Ebola and the ethics of research design

A couple of recent articles stimulated my interest on the ethics of conducting ebola research, particularly with those conducted in low-resource settings and having a randomised controlled design. It interested me enough to write a letter to JAMA, where it got bounced.  I thought about submitting to Lancet, but hip replacement surgery got in my way: just too much work. So it goes. Part of what is fascinating about the topic are the underlying issues regarding whether doing anything other than the 'gold standard' of clinical trials is justified. I am diverting this piece to Global Bioethics Blog just in case a reader finds something of value in it. 

Two recent commentaries in this journal argue for and against conducting randomized controlled trials (RCTs) of new Ebola drugs during the current epidemic.[i] [ii] Those in favor argue that only RCTs can deliver the evidence required to treat future Ebola patients in ways superior to the current standard of care, which is largely palliative. Other observers hold a similar view.[iii] Those opposed to RCTs in this context argue that if patients were randomized to study arms of either (a) a new drug or (b) the baseline 70% mortality rate for Ebola, such a trial would not possess equipoise, because (arguably) the intervention arm would likely provide at least some benefit. In addition, opponents argue that local communities ravaged by Ebola – whose trust in authority, including medical authority, has been profoundly shaken -- are unlikely to accept a randomized controlled trial design.  They therefore advocate for experimental drugs to be offered to patients within non-RCT research designs, even if they have not been tested by the ‘gold standard’ methodology.   

On the face of it, this seems like a conflict between advocates of evidence-based medicine and those who understandably, but misguidedly, want to provide less-than-well-tested drugs to the sick as soon as possible. But while the former position seems rational and impartial, history reveals some unsettling patterns. When the HIV epidemic was raging in the United States decades ago, advocacy groups mobilized aggressively for expedited access to new experimental treatments, bypassing the full FDA approval process.[iv] The alternative then, for many AIDS patients, was death. Current policies surrounding ‘compassionate use’ were borne out of this experience. This raises the question: is it easier to take a hard utilitarian position on the need for RCTs when it is someone else’s epidemic?

The faith placed in scientific knowledge to resolve deep social problems is also part of an old pattern. When health crises occur in developing countries, international efforts often focus on fast-tracking biomedical interventions rather than also engaging the social, economic, and political factors contributing to emerging infections. Citizens of low-resource countries have every right to be skeptical here. Drugs to prevent mother-to-child HIV transmission were tested and developed two decades ago; only 57% of women in sub-Saharan Africa currently have access to it.[v] The same (or worse) can be said for a host of other diseases. Access to new Ebola drugs, should they be successfully developed, will likely to follow the same trajectory. Again, the argument that we should only encourage RCT trial designs for new Ebola drugs, out of scientific concern for the evidence base, may be more compelling in places other than Monrovia.

---

[i]Shaw S. Randomization is essential in Ebola trials.  Lancet 2014; 384: 1667. 

[ii]Adebamowo C, Bah-Sow O, Binka F, et al. Randomised controlled trials for Ebola: practical and ethical issues. Lancet 2014; 384: 1423-1424.

[iii]Joffe S. Evaluating novel therapies during the Ebola epidemic. JAMA 2014; 312(13):1299-1300.

[iv]Dresser R. When Science Offers Salvation. 2001. Oxford: Oxford University Press.

[v]WHO/UNICEF/UNAIDS. Global Update on HIV Treatment 2013: Results, Impacts and Opportunities. http://apps.who.int/iris/bitstream/10665/85326/1/9789241505734_eng.pdf?ua=1