Should Patients be Deprioritised for Self-Inflicted Illnesses?

When, if ever, is it fair to deprioritise treatments for people whose ill health is ‘self-inflicted’ (e.g. brought on by poor diet or heavy drinking)?

The inherent scarcity in any healthcare system forces rationing choices, whether in prioritising expenditure in a limited budget, or distributing scarce transplantable organs. When these decisions must be faced, culpability for illness is often raised as a factor in prioritisation. News articles protesting limited funding for expensive treatments reference cosmetic treatments, smokers, and obese people, implying people in those groups shouldn’t be treatedwhile other patients – presumed to be less culpable – are not having their treatments funded[1]^,[2]. The arguments surrounding whether a patient’s need for treatment is ‘self-inflicted’ focus on two factors. The first is causal certainty – the degree to which we can be sure the patient’s need for treatment results from their behaviour. The second is choice autonomy – how far the patient engaged in those behaviours deliberately and autonomously and the part environmental, genetic, and socioeconomic factors played. I will explore causal certainty and choice autonomy through comparing skin cancer, obesity and Alcoholic Liver Disease. Examining the NHS approach to treating these varying conditions demonstrates a degree of moral incoherency in treatment prioritisation, with punitive, clinical, public health and budgetary outcomes entangled. After establishing the degree to which a condition is ‘self-inflicted’, I will propose alternatives for effecting behaviour change to improve health outcomes that are less coercive than blanket deprioritisation, and conclude that under most conditions it is not fair to deprioritise treatment for patients whose ill-health is ‘self-inflicted’. However, under conditions of absolute scarcity, for example liver transplants for Alcoholic Liver Disease, where any patient not assigned a liver will die, prioritising a patient whose condition was ‘self-inflicted’ could cause significant direct harm to another patient. I will propose a method of deprioritisation to operate as a tie-breaker in limited situations where deprioritisation of some patients might be the least unfair option.

If you need help with a dissertation project for a healthcare or nursing subject take a look at our nursing dissertation writing service. Alternatively, for non-healthcare subjects you can get dissertation help from our siter website UKDiss.com.

In order to clearly define a condition as self-inflicted, a degree of both causal certainty and choice autonomy must be present. Causal certainty, the knowledge that the patient’s behaviour was the direct cause of the condition, is hard to establish. While with, for example, non-melanoma skin cancer, sun exposure is the usual cause, it is possible to get an identical non-melanoma skin cancer without any sun exposure on areas of the body that are always covered. This means that while the balance of probabilities suggests sun exposure was the trigger in a patient with a history of sun-bathing, we cannot establish an absolute causal certainty. In contrast, with Alcoholic Liver Disease, causal certainty can be established. Blood tests can differentiate between ALD and other forms of cirrhosis of the liver, and ALD can only arise in the context of excessive drinking over a long period. Buyx argues that “we want to be sure that we know exactly what actions or behaviours lead to a certain condition before holding patients responsible for the consequences”[3], and where causal certainty cannot be established, it is difficult to justify deprioritisation.

Choice autonomy explores whether the patient had agency and autonomy in the decisions that led to the ‘self-inflicted’ condition. There is very strong evidence that environmental and social factors play a large part in the development of ‘self-inflicted’ conditions, and their progression. Womack identifies that obesity is higher in poorer populations[4], and similarly alcohol-related mortality is higher in more deprived areas[5], while smoking is four times higher amongst the most disadvantaged[6]. Marmot noted in ‘Whose Responsibility’ that levels of harmful drinking are higher in people from higher socioeconomic classes, while alcohol-related harm is more common in people from lower socioeconomic classes[7]. This indicates that class provides a buffering effect, lowering the extent to which harmful drinking actually harms the patients from higher socioeconomic classes.

Walker argues that where healthcare needs rise from an individual’s free choice, society owes them no responsibility[8]. He might apply this to the patient with skin cancer and excessive sun exposure, but he adds that addiction may raise “serious concerns about the extent to which the individual’s choices are voluntary”. While developing an addiction relies on repeated intake of an addictive substance, substance use often starts in childhood – before one necessarily had the ability to project the future consequences of developing an addiction. People who began drinking regularly at under age eighteen were twice as likely to report at least one serious alcohol-related problem compared to people who began drinking regularly at older than eighteen[9], which obviously affects the degree to which the patient can be considered responsible for their situation.

