Are patients morally justified in refusing?

In modern healthcare, significant value is placed on patient choice because of its connection to autonomy. Standard accounts of autonomy in medical ethics lay out three criteria which must be satisfied to say a patient is autonomous with respect to a decision.21 These criteria are: if they act intentionally, with sufficient understanding and free from controlling influence. Where a patient is offered a more environmentally friendly inhaler and they meet the aforementioned criteria with respect to that decision, should the patient decline, then that decision typically ought to be respected out of a concern for autonomy.

One widely accepted exception to autonomous decision-making in contemporary bioethics is, following John Stuart Mill, that this decision will harm others.22 The purpose of environmentally sustainable prescribing is to minimise the harms of climate change. If a patient’s refusal results in harm through climate change this presents a moral justification for thinking that a patient should choose an inhaler that does not contain GHGs. This has significant implications for how primary care approaches switching inhalers. If a patient’s refusal is outweighed by harm to others, the task for general practice is: to identify patients on MDI inhalers for whom a DPI is clinically appropriate, switch them, explain why they have been switched and ensure the patient can use a DPI.

The connection between individual GHG emissions and the harms of climate change are complex. This complexity has been summarised in two salient features of climate change: dispersion of cause and effect and fragmentation of agency.23 Fragmentation of agency refers to the fact that the actions of a single emitter do not alone cause global warming. Global warming is caused by many individuals and institutions acting in concert over time. Dispersion of cause and effect describes the complex relationship between GHG emissions and the climate system. The effects of global warming can be felt long after emissions were first produced and often in other parts of the world, meaning that cause is both spatially and temporally distributed from effect. Walter Sinnott-Armstrong has taken these features of climate change to undermine any individual moral responsibility to mitigate global warming. As an individual’s act of emitting will not alone cause global warming and because of complexity within the climate system through its distributed effects, Sinnott-Armstrong claims individual acts make ‘no difference’ to global warming and so cannot be said to cause harm.24 If this is correct and a patient using their inhaler makes no difference to global warming, then it seems difficult to use harm as a reason to dismiss their autonomous refusal.ii

One response to Sinnott-Armstrong is that if each instance of a GHG emission makes no difference to global warming, then how does global warming occur?25 A peculiar form of causation is required for each additional emission to make no difference and yet the accumulation of emissions to result in climate change. Sinnott-Armstrong cannot be correct that individual emissions make no difference, it is more accurate to say that individuals make minimal difference. Still, such a minor causal contribution to global warming that relies on the actions of others might still be viewed as not causing harm. Causal responsibility associated with using an MDI is not sufficient to underpin moral responsibility even if the difference it makes to global warming is more than zero. In response to this kind of argument, Avram Hiller distinguishes actual from expected harm.26 Stealing a patient’s inhaler such that their respiratory illness gets worse is an example of actual harm. Expected harm denotes events that increase the chance of somebody being made worse off thereby reflecting the statistical and cumulative nature of global warming. Hiller goes on to argue that individuals ought to minimise expected harm as compared with any other easily available alternative.

Even if MDIs do not directly cause harm, their significant global warming potential increases the risk of climate-mediated harms and so in this way is associated with expected harm. The question of how patients and practitioners ought to respond to the emissions of inhalers is important and cannot simply be dismissed as making no difference. Moreover, since there is another easily available alternative (DPIs) then it is pro tanto wrong to use an MDI. For some patients, a DPI is not clinically appropriate, so even though MDIs increase expected harm, a DPI is not an easily available alternative for the patient who is unable to use one. In the next section, I discuss morally justified exceptions. For now, as the expected harm of MDIs is greater than DPIs, and because DPIs are an easily available alternative for many patients owing to their comparable effectiveness to MDIs, a patient is not necessarily morally justified in declining a DPI. This suggests that patients ought to opt for a DPI where this is an easily available alternative, that is, just as effective at managing their respiratory condition.

How should doctors respond to patients’ refusals to switch inhalers?

It might be thought that if a patient is not morally justified in using an MDI over a DPI where the latter is clinically appropriate the response is straightforward: the practitioner simply over-rides the refusal and prescribes a DPI. The patient–practitioner relationship is more complex than this, however, and there are further morally relevant considerations.

One important factor is the impact of changing inhalers on trust. Trust is often seen as central to the doctor–patient relationship.27 Interpersonal trust can be described as an affective attitude towards another that we can in some sense rely on them.28 In healthcare, this translates into a suggestion that the doctor will act in the interests of the patient. Moreover, some suggest that there is an important connection between patient’s consenting and trust. Onora O’Neill argues that ‘Informed consent… is generally important because it can make a distinctive contribution to the restoration of trust’.27 This suggests two ways that over-riding a patient’s refusal might threaten trust. The first is that if informed consent is important for trust, then failing to respect a patient’s refusal simply undermines the distinct contribution informed consent makes to trust. The second is that a patient may feel they cannot rely on a doctor who simply disregards their rejection of an offer to switch. They may withdraw trust from the doctor who they see as failing in their fiduciary role to act in their interests as they seem primarily motivated by environmental concerns (or more cynically, NHS targets).

