When push comes to shove, ethical decision-making and the last ventilator...

We seem to be stuck in a bit of a rut discussing pandemic planning issues at the moment, but the issues are pressing and they warrant discussion and debate. Hamilton Health Sciences (in Hamilton Ontario) has just published its triage guidelines which will come into effect if the hospital gets overwhelmed by patients seeking critical care beds – or ventilators, when the supply of such resources is grossly inadequate to meet the demand. (The complete protocol is published in the September issue of “Health Care Quarterly” a short op-ed piece appeared in the Globe and Mail, Sept 21st) The initial stages of triage are relatively unproblematic; a team of experts will assess the patient’s clinical condition and his or her prospects for survival. So far so good, that the patient is likely to benefit has to be the first criterion for access to any treatment. The elements of that assessment include: “1) Does the patient need critical care? 3) Is more than one organ system affected? 4) Does the patient improve with critical care?” (Globe and Mail Sept 21st numbering added) These are the easy criteria, but let us pause for a moment to look at the elements in turn.

The first stages of triage

1) Does the patient need critical care?

This is completely uncontroversial; patients should not be in critical care if they do not need it (and this is as true outside of a pandemic as it is within one.)

2) Does the patient have underlying illnesses that predict a poor outcome?

Taken one way this is also uncontroversial. For instance, if the patient has an underlying disease process that will imminently kill the patient regardless of the success of the treatment for the flu and its aftermath, then treating the flu is futile. (For more on futility see the post June 4th below.) If the patient’s underlying condition means that the flu treatment is unlikely to be successful then again the treatment is very likely to be futile. However, what happens if the patient’s underlying condition is such that he or she will always be bed-bound or ventilator dependent? What happens if, even if the flu treatment is successful, the patient will survive only to a “poor” quality of life? Is this a “poor outcome” from the perspective of the triage decision-maker? The general point here is that it is all too easy for social (or value) type criteria to slip in under the guise of medical or clinical criteria. If question 2 above is a medical question – there is no ethical problem in having the medical experts answer it. But if it is a disguised value-criterion then it warrants clear discussion as to just what is meant by “poor outcome.”

3) Is more than one organ system affected?

It is not clear how this is not an alternative way of asking question 2.

4) Does the patient improve with critical care?

This seems to be a criterion that kicks in only after the patient has been admitted to critical care, or has been granted access to a ventilator. It really concerns when a patient should be removed from critical care (or a ventilator) in favour of someone who could benefit more (or is more likely to benefit.) Again the issue is more general than merely the pandemic situation. When, and how do we decide that continued treatment is pointless (or futile) how do we decide that enough is enough? But here is where the emergency pressures of a pandemic situation kick in. Is the decision that we have tried enough, that the treatment is not working, independent of the pressure to use that ventilator, or critical care bed for someone else? Usually we want to say that medical decisions about treatment for this patient – are made on the basis of this patient’s needs not on the basis of someone else’s needs. Another way of putting this is to say that I want my physician at my bedside to make decisions for my benefit, not for anyone else’s. For example, I do not want my physician to make a decision to deny me access to a treatment from which I might benefit because someone else might benefit more. (This issue is part of a larger discussion too. Just as we often blur medical decisions with value decisions so too we often blur medical decisions with resource reasons. We do, clearly, and not just in a pandemic, need to allocate scarce resources in the most effective means possible. But the general decisions about what treatments should be made available, and the general criteria for access to this service or that, are not simply medical decisions but rather involve social determinants and what we, as a society are willing to pay for.) It is not clear from the criteria as listed whether health care workers are being charged with taking into account, as they decide for patient X, patients Y and Z who might be able to benefit more.

So much for the easy triage decisions....

What happens when “there are many very sick patients who have a good chance of survival but not enough critical care beds or ventilators?”

Hamilton Health Sciences proposes a set of additional criteria:

1) Does the patient belong to a profession that could help save/care for others?
2) Is it likely that the patient became exposed to influenza through their professional duties?
3) Is the patient pregnant or a caregiver for dependents (children under 18, disabled adults or elders)
4) Is there any evidence that the person is significantly more likely to survive than others?
5) Is the patient significantly younger than the others, thus at risk of losing more years of life?
(Globe and Mail Sept 21st)

These criteria introduce elements that are clearly not merely, or indeed, not even, medical. As such they require a far broader discussion, understanding and agreement than expert technical criteria.

Let’s look at them in turn:

1) Does the patient belong to a profession that could help save/care for others?

I wonder what reaction comes to your mind first. On the one hand this looks like a good idea. Just as in a battlefield triage situation you would patch up the fighters first so that they could continue to fight, doesn’t it make sense to keep the carers going so that they can continue to care? But I know that the cynics amongst you will ask – who made these criteria up – surely the very health care professionals who now find themselves at the head of the queue. And what do we mean by “health care professionals?” Front-line staff, nurses, and physicians would clearly be there, but what about administrative staff? What about the managers charged with leading organizations and the community through the tough decisions required in a pandemic? And what about “professionals?” Front line “professionals” cannot do their work without the support of housekeeping, dietary services and so on would people working in these roles be given priority access too?
What do you think? Would you accept this criterion as a fair and just way of allocating a scarce resource? Who would you include in this priority professional list? (Your local MP?) Does it matter that you are (or are not) a member of the health care professions (industry)?

