Ethics of Ebola and Potentially Life-Saving Experimental Treatment

Summary: Patients potentially infected with Ebola have the right to experimental treatment outside of randomised controlled trials. Consent should be sought while they are competent if it is anticipated that they will lose competency through the disease. In the most extreme emergencies, experimental treatment could be given without consent in the public interest. However, in such extreme emergencies, such if the virus were to mutate to become very highly transmissible, selection of patients for such trials should be on a principle of justice. This would generally require selecting patients with the worst prognosis as initial subjects. Whilst the current outbreak has been declared an international health emergency, it does not warrant over-riding standard ethical principles of consent and autonomy.


WHO are convening a meeting next week to discuss the use of untested treatments for the current Ebola outbreak, which it has declared to be an international public health emergency.

Dr Marie-Paule Kieny, Assistant Director-General at the World Health Organization explained “We are in an unusual situation in this outbreak. We have a disease with a high fatality rate without any proven treatment or vaccine,”

With a fatality rate of recent outbreaks at around 50%, and current treatments consisting of little more than rehydration, it is easy to see why this experimental approach is an attractive option. Jeremy Farrar, David Heymann and Peter Piot argued convincingly in favour of such measures.

However, there are a number of diseases which meet Dr. Kieny’s criteria of “a high fatality rate without any proven treatment or vaccine”. While Kieny is referring to infectious disease, sufferers of other diseases have campaigned for many years for access to untested medicines. Along with several colleagues, I wrote a paper on this subject with Les Halpin, who recently died from motor neurone disease, having campaigned for access to medicines for many years

The case for

I have argued that, when a patient is facing a certain death from an incurable disease, the relative safety that randomised controlled trials provide is unnecessary- especially when the payoff is a placebo group of patients who receive no benefit at all.

One common reason in support of randomised controlled trials is safety. This is true for most patients: patients for whom there are other good treatment options, or whose disease is not itself life-threatening. But for some patients, this concern is a cruel irony. Their disease is so overwhelmingly unsafe that any risk is outweighed by potential benefits. Udo Schuklenk reports Martin Delaney, a trial patient for AIDS who described the situation vividly:

It is as if I am in a disabled airplane, speeding downward out of control. I see a parachute hanging on the cabin wall, one small moment of hope. I try to strap it on, when a would be government employee reaches out and tears it off my back, admonishing, ‘You cannot use that! It does not have a Federal Aviation Administration sticker on it. We do not know if it will work’

Against: Coercion, Exploitation and Injustice

If a patient faces losing their life, they are bound to be drawn to any hope that is offered. Is this coercion?

To offer patients an untested drug when there are no other options available is not coercive. The patient, at least in the early stages of the disease, still has the capacity to make an autonomous choice. It is sometimes argued that offering experimental treatments to dying patients is coercive as they have no other option. This is a mistake. When an offer is made with the status quo as an option, any choice for the intervention is based on an evaluation that it is an improvement on the status quo, that is, that it is the rational choice.

What people really mean when they raise the spectre of coercion is that such patients are exploited. Exploitation occurs when there is background injustice and a person is made an unreasonable or unfair offer that they would not otherwise accept, except for the poverty or injustice of their circumstances. Examples are when people living in unjust poverty are paid low prices for their organs or risky labour.

But there is no exploitation in offering potentially lethal or dangerous experimental treatments to people who have a high chance of dying anyway. Exploitation would be if we paid poor people small amounts of money to engage in challenge studies or if they were given drugs with no reasonable prospect of improvement simply to determine physiological reactions. But for those who are afflicted with Ebola, or potentially afflicted, they are in a bad situation.

In the case of some infectious disease outbreaks, the origin of the infectious disease may be in part, even substantial part, because of some prior or background social injustice – say poverty, poor housing or lack of effective public health. In such cases, there are strong reasons to relieve injustice. But once an outbreak has occurred, it is not exploitation to offer participation in experimental trials.