Even in the context of addiction there are points at which people could be expected to take, or have taken, responsibility, and these points illustrate some degree of choice autonomy. In the adult with capacity, they could have avoided their need for treatment for alcoholism by not starting to drink alcohol. Drinking alcohol is the first step towards developing alcoholism, and most adults understand the risks and chose to begin drinking anyway. Glannon argues that this makes alcoholics causally responsible for not having acted to avoid alcoholism, and morally responsible for increasing demand for limited medical resources[10]. Another point of responsibility is when the patient is diagnosed with alcoholism, and thus warned that they had an illness they must learn to manage. Moss and Siegler take the perspective that the alcoholic is responsible for their behaviour once they have the condition known as alcoholism, just as the diabetic is responsible for managing their sugar consumption and insulin[11]. This is coherent with the idea common across transplant units that the alcoholic must stop drinking for six months before being listed for transplant, and is a more forward-facing perspective, allowing for the alcoholic to come to understand their condition and change their behaviour.

The impact that socioeconomic class, environment, and initiating age have on addiction is considerable and makes it impossible to conclusively establish that full choice autonomy was present with any addiction. These factors must be given considerable weight when discussing choice autonomy, or we risk punishing patients twice – first by creating the environment that encouraged the development of the condition, and secondly by calling it ‘self-inflicted’ and refusing to treat it. Voight argues that “the more difficult or painful it is for an individual to make a best choice less reasonable it is to expect that she will make that choice”[12]. This allows us to develop a partial concept of responsibility, acknowledging that patients from different backgrounds might have had access to different resources to help them avoid developing these conditions, and different access to treatment.

The NHS Constitution establishes responsibility at its heart, asking that patients “recognise that you can make a significant contribution to your own, and your family’s, good health and wellbeing, and take personal responsibility for it”[13]. This emphasis on personal responsibility for behaviours is an important encouragement in a system where healthcare is publicly funded and subject to budgetary restrictions. However, historically a non-punitive approach has been taken which recognises the multi-factorial causes of many conditions, thus the emphasis on personal responsibility has been encouragement rather than conditionality.

However, recently this has been taken further, and several Clinical Commissioning Groups (CCGs) ration and deprioritise treatment for smokers and for overweight and obese people[14]. This is a replication of Feiring’s argument that “the obese patients suffering from X should not be held fully responsible for her situation at the sickbed and be denied treatment. Assigning longer waiting time for treatment of obese patients than their normal weight counterparts may be one way of holding people partly responsible for their lifestyle choices”[15]. This is an active deprioritisation for people whose conditions, as we have established already, may lack causal certainty and are likely to lack full choice autonomy also.

The messaging around the justification for delaying and denying treatment is inconsistent. It contains a mixture of appeals to personal responsibility and behavioural change, and clinical arguments, which range from being described as “weak and unclear”[16], to “potentially justifiable”, although with evidence gaps[17]. The Royal College of Surgeons’ guidance opposes blanket weight or BMI limits for surgery and suggest that these restrictions are being brought in not to improve outcomes, but to save money[18]. While there may be very strong clinical reasons for, in the case of some obese people and smokers[19], insisting on weight loss or quitting smoking, this should be an individual clinical decision[20]. The goal of “freeing up limited NHS resources for priority treatment”, as stated by East and North Hertfordshire CCG[21], should not require the wholesale banning of treatment for people with specific health conditions, especially where it cannot be proven that the patient bears either causal or autonomous responsibility.

The proposals used by Vale of York NHS trust were to delay smokers by up to six months while they tried to quit – but then to proceed with surgery. This suggests that the concern wasn’t the clinical outcome of the surgery, but to try to use the threat of a delayed surgery to persuade smokers to quit[22] – and thus to serve as a warning to other smokers that this could be applied to them also. It serves the public health goal of signalling that access to elective treatments for patients will be limited if they are obese or smokers. This may confer a stigmatising effect as encouraged by Callahan, who argued for “social pressure” to be “brought to bear” against the obese, as it was smokers[23]. By arguing for cutting expenditure on surgeries for obese people and smokers, they are making it unacceptable to be obese or a smoker and require medical treatment[24].

Alternative, less coercive options for improving public health and incentivising health behaviours exist. Womack argues that public health has a powerful toolset for responding to obesity, and proposes using that to change default options and restrict choice of more unhealthy foods[25]. Alongside that can run a series of taxation options – such as the sugar tax recently implemented in the UK, which is believed to have worked to reduce the sugar content in soft drinks, although whether that will have the desired health implications is still unknown[26]. Azevedo and Vartanian highlight the risk that taxing certain foods will disadvantage the economically disadvantaged, and propose prioritising interventions that alter the physical environment and that focus on encouraging healthy behaviour. While flat taxation without subsidy risks punishing poorer people more than the wealthier – who can absorb a price increase and continue to purchase their preferred option – it is still less coercive than deprioritisation[27].

