When it comes to determining how healthcare resources should be allocated, there are many factors that could—and perhaps should—be taken into account. One such factor is a patient’s responsibility for his or her illness, or for the behavior that caused it. Should we take into account, for instance, the fact that a patient’s lung cancer is the foreseeable result of a lifetime of smoking, or that a patient’s heart disease is a foreseeable result of a lifetime of unhealthy eating habits? Call policies that take such considerations into account Responsibility-Sensitive Healthcare Policies, or RSHPs for short. On such policies, patients would, to some extent or other, be held accountable for their unhealthy lifestyles and/or their outcomes.
Some defend RSHPsFootnote 1; while others reject them. One prominent line of attack gives reasons for thinking that patients sometimes, or perhaps always, lack responsibility for their health-related actions, or their outcomes, in virtue of lacking the requisite sort of control over them.Footnote 2 Another worry is that patients might lack the sort of awareness of their actions, and their consequences, required to be responsible for these lifestyles and their negative health outcomes.Footnote 3 Further, even if we suppose that some patients can and do have the requisite control and awareness over their actions and their outcomes, determining whether any particular patient has them is—some hold—either something that we cannot do, or involves doing things that we should not do (for example, being objectionably intrusive, or undermining the doctor-patient relationship).Footnote 4
For the purposes of this paper, we will assume that human agents can sometimes freely and knowingly engage in health-risking lifestyles. Because of this, they can be responsible for having some of these lifestyles, and for some of the health outcomes of these lifestyles. However, we will also assume that, when a patient’s lifestyle is first discovered or suspected by the healthcare system, healthcare authorities are typically not in a position to know whether this particular patient is responsible for the illness that may be a result of her previous behavior, nor whether she met conditions on responsibility for the behavior itself. This may be because the healthcare authorities do not have enough evidence about the patient’s behavioral and medical history, and perhaps because gathering such evidence would be impractical or undesirable.
Even when making this second assumption, one might still be justified in taking responsibility into account when making healthcare decisions. This is because there are other ways of attaining knowledge about a patient’s responsibility for their unhealthy behavior. How might this be done? By implementing a policy of presenting patients with what we will call Golden Opportunities (GOs): opportunities to make a change to a healthier lifestyle at reasonable cost, and accompanied with appropriate support, should the patient choose to attempt the change. We elaborate on these features below, but some examples of possible GOs include offering, with appropriate support, e-cigarettes to smokers, methadone replacement therapy to heroin addicts, and nutritional or exercise programs to individuals at risk of type 2 diabetes or arterial disease (Savulescu 2017).
The reasons often given for uncertainty about the patient’s responsibility for her lifestyle stem from obstacles the agent may face in changing her lifestyle. Some patients may face obstacles to making a lifestyle change which reduce, or eliminate, their responsibility for the lifestyles in question. Given that we may not know, of a particular agent, whether, or to what extent, she faced such obstacles, we may not know whether she is responsible for her lifestyle. One of the aims of GOs is to help patients avoid as many of these obstacles as possible.
Once a patient is presented with a GO, we are in a significantly better position to determine the patient’s responsibility for at least some of her health-relevant choices—namely, whether to take or refuse the opportunity, and (at least for a period) whether to stick to the healthier lifestyle that she is offered. We also thereby remove some of the obstacles to implementing an RSHP. If the patient is responsible for the choice, and her choice was to decline the GO, we may then be justified in holding her responsible for that choice—and for whether she subsequently sticks to it—even though we are still not justified in holding her responsible for her behavior prior to the offer (Savulescu 2017). It is also plausible that, as others have argued, once the agent is responsible for declining the GO, “she cannot reasonably complain to being given lower priority on the waiting list” (Feiring 2008, p. 35) and “should be considered to have a weaker claim on resources than others who either did comply or never had a high-risk lifestyle in the first place” (Bærøe and Cappelen 2015, p. 838).
In order to determine whether GOs can indeed empower individuals, and reliably justify belief in the claim that the patient is responsible for some of her health-relevant choices and behaviors, we will need a fuller account of what a GO is than has been provided to date. In this paper, we develop the notion of a GO. Our main goal in doing so is to show how GOs might reliably help patients overcome obstacles to making the relevant change, and thus provide justification for the belief that such patients are responsible for some of their health-relevant choices and behaviors.
Before moving on, we wish to make some clarifications.
First, we should clarify what we mean by “responsibility.” This term is used in different ways in this debate. For instance, some speak of individual responsibility, others of personal responsibility, others of moral responsibility, and other still of “responsibility,” unqualified. We do not wish to take a stance on which of these notions is the relevant one in this debate, and simply stick to the general “responsibility.” What is important for our purposes is that responsibility has a control, and perhaps an epistemic, condition. The different sorts of responsibility often mentioned are often thought to have one, or both of these conditions.
However, there is also a different way of distinguishing between “types” of responsibility. Sometimes people contrast forward-looking, or prospective, responsibility with backward-looking responsibility. It seems to us that these terms are not used uniformly, though they are often intended to mark a difference similar to what we will call the obligation-sense and the accountability-sense of responsibility. Consider the claim that a doctor has a responsibility to his patients. Here, “responsibility” seems to refer to some sort of obligation or duty that a doctor has to his patients. Call this the obligation-sense of responsibility. Contrast this with the claim that John is responsible for failing to stop at the red light. Here, “responsibility” is not being used to refer to some obligation that John had; he certainly did not have an obligation to fail to stop at the red light. Here, we mean that John had the sort of control over his behavior, and awareness of (or capacity to be aware of) the relevant features, such that responding to his failure in a certain way is now appropriate. Call this the accountability-sense of responsibility.Footnote 5 The sort of responsibility we are concerned with in this paper is this accountability-sense; GOs help to justify the belief that an agent is accountable for his choices and behavior, and RSHPs hold people accountable for their choices, behavior, and perhaps their outcomes.