We have no third option to offer these patients, no safe tested cure. Therefore, offering an untested, possibly unsafe option is just that, an offer. A patient may, reasonably and autonomously, refuse the offer and they will be in the same situation as they were before.


As the disease progresses, however, it is possible that patients will no longer be competent to make an autonomous choice. One solution is to seek consent at the earliest stage, when Ebola potential victims are identified with risk factors and early symptoms. A potential victim could say, for example, if I do develop the symptoms and signs of Ebola and become unconscious, I agree to experimental treatment X with risk factors A, B, C

What if the treatment is most effective in the early stages, before the person is sure he or she has Ebola? Such a person might have travelled to a danger zone but just have a cold. The critical issue here is clear explanation of the risks of having the infection, and the risks and benefits of the intervention, insofar as they are known. In some cases, there may be great uncertainty about both risks. Still, people should be given the opportunity to make their own decisions.

Could there Ever be a Situation Where it is Acceptable to Give Experimental Treatments Without Consent?

In the most extreme potential cases, experimentation without consent might become necessary in the public interest. For example, imagine that Ebola mutates and becomes as transmissible as flu, with millions dying, or if a bioterrorist genetically engineered such an infectious strain. Extreme measures would then need to be taken. This might require a higher confidence that any new intervention is of relatively low and acceptable risk, compared to the disease and conservative management. Higher risk strategies could be employed with the consent of the patient. This is possible when potential victims are identified with risk factors and early symptoms.

The most difficult case ethically is where experimental treatment is possibly very dangerous, even lethal, and must be given when the patient is incompetent. In such a case, it may be that patients are used, at least in part, as a way of developing effective treatment for future patients. But if the numbers of future fatalities are large enough, such potentially lethal experiments on incompetent patients might be justified. But how should patients be selected as recipients of this potentially dangerous but potentially life-saving treatment?

The answer is according to some principle of justice.

According to one theory of justice, egalitarianism, people should be treated equally. Some kind of lottery should be employed to equally allocate the intervention to potential recipients.

According to utilitarianism, those with the worst prognosis should be the first recipients since they have the least to lose – they are most likely to die . However, if one group is more likely to benefit, utilitarianism would favour this group. These reasons may pull in opposite direction and, overall, utilitarians will have to balance these.

A third theory of justice is prioritarianism – this involves giving priority to the worst off. In this case, this will be those with the most to lose, that is, those with the worst prognosis. This coincides with one utilitarian justification and speaks in favour of giving greater weight to gravity of prognosis over likelihood of benefit, at least in the first phases of experimentation.

The Decision We Face Today

Fortunately, the Ebola outbreak is not of such proportions and consent for experimentation can reasonably be obtained early in the course, even if treatment is only given in the later stages. If patients consent to treatment early in the course, there is a downside – patients who might not have died may be exposed to the risks of experimental treatments. The expected harm is lowest as prognosis deteriorates. This is an argument (not conclusive) for administering interventions later, rather than earlier. There may be countervailing reasons of potential efficacy of early intervention that could outweigh these.

In this time of crisis, I believe we should allow Ebola patients to decide whether to use the untested medications. But if we do, we should ask ourselves why we continue to deny this to other patients with rare, incurable, fatal diseases who are asking for the same freedom.

  • Facebook
  • Twitter
  • Reddit

4 Responses to Ethics of Ebola and Potentially Life-Saving Experimental Treatment

  • Sarah says:

    “I agree to experimental treatment X with risk factors A, B, C”: isn’t the trouble here (in the case of ebola) that we don’t know risk factors A, B and C?

    As an ebola patient I would know I have a roughly 50% chance of dying but that equally, I could also recover and be perfectly healthy for the rest of my life. But I don’t know whether the drug will be as safe as paracetemol (in the right dose) or as dangerous as the Northwick Park drugs and have a 100% chance of killing me, or will be ineffectual but will harm me in the longer term if I happen to survive.