A second related issue is the patient’s interests. It is well established that doctors ought to act in the interests of their patients. There is evidence that suggests switching inhalers without discussion leads to reduced effectiveness and deterioration in asthma control through poor inhaler technique and patient dissatisfaction,29 30 not to mention as part of the proceeding issue of trust. If switching inhalers without a patient’s input worsens the patient’s health, there is a failure to promote their interests. Patients’ interests provide a weighty consideration against the pro tanto obligation to minimise expected harm. Nevertheless, this does not automatically lead to a patient being able to opt for an MDI because without it they will be worse off. We can distinguish changing treatments without discussion and without consent. The former is the subject of the aforementioned studies29 30 and predictably leads to a deterioration in care such that changes without discussion should be avoided. However, there are no studies of the effects of a change in treatment that a patient does not agree with but has been discussed with them. It is possible that non-consensual switches of this kind will result in poor disease control and if this is the case would outweigh the pro tanto obligation. Yet, there does appear to be opportunities to mitigate any potential risks through an appropriate safety net, planned follow-up and a proviso that an MDI can be represcribed. There is no expectation that healthcare professionals promote the patient’s interests at all costs, meaning that minimal and considered risks may sometimes be acceptable.

Trust and beneficence are important moral aspects of doctors’ management decisions in addition to environmental concerns. These important considerations represent morally salient reasons to over-ride a duty to choose options that minimise expected harm. As a pro tanto obligation, the need to minimise expected harm can, for the doctor, be over-ridden by a need to maintain trust or to protect their patient’s health. Moreover, as is discussed further in the third section, these aspects may be used to shape the available options to minimise expected environmentally mediated harms. Specifying exactly when these conditions will obtain is difficult. Nevertheless, as longer term relationships with patients remain common within primary care, practitioners ought to be capable of gauging where trust is being eroded. In terms of the effects on health, practitioners should weigh the interests of the patient against the expected harms with both parties being prepared to consider changing inhalers because of the expected harms without consent but following a discussion if it is thought that the risk to the patient is manageable and minimal. It may not always be easy to say exactly when the risk is minimal but the interests of the patient are a weighty concern and so any reasonable concern about worsening of the patient’s health will outweigh the pro tanto obligation.

Cost as a barrier to switching inhalers

One reason for hesitation in prescribing a DPI is cost. The issue of whether DPIs are more expensive than MDIs is complex. The most comprehensive study on this question essentially answers that it depends on which inhalers are being swapped.31 Changing MDIs to certain brands of DPIs could cost an extra £12.7 million annually for every 10% of MDIs changed. Targeted prescribing of the cheapest equivalent DPI instead of certain MDIs could lead to cost savings: £8.2 million annually for every 10% of MDIs. While some of these costs could cancel each other out depending on the changes, there is certainly the potential for DPIs to increase costs for the NHS. What makes the calculation especially difficult is accounting for wider effects of switching like reducing exacerbations of chronic respiratory illness and hospitalisations, which also carry a cost. The answer to the health economics of maximising DPI prescribing is far from settled. If it turns out that environmentally sustainable inhalers are costly this will pose a substantial barrier to change. I consider whether increased financial costs are justified.

Typically, we might think that the NHS is liable for higher costs where this brings about sufficient health benefits. As DPIs are just as effective as MDIs there is unlikely to be any additional direct health benefits to the patient. Nonetheless, there may be wider health benefits. Climate change mitigation carries health benefits so it might seem justified for the NHS to accept these costs. However, most of these benefits will be felt in low-income and middle-income countries. Any health benefits of climate change mitigation for the UK are likely to occur long after inhalers are changed. As climate change is disproportionately caused by wealthy countries and the harms felt most by the disadvantaged, this has led many to agree that wealthy countries ought to bear the burden of paying the costs of climate change mitigation.32 Nevertheless, it is not obvious that healthcare should accept such costs, even if it is a healthcare system in a wealthy country. We might ask whether it is fair for a healthcare system to accept higher costs to reduce its carbon footprint if this makes no direct difference to patients locally, instead producing spatially and temporally distributed benefits. This raises a question of distributive justice regarding the fair share of the benefits and burdens of tackling climate change for a publicly funded healthcare system. Clearly, this is a rather broad question, but it is highly relevant in considering the topic at hand. Space limits an in-depth discussion and hence my comments are necessarily brief and focused on inhalers. Nevertheless, some of the arguments may have broader relevance to discussions of healthcare’s role in mitigating climate change.

