Opt-out and respect for autonomy

Opt-in policies, which require people to actively register as organ donors, are often thought to better respect people's autonomy than opt-out policies since they aim to secure people's actual consent to organ retrieval. In many jurisdictions moreover, people's decisions to opt in constitute legal consent to retrieval.

Gill argues that the principle of respect for autonomy does not require consent for organ retrieval, but instead requires that care providers act in accordance with the wishes of the deceased.3 Since more people prefer organ retrieval than not, opt-out better respects people's autonomy than opt-in.

To make this argument, Gill first distinguishes between two models of respect for autonomy. On the non-interference model, ‘it is wrong to interfere with a person's body unless that person has given us explicit permission to do so’.3 On the respect-for-wishes model, ‘we ought to treat a person's body in the way that he wishes it to be treated’.3 Gill argues that the non-interference model should govern our treatment of competent individuals, but should not govern our treatment of brain-dead individuals.3 If the non-interference model governs our treatment of brain-dead individuals, we would not be permitted to treat their bodies in any way other than those ways they had explicitly authorised—an implication that is absurd since we must treat their bodies in some way.3 Since organ procurement decisions concern the treatment of brain-dead individuals, we must opt for the respect-for-wishes model and implement the organ procurement policy that results in the fewest mistakes about people's preferences.3 Since most people prefer to be donors, opt-out, compared with opt-in, will result in the fewest mistakes and so will be more respectful of people's autonomy.3

Gill is correct that the non-interference model is not sufficient to decide the treatment of brain-dead individuals. But, Gill's argument misconstrues the nature of the policy problem posed by organ donor registration. Gill presents this problem as one of deciding what we should do with brain-dead individuals. As Gill puts it, “The first thing to note about a situation in which we are trying to decide whether or not to remove organs for transplantation is that the person whose body houses the organs in question is, when we are trying to make that decision, brain-dead.”3 Gill continues,If, then, our goal is to respect the autonomy of brain-dead individuals, we have no choice but to operate under the respect-for-wishes model of autonomy. And according to the respect-for-wishes model, we ought to implement the organ procurement policy that results in the fewest mistakes. If, therefore, presumed consent will result in fewer mistakes than the current system, presumed consent will be more respectful of autonomy than the current system.3

For Gill therefore, the problem that organ donor registration policies are meant to address is one of surrogate decision making for brain-dead individuals.

But, this is mistaken. Organ donor registration policies of course have implications for this type of decision making, but the question they address is a different one. These policies offer ways of ascertaining presently competent people's preferences regarding a treatment question care providers will face when these people are no longer competent, namely whether to retrieve their organs. The question organ donor registration policies aim to answer is thus neither, “how should we treat this competent person?” nor “how should we treat this non-competent person?” Rather, it is “how should we register presently competent people for an intervention that will occur—if it does—when they are no longer competent?”

What does the principle of respect for autonomy require in situations like this? Does it require that agents take steps to secure the actual consent of those presently competent persons for an intervention they may undergo when they are not competent? I claim that it does, when three conditions are satisfied: (1) it is not unreasonably costly for agents to secure consent; (2) there is a real possibility that decisions regarding the treatment in question will need to be made; and (3) the choice of treatments is likely to matter to people.

Consider first that the principle of respect for autonomy requires that we treat non-competent individuals who were previously competent in accordance with their previously expressed authorisations. Thus, in cases of surrogate decision making, care providers ought to follow advance directives, before turning to the principles of substituted judgment or best interests.9 The underlying principle here is that we best respect non-competent people's autonomy by treating them in accordance with their competent authorisations. Since agents best respect non-competent people's autonomy by treating them in accordance with their competently expressed authorisations, in cases where agents know that decisions will need to be made in the future for non-competent individuals who are presently competent, it follows that agents better respect their autonomy—that is, recognise the value of their governing their own lives—by asking them to authorise future treatments, rather than relying on the principle of substituted judgment or best interests once they become non-competent.

