Abstract
Arguably the most concerning new public health threat since Ebola is the Zika flavivirus, carried primarily by Aedes mosquitos. First identified in the monkey population of Uganda in 1947, an association between human maternal Zika infection and both the Guillain-Barré syndrome and microcephaly (or other problems of the central nervous system) among their newborns became evident only during the Brazilian outbreak of 2015. 1 As of August 2016, 70 countries have declared the presence of Zika virus transmission. 2
There are no vaccines (yet) available, with pregnant women (currently thought to carry the burden of disease) facing few options once infected, namely abortion, adoption, or lifelong care with or without financial and social security. 3 Any emerging technology’s effects are largely unknown, although the US National Institutes of Health recently started human trials of a DNA vaccine on 80 healthy subjects. 4 If all goes well with initial immunogenicity data, phase 2 trials hosted by Zika-endemic countries are expected to start early in 2017.
As with Ebola (for which drug treatments were more developed by the time of the outbreak), typical risk–benefit analyses may be of some, if limited, value because they rely heavily or entirely upon what can be quantified. The idea of the precautionary principle, on which most restrictive regulation of new drugs is based, seems to offer a way forward by introducing the value of caution unless and until there is evidence that these drugs are safe. Better safe than sorry. However, in times of public health crisis, from the HIV epidemic of the 1980s and Ebola of 2014, the precautionary approach itself increasingly requires justification.
Precaution, in the case of new medicines, lies in studying the drugs to give precise estimates of risks (and expected benefits) for later cost–benefit calculation. New medicines are restricted to these studies before allowing wide distribution on grounds that new drugs carry the possibility of causing severe harm. There is growing discussion, in the academic literature, of what nature of harm should qualify regulators to activate precautionary measures. Historically, the effects of thalidomide were enough of a catastrophe to warrant introducing regulation and precautionary measures for studying before distributing widely. 5 At this stage, it is probably unnecessary to explore the types of harm, their possible severity and whether they can be ‘reversed’, except to note that the drug was initially developed to treat nausea in pregnant women.
Sometimes, precautionary measures may, however, themselves be equally or more harmful – in terms of whatever criterion for precaution is chosen – by holding back potentially beneficial treatments while they are fully studied. Some critiques of the precautionary principle have taken issue with the logic of the principle, claiming that precautionary measures always have the potential to cause the very harm the principle is intended to prevent, thereby making the principle incoherent and hence irrational. 6 Precautionary measures, however, may not have similarly uncertain effects as the new technology itself, so may not necessarily themselves qualify for a level of precaution. During a public health crisis on the scale of Zika, it is easy, though, to see that this worry about precautionary measures themselves could become factually true. 7
The principle rests on the options between caution and the status quo being distinct and incompatible, such as in the choice to restrict a new medicine or to distribute it widely. During the Ebola crisis the development of new drug treatments was dramatically speeded up to meet the scale and severity of the crisis. 8 There was considerable debate over the apparent tension between allowing many patients to have access to investigational treatments while generating enough knowledge to learn what works and what does not. However, there is still little consensus over what evidence is required before new technologies are made more widely available, and how to obtain this information. The convention (since Thalidomide) is to run a randomized controlled trial of a new drug against best available care, which implies restrictions – if not on access to the trial, then on access to the investigational drug by virtue of the randomization process. Methodologically, any drug with a large effect size, perhaps sufficient to reduce the risk of a positive result really being caused by confounders or biases, would not need such elaborate and time-consuming evaluation. 9 Only drugs having moderate yet worthwhile effects would require such demonstration. During Ebola, however, there was an additional methodological problem raised by the variation in historical case fatality rates. 10 For Zika, we do not even know which disease outcomes are of principal interest, (and whether they vary to the same degree as mortality did in Ebola), let alone have experimental treatment options to test. Together with the patients being a particularly vulnerable group, there are further reasons to be cautious, meaning that the development of widely available vaccines and treatments for Zika is still a long way off.
To attend to the source of the threat, there is considerable effort to better control the mosquito population. However, following a long history of chemicals once used to combat another mosquito-borne disease, malaria, the possibility of new technologies directly harming human health and the environment also activates the precautionary principle. Standard measures now include the use of insecticides such as DDT, as well as removing stagnant water. 11
A British firm has also genetically modified the culprit mosquito to try to keep the population at a level so low that Zika transmission becomes unlikely. 12 Commercial release of modified mosquitos began in Brazil in 2014, and is currently planned in Florida, against some local opposition. A recent US Food and Drug Administration report concluded that any adverse ecological impact of the new technology would not to be significant, as each mosquito is able to travel over a relatively short range, making wider impact unlikely. At the same time, early empirical evidence suggests that the rates of disease are, indeed, reduced by releasing modified mosquitos. 13
As Christian Munthe argues, adhering to ideas of proportionality is crucial for blocking the objection that the precautionary principle is incoherent or self-defeating, which becomes a normative guide under uncertainty, not a decision tool. 14 Proportionality itself, however, is difficult to pin down, especially as it rests on assumptions about relative costs and benefits, even if they are not yet quantifiable. The reasons for being more cautious before injecting vulnerable humans with experimental vaccines than before releasing genetically modified mosquitos in the local environment seems proportionate to the problem of Zika. Ironically, though, the problem of new infectious diseases is thought to be at least partially a result of the climate change which human industrial activity has brought about, and against which precautionary measures might have averted. 15 To be cautious when trying to manage the symptoms from previous lack of caution may risk a pendulum swing from one extreme to another.