Second, a patient offered a GO could be responsible for a variety of things: (1) the choice to accept/reject the offer, (2) the individual actions and omissions that constitute the lifestyle (new or old), (3) the lifestyle (that is, series of actions and omissions) itself, and (4) the health outcomes of the lifestyle. Our first assumption – that agents sometimes freely and knowingly engage in health-risking lifestyles – implies that individuals can be responsible for 1–3, and strongly suggests that they can be responsible for 4. For the majority of this paper, we will be concerned with patients’ responsibility for their choices after being presented with the GO (1), and for their ensuing lifestyles (2, 3), not their responsibility for the negative health outcomes of that behavior (4). When we speak of “the patient’s responsibility for her lifestyle,” or of her “responsibility for accepting or declining the GO” we intend to speak of the patient’s responsibility for one or more of 1–3, but not 4.
In taking this focus we do not mean to imply that agents are not, or cannot be, responsible for the outcomes of their lifestyles; nor do we mean to imply that a healthcare policy which held patients accountable for these outcomes would be unjustified. The main reasons for our focus on 1–3, as opposed to 4, are twofold (we elaborate on these further below). First, GOs can result in patients’ having the requisite control and knowledge over their choice and actions; yet having the requisite control over the outcomes of their lifestyles is a different matter, since there are various other factors involved which influence whether the lifestyle will result in a given outcome. Thus, GOs may not be as good at providing agents with the requisite control over the outcomes of their lifestyles. Second, given this point, and some further difficulty in acquiring evidence about the causal factors leading to an illness, we think the epistemic situation concerning responsibility for the outcomes of lifestyles will be a much more complicated matter.
Third, we wish to clarify that our main focus will be on GOs as a component of an RSHP, intended to solve some problems faced by such policies. In section 3, we suggest how such a policy can make use of features of GOs; yet we do not intend to develop a full version of such a policy. Further, we do not mean to imply that offering GOs would be desirable only insofar as it helps to justify such a policy; even if one thinks that RSHPs are always unjustified, one would still have reasons to implement a program of GOs, insofar as one wants to empower individuals to make changes in their lifestyles.Footnote 6
Fourth, when we discuss RSHPs, we will be focusing on policies that make use of the knowledge provided by GOs, and are sensitive to responsibility for a lifestyle after the offer. Our discussion will be limited to considering the possibility that the responsibility of patients offered GOs might be one factor among others that determines the priority given to a patient, or it might modestly affect the proportion of treatment costs that the patient will be required to bear for health outcomes related to the lifestyle, or it might modestly increase the cost of premiums.Footnote 7 When we speak of “holding the patient responsible,” we are referring to, and only to, these practices.Footnote 8
We will not consider or defend a policy that recommends the absolute exclusion of some patients from the health care system or from public funding for healthcare. A policy of absolute exclusion from treatment or complete denial of funding, solely on the basis of responsibility, would face a host of objections which we do not address in this paper, nor are we convinced that they could be addressed adequately (Cohen and Benjamin 1991; Shiu 1993; Anderson 1999; Feiring 2008; Segall 2009).
Finally, a point about the sorts of health-related lifestyles we have in mind. RSHPs face what is sometimes called the Universalization Objection. Proponents of this objection argue, roughly, that holding people responsible for unhealthy lifestyles would require that we do so for many more lifestyles than are usually considered, and this would lead to absurd results.Footnote 9 For instance, if we are to hold smokers responsible for smoking then, to be consistent, we should also hold responsible a physician who contracts Ebola while participating in a Doctors Without Borders project in an area where Ebola is prevalent; after all, both the smoker and the physician knowingly and willingly engaged in behavior that incurred risks to their health. But holding the physician responsible, say, by increasing the proportion of treatment costs that he is required to bear, would be absurd. The challenge is to delineate those health-related lifestyles which it would be appropriate to hold people responsible for in a way that avoids these implications.
For our purposes, we will remain neutral on how best to meet this challenge, focusing instead on paradigm cases of the sorts of lifestyles that are typically considered in this debate. However, one of us has addressed this objection in the past when introducing the original account of GOs (Savulescu 2017). The suggestion there was that one major factor will be whether the risks incurred by the lifestyle are unreasonable. The physician who contracts Ebola did not incur an unreasonable risk, whereas the smoker, at least once she has been presented with a GO, does incur an unreasonable risk by continuing to smoke. Determining whether a risk is reasonable or not will depend on a variety of factors. For example, the objective values of the costs and the benefits, the likelihood of the potential costs and benefits, whether there are means of minimizing likelihood of costs while maintaining benefits, etc. (Savulescu 2017).Footnote 10
In the next section, we further develop the notion of a GO, offering a general framework for what a GO program would look like, and what it should aim to achieve if it is to form a component of an RSHP. In section 3, we make use of some of the discussion of GOs to suggest some desiderata for an RSHP policy that makes use of GOs. In section 4, we discuss areas for further development. We then offer some concluding thoughts on how our discussion fits into the broader context of the debate on RSHPs.