Now, the principle of respect for autonomy would be too demanding if it required that agents asked presently competent people—who may be non-competent in the future—to authorise particular courses of treatment regarding all manner of situations. But, it does seem to require that agents seek people's authorisations under the conditions I outline above: (1) it is not unreasonably costly for them to do so; (2) there is a real possibility that decisions regarding the treatment in question will need to be made; and (3) the choice of treatments is likely to matter to people. The following two cases, I suggest, support the claim that agents should seek people's authorisations when these three conditions are satisfied.Surgery: Dr Herrera is discussing a future surgery with his patient Kim. Dr Herrera knows that one possible consequence of the surgery is that Kim may need a blood transfusion. Dr Herrera won't know whether this transfusion is necessary until the procedure is well underway and Kim is unconscious. Dr Herrera knows that a not-insignificant number of people in his community are members of the Jehovah's Witness faith. Dr Herrera has only met Kim recently and does not know what her faith is. Does he have an obligation to ask Kim to consent to a blood transfusion prior to the procedure?Alzheimer's: State-sponsored investigators operating in North Carolina have found it increasingly useful to conduct research on patients with Alzheimer's disease. As a result, the surrogate decision makers for Alzheimer's patients are routinely asked to enrol these patients in research protocols. In response to this development, policy makers are considering whether to include an option for research participation on standardised advance directive forms. Does the government of North Carolina have an obligation to do so?

I suggest that Dr Herrera in surgery and the government in Alzheimer's have a duty to seek the authorisations of the respective patients for the treatments in question. Doing so better respects patient's autonomy, and does not impose significant burdens on those charged with securing these authorisations.

With respect to the question of organ donor registration, the claim that agents ought to secure the authorisations of presently competent people for the treatment of their non-competent future selves under the conditions I outline above implies that governments should enact policies that seek to secure people's consent to donation. Organ donor registration policy satisfies conditions 1–3 since (1) the implementation of opt-in or mandated choice policies is not unreasonably costly to government agencies; (2) decisions regarding procurement must be made regularly; and (3) many people have strong preferences regarding the use of their organs. Contra Gill therefore, opt-out policies, which do not aim to secure people's consent to donation, fail to adequately respect people's autonomy.

This does not mean that consent is a necessary condition of permissible organ retrieval. My claim that respect for autonomy requires governments to take steps to secure people's consent to organ procurement does not imply an answer to the question of the permissibility of organ retrieval from those who have not consented. My own view is that it is permissible to retrieve the organs from people who have not consented to donation in cases where they have not explicitly dissented from organ procurement, and they have expressed a clear preference for procurement. Such organ retrieval is permissible, I argue, since it is permissible to intervene with a person's body without her consent in cases where (1) she is not able to consent to the intervention now or in the future, and (2) she has expressed a clear preference for the specific intervention in question.10 Since a chief role of consent is to protect people's ability to govern their lives and interactions with others in accordance with their preferences, in cases where (1) and (2) are satisfied, treating people in accordance with their expressed preferences best realises this value. This principle also explains why it can be permissible for surrogate decision makers to authorise interventions for an incapacitated patient even in the absence of an advance directive, and why surgery is permissible in emergency situations.^iv Importantly, this argument does not imply that organ retrieval is respectful of people's autonomy in cases where they have not expressed a clear preference for donation. In these cases, there is no expressed preference that licenses care providers to intervene with a person's body without her consent.10

Gill argues that the fact that opt-out does not aim to secure people's consent to donation does not imply a failure to respect people's autonomy. I have argued that this claim is false: respect for autonomy implies that governments have a duty to implement a donor registration policy that aims to secure people's consent to retrieval.

Do opt-out policies secure people's consent?

Saunders questions an assumption that Gill concedes, namely, that opt-out policies do not secure people's consent to donation. Saunders argues that such policies do secure people's consent, provided that certain conditions are satisfied.4

Saunders first draws a distinction between express and implicit consent. The former occurs when people actively give their consent, for example, by signing an informed consent form or replying with an affirmative ‘yes’ to a request to touch their person.4 Implicit consent occurs when a person's inaction indicates consent; for example, when the chair of a meeting says that silence indicates acceptance of a motion, and one remains silent. As Saunders makes clear, express and implicit consent are forms of actual consent, in contrast to presumed consent.

Saunders argues second that under opt-out systems, those who have not opted out have implicitly consented to donate their organs, provided that (1) it has been clearly communicated that their silence shall be interpreted as consent; and (2) it is not unreasonably costly to opt-out.4 Saunders claims that most existing opt-out systems satisfy these conditions.4

Saunders is correct that implicit consent can be valid consent. However, if opt-out policies are to constitute a system of valid consent, a further condition must be satisfied: all must understand that their silence will be interpreted as consent. This condition, I argue, is highly unlikely to be satisfied in many jurisdictions.

To see why this further condition must be satisfied, it is helpful to consider the widely accepted account of valid informed consent. On this account, consent is valid if and only if five conditions are satisfied: (1) the person giving consent has the capacity to do so; (2) the relevant information has been disclosed to her; (3) her decision is voluntary; (4) she understands the relevant information to a sufficient extent; and (5) she gives consent.11 ,12 To satisfy (5), one must offer a token of consent—which can be silence in certain circumstances—and understand that one is offering a token of consent, that is, that one is publicly authorising a course of action.11 ,12 The argument for this claim is simple. To say that X gives valid consent to Y Φ-ing, where Φ-ing involves Y using X's person or property, is to say that X autonomously authorises Y Φ-ing.11 X can only autonomously authorise Y Φ-ing if X understands that she is authorising Y to Φ.