    This seems to me to be different and more difficult ethically than in diseases such as motor neurone disease or AIDS (prior to effective treatment) where the patient knows the disease is fatal, and has some idea of the likely progression time of their disease. Therefore, even if the risks of the experimental drug are equally unknown as in the ebola case, the only relevant potential risk is that they will die sooner than they would have anyway which, although a difficult choice to make is more akin to the ideal situation of being able to say “I agree to experimental treatment X with risk factors A, B, C”. They can pick themselves the point in the disease at which they are prepared to take that known risk.

    • Julian Savulescu says:

      Thanks Sarah. You are right. The cases are different. But as long as that is explained to patients, it should be up to them to decide, not their doctors, health authorities or governments. Some people will opt for the uncertainty of the disease, others for the uncertainty of the medicine. Life is uncertain – being a person is about making your own choices about the risks you take, and those you don’t.

  • Owen Schaefer says:

    Hi Julian,

    I agree that experimental treatments should be made available, given the abysmal current standard of care. There is some analogy to the right-of-access claims that have been successful in the case of HIV, though the present case of Ebola is much more acute.

    However, I *do* worry about a further issue – that distributing experimental treatments will constitute de facto experiments, with observed results used to further develop the treatments. Patents to such treatments are, I understand, privately held, and the company with the successful drug could reap significant financial rewards. Additionally, participating in such de facto experiments *would* be exploitative and maybe even coercive, if the offer is ‘you can receive this drug, so long as you agree to have your results made available for analysis’ – essentially, that the recipients take part in an observational study. The provider would be taking advantage of a desperate situation, with potentially significant financial gain if the results are positive. This becomes coercive if we accept a right to access regardless of experimental participation.

    The proper remedy is not to deny access, of course, but to be very careful that receipt of the treatment is not contingent on experimental participation. One way is to simply restrict results from being used for experimental/development purposes. But this is too strict – there is some public interest in developing Ebola treatments quickly, and more data on effectiveness of current remedies would be valuable (indeed, if the treatment has severe toxic side effects, we would want to know ASAP). Better, then, that recipients’ data *can* be used – but they must have the available option to deny drug developers access to their results, while still being free to receive treatment. Perhaps a two-stage consent process, something like this: Step 1, you get consent to the treatment itself (laying out known benefits/risks). If they agree, they are assured they are now guaranteed treatment no matter what they say in step 2. Step 2, they are asked whether or not they would be willing for their results to be released to the drug developer, which may lead to better treatments in the future and which may lead to profit for the company in question (of which the participant will receive nothing).

    A more radical solution is profit-sharing. This becomes more plausible in the extreme case mentioned above, where consent must be waived, but perhaps should be extended to cases of regular consent as well. Again, participants’ desperation for a treatment could be facilitating the developer’s profits – exploitation is a serious risk. To address this exploitation, we may want to recognize that participants are owed additional compensation for their contribution, if the treatment turns out to be significantly profitable. This approach has its own downsides; it would likely dampen the profits of the developers, disincentivizing development of neglected diseases like Ebola. But it may nevertheless be a demand of justice.

    • Julian Savulescu says:

      Thanks Owen but I disagree that patients should be able to withhold their data. Every person receiving any experimental treatment must provide the data related to that treatment on a publicly accessible database – otherwise we can’t confidently work out if treatment is effective and we risk publication bias. Every patient has a moral obligation to be a data point in the progress of medicine. The issue you raise of patenting and commercialisation is different. How treatments ought to be developed and priced is a separate issue over access to information and knowledge. Companies, by the very nature of capitalism, will aim to make a profit out of R and D. Perhaps there should be caps on this. Or perhaps governments ought to buy the products. Or perhaps the State ought to invest more in R and D making its results available to all. These are all interesting issues but every time a drug is given, whether in an Ebola experiment or not, data ought to be systematically collected and made publicly available, with appropriate anonymisation. So much more could be achieved, and so many disasters averted, if we used our now enormous data mining capacities to understand just what we are doing and what we could do.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use the <em>, <strong> and <blockquote> tags. Links have been disabled to combat spam.

Notify me of followup comments via e-mail. You can also subscribe without commenting.