There are numerous reasons not to deprioritise patients whose conditions are deemed self-inflicted, in most scenarios. Given the known importance of genetic, socioeconomic and environmental factors on the development of obesity, alcoholism, and nicotine addiction, it would be difficult to define exactly how much of the development of the condition someone was responsible for, both in terms of causal certainty and choice autonomy. Even with skin cancer, whose cause can often be established on the balance of probabilities and where there are not concerns of addiction impinging on choice autonomy, the decision to deprioritise those patients wouldn’t prevent harm to other identifiable patients and thus cannot be justified. Where deprioritisation is a result of clinical evidence – for example improved outcomes in patients supported to lose weight – it is justifiable. However, where it is a cost-cutting exercise it is not fair on a patient to deprioritise them for something where the extent to which they were responsible for it was doubtful, the clinical evidence for deprioritising them was doubtful, and there was no compelling reason to deprioritise that patient in order to treat another who had no culpability in their development of a condition

There is one scenario which I believe to be qualitatively different to an extent such that it should be approached as a different question, and that is liver transplants in people with Alcoholic Liver Disease. These exist in a different context to other ‘self-inflicted’ conditions like obesity. When a patient needs a liver transplant, there is no alternative (such as kidney dialysis), the patient either receives a liver or they die. Due to the scarcity of livers, patients frequently die without receiving a needed transplant. Listing patients with ‘self-inflicted’ conditions for transplant increases the number of patients competing for a limited resource, and thus deprioritising some of those patients under specific circumstances may be the least unfair option. It is under these conditions of scarcity that this moral argument is developed. If supply could be increased to eliminate the organ scarcity, then the arguments in favour of deprioritisation would no longer apply, but methods for increasing supply are beyond the scope of this essay.

With Alcoholic Liver Disease, causal certainty has been established – we know that excessive intake of alcohol caused the condition to develop. We also know that while choice autonomy is clearly impeded, there are the points prior to developing alcoholism and once diagnosed with it where the patient could have mitigated their need for treatment. Causal certainty plus a degree of choice autonomy suggest that the condition is ‘self-inflicted’, and that the patient bears some moral responsibility for that.

In order to justify deprioritising these patients we must not just consider that some people might suffer some harm – but that existing people are suffering harm to the extent that it overrides the reasons against deprioritisation outlined earlier – such as the influence of genetic, environmental, and socioeconomic factors on the development of alcoholism. Where there are limited livers available and one patient receiving one could cause another identifiable patient direct harm equating to death, the fact that one patient has a degree of responsibility in their requirement for a transplant while another doesn’t may become relevant. The ‘harm argument’ outlined by Sharkey and Gillam suggests that patients whose conditions were not self-inflicted would experience harm if patients whose conditions were self-inflicted were treated ahead of them[28]. Where that harm could be mitigated by increased funding, it would not be fair to deprioritise patients, but where the harm cannot be mitigated, deprioritising might be the least unfair method of treating these patients.

Decisions about transplantation are currently made on purportedly clinical criteria. The Model for End-Stage Liver Disease (MELD) criteria are the most commonly used. These assign livers to patients based on the urgency of the patient’s need for a liver. Typically they are applied with a requirement of a period of six months completely abstinent from alcohol before listing patients for transplant[29]. This serves multiple functions – some clinical, including allowing the liver to heal which might avoid a need for transplant and allowing the patient to join a structured treatment programme. Other functions are more punitive – the abstinence criteria functionally exclude patients with Severe Alcoholic Hepatitis whose lifespan without transplant is measured in weeks from the transplant list. During the period of required abstinence, patients are typically required to sign a contract that says they will not drink any alcohol and are willing to submit to blood and urine tests.  Anyone who drinks any alcohol is permanently refused listing for transplant, or is permanently delisted[30].

These criteria do not have a strong evidence base. The requirement for six months of abstinence arose from three poorly controlled studies[31] and is, in its details, difficult to defend as a clinical criteria. While less than 6 months abstinence pre-transplant was a significant factor associated with harmful relapse[32], there is also an association between post-transplant relapse and less than 12 months of sobriety pre-transplant[33].