2) Is it likely that the patient became exposed to influenza through their professional duties?

This one is interesting because it looks not to the future (to benefits that might accrue by saving this person rather than that) but instead it looks to the past. It recognises that we, as a community, expect more things from some groups rather than others and, as such, we ought to compensate those people for the additional burdens they carry. Because it looks to the past it has a completely different grounding from moral reasons that look to the future. Most of the criteria used in triage decision-making are forward looking (“consequentialist” in the ethics jargon) they seek to maximise the good by doing those things that are most likely to have the best outcomes – that are most likely to bring about the most good.) This criterion looks back, in effect recognising an obligation to those we ask the most from. It is entirely possible that in following this principle we could, in any particular case make a decision that has worse consequences than an alternative. This type of backward looking reason is unmoved by consequences.

We could take this type of backward looking reason a step further. What happens if there is a programme of vaccination for, for example, health care workers that is encouraged but not mandatory. (See the post Sept 22nd below.) Let’s say you chose not to be vaccinated and you contracted influenza, probably as a result of your close professional contact with a high-risk group. You now need a ventilator. You meet the criterion above – but you chose not to accept a vaccination that could have been useful. Should you have access to the ventilator?

What do you think? Is this a good criterion? Who should it apply to? Should it apply to people who chose not to be vaccinated? What do you think – in general, of introducing backward looking criteria in health care decisions?

3) Is the patient pregnant or a caregiver for dependents (children under 18, disabled adults or elders)

This is another version of the social usefulness criterion that we saw in number 1. In a situation where many require care we should focus our attention on those that provide the care, so that they can return as quickly as possible to their roles as caregivers. But why is providing care to others the principal social role that needs protecting. Wouldn’t we, in a pandemic with the sorts of effects we are dreaming in our nightmares, need police to keep order in the streets, security guards to protect hospitals and stocks of Tamiflu, truck drivers to deliver supplies, funeral directors, morticians and gravediggers, administrators and managers and politicians to make the priority decisions we need to keep things functioning – and on and on and on (oh and ethicists of course to ask the difficult questions.)

Again, what do you think? Are there some occupations or roles that are more important than others in a pandemic situation? If so, what are they and should members of those professions be pushed to the head of the queue?

4) Is there any evidence that the person is significantly more likely to survive than others?

This one seems strangely out of place in this list as it seems to be just an extension of the medical criteria we saw in the first set of triage criteria. But it does take things an interesting step further. Typically we make medical decisions on the basis of need. The first criterion is that the person needs, and could benefit from access to critical care or a ventilator. This ups the ante – now it concerns people who are most likely to benefit. Who would these people be? We don’t know yet, of course, but we can make some educated guesses. In a novel flu strain there may be some benefits to having been exposed to other strains of flu. That is more likely to be the case for people who are older rather than the young. And amongst that older group there would be benefits to being generally healthy. So, as a group it could be the case that older otherwise healthy people would be more likely to survive than others.

So again, what do you think? This is a return to a consequentialist style approach – we are seeking to maximise the good and we do that by treating those most likely to survive.

5) Is the patient significantly younger than the others, thus at risk of losing more years of life?

This again, is a consequentialist form of reasoning. If you save a young person’s life rather than an older person’s you are likely to get a better return, at least in terms of years lived, from the young person who may live a long time, over the older person who has fewer years to live. But is more necessarily better? Is it better to try to save the young delinquent, drug-abusing teenager, over the older, at-the-peak-of-her-career, cancer researcher? Again, what seems straightforward and uncontroversial – a prejudice in favour of the young, seems far more difficult to defend on second reflection.

Conclusions

We are exploring new and dangerous territory. The stated objectives of proposing these triage criteria is to remove the decision from the idiosyncratic whim of an individual triage officer and share the decision amongst a group that in turn can ground the decision in a set of principles. These are important objectives. And, the criteria that are proposed have some common sense behind them – isn’t it a good idea to do the most good that we can – and don’t we, as a community, owe something to those who provide care on our behalf. But, we give up a great deal as we travel in that direction. We pride ourselves on our values of equality and diversity. We do not believe that any one is intrinsically more worthy of social benefit than anyone else, and we celebrate diversity of values and difference in ways of life. In Canadian health care that is put into practice through equality of access and care based on need. If you need the service, and it is available you will be put on a waiting list (which, in effect is a way of fairly distributing a scarce resource through random allocation.) Access to the service does not depend on who you are, ability to pay, or your job or role, or even why or how you contracted the disease in the first place. Should we abandon those principles in the emergency of a pandemic? If we do abandon them we may well achieve a greater good for the community, but at what price to the values that made that community what it is in the first place?

As always I would welcome you comments.