https://pubpeer.com/publications/B4044A446F35DF81789F6F20F8E0EE
- Food and Drug Administration. Hydroxychloroquine FDA Drug Label [Internet]. [cited 2020 Apr 1]. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/009768Orig1s051lbl.pdf
- Mills N, Donovan JL, Smith M, Jacoby A, Neal DE, Hamdy FC. Perceptions of equipoise are crucial to trial participation: a qualitative study of men in the ProtecT study. Control Clin Trials. 2003 Jun;24(3):272–82.
- Rooshenas L, Elliott D, Wade J, Jepson M, Paramasivan S, Strong S, et al. Conveying Equipoise during Recruitment for Clinical Trials: Qualitative Synthesis of Clinicians’ Practices across Six Randomised Controlled Trials. PLoS Med. 2016 Oct;13(10):e1002147.