Subjecting the decision on whether to offer alcoholics a transplant to apparently normative grounds, when those grounds cannot be defended on a closer look at the evidence, is a handing-off of moral responsibility which seeks to evade choosing and defending a position openly, instead obfuscating. By requiring a period of absolute abstinence the architects of this policy demonstrate that they believe the alcoholic has the ability to choose not to drink, and to understand the long-term consequences of being delisted were they to drink. This fits with the position of Moss and Siegler that an alcoholic once diagnosed has the obligation to stop drinking if they wish for access to transplant[34], and with Feiring’s forward-looking conception of taking responsibility for future actions[35]. This also demonstrates that responsibility is already incorporated into the liver transplantation process, albeit in an inflexible way that (after the six month sobriety test for listing) either permanently delists the ALD patient (if they have a single drink) or treats them the same as any other patient (if they remain sober).

Veatch suggests incorporating responsibility instead through a reduction of the MELD score for alcoholic patients, which would serve to deprioritise the alcoholic patient marginally compared to non-alcoholic patients[36]. This would not lead to a fairer outcome, as it would undermine the function of the MELD score of minimising mortality without leading to an increase in utility, or even a guarantee that the survival of patients without ALD would increase, as livers might be given to patients who were less suitable for those organs, solely because those patients weren’t alcoholics. Given the undersupply of transplantable livers, this could also serve to absolutely deprioritise patients with ALD as there will always be more patients requiring livers than there are livers. Glannon describes this as a tiebreaker – but it doesn’t identify a tie, let alone break one[37].

I propose instead an actual tiebreaker based on the UK’s new Transplant Benefit Score model[38]. The new model prioritises transplants for patients on the ‘super-urgent list’ (people who will not live long without an urgent transplants), before modelling based on donor and recipient who is a good match for the liver, who will live the longest with a transplant and who can survive the longest awaiting one in order to maximise utility. It incorporates significant psychosocial assessment to identify whether alcoholic patients are likely to return to alcoholism, but is willing to list patients with less than 6 months of sobriety where this is necessary. The tie-breaker would operate only if:

1. Two patients are equivalent matches for a liver based on donor and recipient criteria
2. They have anticipated outcomes within a broadly similar ballpark (this should be clinically defined but should be strict)
3. One of these patients has alcohol-related liver failure and the other patient does not
4. The patient with alcohol-related liver failure has passed all the psychosocial testing and is deemed a good candidate for transplant
5. There are no extenuating criteria in the development of the alcohol-related liver failure that might limit the degree of responsibility the patient holds, or demonstrate that the patient had fully taken responsibility. Extenuating criteria might include: patient lacking in capacity, patient developing alcoholism as a child or young teenager, patient having stopped drinking entirely many years prior to diagnosed liver dysfunction

If the tie-breaker is operative, the liver should be given to the non-ALD patient ahead of the ALD patient. This restores non-ALD patients their position on the transplant list only where this can be done without a significant increase in mortality or decrease in utility. It ensures that punitive deprioritisation imposed as a blanket policy in a way that would overall prevent ALD patients from receiving transplants. It is sensitive to individual circumstances in that it only comes into play where two patients are in similar levels of need, and allows room for extenuating circumstances that might limit the responsibility of the ALD patient for their situation. Deprioritising ALD patients in a way that might decrease utility or increase mortality would not be fair, and nor would a blanket policy of deprioritising those patients. However, it would also not be fair for patients who had no way of avoiding their need for a liver transplant to die awaiting a transplant when someone who could have avoided their need for transplant was given the organ instead. Balancing these two competing perspectives, it seems least unfair to deprioritise ALD patients only where it is a life-or-death choice between the two patients.

With most conditions deemed ‘self-inflicted’ the arguments against deprioritisation clearly outweigh the arguments in favour. When neither causal certainty nor choice autonomy can be definitively established, and socioeconomic and environmental factors often heavily influence the degree of responsibility the patient has for their condition it cannot be fair to deprioritise the patient. However, there are scenarios that differ, and the clearest of these is regarding liver transplants for patients with ALD. As offering a transplant to one patient may be a death sentence to another, a settlement must be reached that manages the impact of patients whose conditions are ‘self-inflicted’ on other patients. Transplanting livers solely to patients without alcohol-induced conditions would be excessively punitive given the socioeconomic and environmental factors affecting addiction. On the other hand, ignoring the fact that some patients could have avoided their need for transplant while others had no choice is unfair on those patients whose liver failure was not self-inflicted. The imperatives to maximise survival and utility whilst developing a fair distribution model that recognises this are in tension, and my solution highlights that maximising survival and utility is the higher moral priority. Deprioritisation should only occur when it doesn’t interfere with this, hence proposing that responsibility serve as a tie-breaker between patients with broadly similar anticipated outcomes. Any attempt to move beyond that into an overall deprioritisation presumes too much voluntary capacity in the development of alcoholism and prioritises punishment over optimum utility. It is important that we do not cross that moral rubicon, and that we approach the question of priority-setting with a humility that recognises we are doing the best we can with the resources available to use – and that much about the neurological and genetic basis for addiction is yet unknown. This demands an open-minded gaze at the evidence, and a willingness to alter our position should new evidence come to light that indicates that addiction is more or less voluntary than we had presumed.