This is a problem for Saunders’ argument since even if the government satisfies his condition (1), that is, clearly communicates to all residents that their failure to opt out constitutes consent to donation, this is not sufficient for valid consent. It must also be the case that all residents understand that their silence constitutes a token of consent to donation. I am sceptical that governments of large pluralistic countries such as the USA or UK can achieve such a level of understanding, for example, by means of some sort of public information campaign.

Consider first that even in contexts highly conducive to the communication of information, it is often difficult to achieve understanding. For example, in the context of biomedical research, the informed consent process often involves staff carefully explaining the chief elements of an informed consent form to potential participants. However, misunderstandings among participants concerning basic features of trials—for example, risk and benefits of participation and the purpose of the trial itself—are widespread.13 Indeed the concept of the ‘therapeutic misconception’ has emerged within the research ethics literature to label a common form of such misunderstandings, namely, the failure to understand the differing imperatives of research and care and the false attribution of therapeutic intent to research procedures—for example, randomisation, blinding, determination of dosage and the use of placebos.14 My point here is not that participants who misunderstand in these ways are not giving valid consent—whether they are or not depends on how we interpret condition (4). Rather, it is that even in contexts highly conducive to communication, misunderstanding is often widespread.

Things are likely to be far worse in the context of public information campaigns.15 Consider the Affordable Care Act, a major policy innovation in the US healthcare system, the subject of endless debates within Congress and in a variety of media, and the subject of a massive public advertising campaign featuring celebrities and even President Obama himself. Still, the latest Kaiser Health Tracking Poll found that among the uninsured—those targeted by the law—only 56% are aware that the healthcare law creates health insurance market-places; only 42% are aware that the healthcare law provides financial help to low-income and moderate-income Americans; and only 44% say they have enough information about it.16 To return to the case of organ donor registration policies, although Saunders claims that most opt-out policies satisfy his two conditions, including the condition concerning clear communication of the policy, a recent Eurobarometer survey finds that only 28% of Europeans know the regulations in their country governing the donation and transplantation of organs.17 The numbers in European countries employing opt-out policies are similarly low: Austria 19%, Belgium 33%, France 33%, Italy 36%, Hungary 23%, Spain 21% and Sweden 33%.17

Saunders considers something like this objection, namely, that his position “involves a presumption that all are aware that silence will be interpreted as consent.”4 In response, he argues that even if some are unaware of this, it does not follow that their consent is not valid, “in keeping with the general position that ignorance of the law is no excuse.”4 Saunders’ point here seems to be that people need not understand that their silence constitutes a token of consent for their consent to be valid.

In response, consider first that this position directly conflicts with the widely accepted account of valid consent I reference above, according to which one must understand that one is giving consent for the consent to be valid. Consider second that it is not clear how Saunders’ appeal to the claim that ‘ignorance of the law is no excuse’ helps show that such understanding is not necessary. This claim is relevant to the question of whether a legal wrong has been committed or not; it's not relevant to the question of what constitutes valid ethical consent. Consider an example:Opt-out: Peter is a 45-year-old Austrian citizen. Austria has an opt-out policy, but Peter does not fully understand what this means. The Austrian government recently conducted a public information campaign informing citizens about this policy, but, while Peter saw some of the government ads, the issue is complex and he didn't pay sufficient attention. He is therefore unaware that his failure to opt-out means that he is registered as an organ donor. Unfortunately, he has a terrible accident that leaves him brain dead. Care providers try to contact Peter's family, but his parents are deceased, he is unmarried, and has no children or siblings. His preferences regarding organ donation are unknown. Under Austria's opt-out policy, care providers are authorised to remove Peter's organs and they do so.

The claim that ‘ignorance of the law is no excuse’ implies that Peter's care providers are not guilty of any legal wrongdoing and that Peter has no legal claim against them (were it possible for him to make such a claim). But, this claim does not imply that Peter gave valid ethical consent to the donation of his organs, that is, consent understood as autonomous authorisation, and it seems deeply counterintuitive to claim that he did since he did not understand his failure to opt out to be a token of consent.

Saunders might argue here that it is the legal perspective that matters since opt-out policies are enacted by governments. But, this response is not available to him. His claim is that opt-out policies secure people's ethical consent to donation, not that a failure to opt out can be construed as consent by a legal system.