The question of how to share the burdens of climate change mitigation is an issue of ‘burden sharing justice’.33 34 Broadly, philosophers have discussed two principles of burden sharing justice to assess what a fair share is: a polluter pays principle and ability to pay. Each of these could be used to justify the NHS accepting higher costs for environmentally sustainable inhalers in the absence of a direct health benefit and so I will discuss each in turn.

‘Polluter pays’ is a straightforward principle that states that those causing global warming should pay the costs of fixing the problem. Henry Shue explains the reasoning behind this:

All over the world parents teach their children to clean up their own mess… If whoever makes a mess receives the benefits and does not pay the costs, not only does he have no incentive to avoid making as many messes as he likes, but he is also unfair to whoever does pay the costs.35

Unmitigated global heating will be paid for disproportionately by those who are vulnerable and contributed the least. This principle applies easily to the problem at hand: if MDIs contribute to global warming, then the NHS should pay the cost of mitigation including the cost of using more expensive inhalers.

Two issues arise with a polluter pays principle. The first is that it is not entirely obvious that the NHS is the polluter. Perhaps it is, but plausibly it could also be the patient using the MDI, the prescriber, the pharmaceutical company and so forth. If we are not convinced that the NHS is the polluter, it may be considered unfair that the NHS pays. Worse still, it is difficult to make the NHS as an institution pay because the funds for more expensive inhalers must be found either through the taxpayer or from elsewhere within the NHS. As it is patients or taxpayers who ultimately pay, we may be especially concerned by a disconnect between who the polluter is and who pays. This leads to the second concern. Some claim that ‘subsistence’ as opposed to ‘luxury’ emissions are an exception to a polluter pays principle. Shue argues that to treat all emissions as equal regardless of their purpose is to ‘ignore the fact that some sources [of greenhouse gas emissions] are essential and even urgent for the fulfilment of vital needs and other sources are inessential or even frivolous’.36 It might seem obvious that GHGs emitted in the pursuit of respiratory health are the very epitome of subsistence emissions and therefore the NHS is not liable for the costs. If health emissions are considered subsistence and therefore exempt, then MDIs are the obvious choice for they are just as effective and perhaps cheaper. The distinction starts to fall apart depending on how we interpret ‘essential’, however. Most would agree that health is a vital need, the problem is whether MDIs are essential when DPIs are also widely available and carry a lower carbon cost. MDIs are not essential to fulfil this vital need in the sense that we can achieve the same health ends with DPIs, but MDIs are essential if the costs are especially high in the context of a resource-limited NHS.

The concept of subsistence emissions seems to go too far in idealising any health-related emissions as essential. However, this does clarify two important insensitivities of a polluter pays principle. First is an insensitivity to the vital needs that healthcare fulfils generally. The second is the magnitude of the costs that healthcare might be forced to pay by accepting a polluter pays principle. Taken together, the problem is that simply saying that healthcare ought to pay to fix its own mess might force healthcare into accepting higher costs that ultimately limit its ability to meet its wider morally valuable goal of protecting and promoting health. If changing inhalers is very costly and healthcare pays because of a polluter pays principle, funds spent on climate mitigation cannot be spent on healthcare elsewhere creating the potential for opportunity costs. MDIs are then essential in helping a healthcare system meet a vital need with limited resources.

An ability to pay principle can be used to avoid these problems. Rather than polluters paying in proportion to their emissions, or whatever the costs, responsibility could be distributed based on an ability to shoulder the cost. 'Cost' is usually conceived of as financial costs, so the wealthy pay. One might object at this point that ability to pay does not take us any further. It is well known that the NHS is underfunded and struggling to meet demand after years of austerity, and now a pandemic. When one looks to the wealthy to pay the bill of mitigating climate change, the NHS hardly features highly. If ‘ability’ means spare or large amounts of wealth, it seems the NHS does not have the ability to pay for more expensive environmentally sustainable inhalers given the current political context.

This interpretation sets the bar too high on ‘ability’. Rather than ability, meaning that one is easily able to pay to mitigate climate change, it could have a more moderate reading. To say one has an ability to pay means that one has sufficient resources to mitigate climate change without having to sacrifice too much of comparable moral importance. In this way, ability to pay has much in common with a ‘duty of easy rescue’.37 This states that if the cost of engaging in some activity is small but the harms averted great, then one is obligated to act. A more moderate ability to pay principle suggests that the NHS has a duty to pay the costs of mitigating climate change insofar as the costs are not too significant given the grave nature of unmitigated climate change. In the overall NHS budget, the resources needed to pay for DPI devices are not particularly burdensome based on the calculations from Wilkinson et al.31 Even if they were higher, the NHS should be prepared to accept these costs given the importance of preventing global warming insofar as this does not significantly threaten the NHS ability to protect and promote health.