References

[1] Devlin, ‘Kidney Patients Denied “too Expensive” Life-Extending Drugs’.

[2] Martin, ‘Betrayal of 20,000 Cancer Patients: Rationing Body Rejects Ten Drugs (Allowed in Europe) That Could Have Extended Lives’.

[3] Buyx, ‘Personal Responsibility for Health as a Rationing Criterion’.

[4] Womack, ‘Public Health and Obesity’.

[5] Erskine et al., ‘Socioeconomic Deprivation, Urban-Rural Location and Alcohol-Related Mortality in England and Wales’.

[6] Hiscock et al., ‘Smoking and Socioeconomic Status in England’.

[7] Marmot, ‘Whose Responsibility?’

[8] Walker, ‘Who Do We Treat First When Resources Are Scarce?’

[9] Morleo, Jones, and Bellis, ‘Alcohol Age of Initiation and Long-Term Impact’.

[10] Glannon, ‘Responsibility and Priority in Liver Transplantation’.

[11] Moss and Siegler, ‘Should Alcoholics Compete Equally for Liver Transplantation?’

[12] Voigt, ‘Appeals to Individual Responsibility for Health’.

[13] ‘The NHS Constitution for England’.

[14] Pillutla, Maslen, and Savulescu, ‘Rationing Elective Surgery for Smokers and Obese Patients’.

[15] Feiring, ‘Lifestyle, Responsibility and Justice’.

[16] Womack, ‘Reviewing the Evidence for Restricting Elective Surgery for Obese Patients’.

[17] Pillutla, Maslen, and Savulescu, ‘Rationing Elective Surgery for Smokers and Obese Patients’.

[18] ‘Smokers and Overweight Patients: Soft Targets for NHS Savings’.

[19] ‘Joint Briefing: Smoking and Surgery’.

[20] Dehn, ‘Joint Replacement in the Overweight Patient’.

[21] Donnelly, ‘NHS Provokes Fury with Indefinite Surgery Ban for Smokers and Obese’.

[22] Campbell, ‘Surgeons Attack Plans to Delay Treatment to Obese Patients and Smokers’.

[23] Callahan, ‘Obesity’.

[24] Gillett, ‘Obese People and Smokers Banned from NHS Surgery in Cost-Cutting Bid’.

[25] Womack, ‘Public Health and Obesity’.

[26] https://www.bbc.co.uk/news/health-43372295

[27] Azevedo and Vartanian, ‘Ethical Issues for Public Health Approaches to Obesity’.

[28] Sharkey and Gillam, ‘Should Patients with Self-Inflicted Illness Receive Lower Priority in Access to Healthcare Resources?’

[29] Bramstedt and Jabbour, ‘When Alcohol Abstinence Criteria Create Ethical Dilemmas for the Liver Transplant Team’.

[30] Bramstedt and Jabbour.

[31] Singhvi et al., ‘Ethical Considerations of Transplantation and Living Donation for Patients with Alcoholic Liver Diseases’.

[32] Pfitzmann et al., ‘Long-Term Survival and Predictors of Relapse after Orthotopic Liver Transplantation for Alcoholic Liver Disease’.

[33] Gedaly et al., ‘Predictors of Relapse to Alcohol and Illicit Drugs after Liver Transplantation for Alcoholic Liver Disease’.

[34] Moss and Siegler, ‘Should Alcoholics Compete Equally for Liver Transplantation?’

[35] Feiring, ‘Lifestyle, Responsibility and Justice’.

[36] Veatch, ‘“Just Deserts?” A Response to Daniel Brudney’.

[37] Glannon, ‘Responsibility and Priority in Liver Transplantation’.

[38] ‘New System for Allocating Livers for Transplant Is Now Live’.