It is a sad truth that we sometimes spend as much time deciding what not to do when treating elderly patients as we do determining what options are appropriate. The early part of each admission is straightforward enough – we take a history, perform an examination, form a list of possible diagnoses and initiate treatment of some sort. But as we walk away at the end of the day, part of our duty is to consider what will happen if the patient deteriorates. If, in the middle of the night, or over the weekend, they suffer some sort of collapse – what instructions should we leave?
In most cases we do not need to leave specific instructions, because the hospital is set up to deal with such emergencies. The patient’s worsening condition will be identified by a nurse, the on call team will be bleeped (perhaps a ‘MET’* call will be put out), resuscitative measures will be taken and they may even be transferred to the high dependency or intensive care unit. If their heart stops they will undergo cardio-pulmonary resuscitation (CPR). That is the default. For many though, such a sequence of events will prove disastrous.
The reasons why very elderly or already frail patients should not undergo resuscitation have been rehearsed on this blog before, but many of those arguments apply to ICU admission too. The potential benefits are so slight that the definite burdens (such as prolonged ventilation, tracheostomy insertion, connection to a dialysis machine via central venous catheters, critical care neuropathy) cannot be justified. So, it is necessary to determine in advance whether those options are offered.
This is reasonably straightforward if the patient is alert and understands the issues. Unfortunately, a large proportion of acutely unwell elderly patients do not have the capacity to engage in such a conversation when they arrive in hospital. We therefore turn to the family, and by combining what we learn about their pre-existing wishes with what we anticipate are the possible benefits of organ support, we set ‘ceilings of treatment’. We try to work out what the patient would want if they could talk to us clearly. We make an informed guess, then we write in the notes phrases such as:
Not for ICU
Not for MET
Not for escalation
Ward based care only
Not for filtration
Not for NG tube
Not for NIV**
Not for ventilation, for NIV only…
Not for this, not for that. A typical ward round or on-call shift might require many such decisions, and it can feel overwhelmingly negative. Doctors are trained to treat actively, but increasingly, as the population ages and more and more patients are admitted to hospital toward the end of the expected human life span, they have to be protect them from the full gamut of invasive medical procedures. And the less information they have about the patient’s preferences, the greater the weight of responsibilty in setting those ‘ceilings’.
How then to change the emphasis and accentuate the positive, focus on what we can do, and not what we cannot?
It’s all about anticipation. I am always amazed when patients in their nineties, or their families, express surprise when I bring up the possibility that they might die. It seems never to have crossed their minds! This natural reluctance to consider death has been challenged. Dying Matters (a coalition led by the National Council for Palliative Care) highlights that,
81% of people have not written down any preferences around their own death
Nearly two thirds (63%) of us would prefer to die at home, yet of the 500,000 people who die each year in England, 53% die in hospital.
They request that people consider,
The type of care you would like towards the end of your life
Where you would like to die
Whether you have any particular worries you would like to discuss about being ill and dying
Whether you want to be resuscitated or not
‘It’s OK to die’, an organisation in the United States set up by Monica Williams Murphy MD, advocates a similar set questions. Imagine if every patient came into hospital with this information. No more guess work. No more stressed families trying to explain what their elderly relative would have wanted, feeling the weight of responsibility even when reassured that they are not being asked to make ‘life or death decisions’.
Engaging in these questions may not be natural. Those with manifestly terminal conditions may be drawn into such discussions by their families, palliative care teams or general practitioners (GPs, family doctors), but those who are in their ninth and tenth decades without end stage disease are not necessarily going to dwell on their mortality. It falls to their GPs to initiate the discussion.
The concept of the Advance Care Plan (ACP) is well established, albeit in the context of life limiting conditions. The document ‘Advance Care Planning: A Guide for Health and Social Care Staff’ specifies that,
The difference between ACP and care planning more generally is that the process of ACP will usually take place in the context of an anticipated deterioration in the individual’s condition in the future, with attendant loss of capacity to make decisions and/or ability to communicate wishes to others.
It might seem then ACPs are not intended for people in a stable condition who are in no imminent danger of deteriorating. The document does however include this common, ‘non-terminal’ example scenario:
Mrs Carter – An 81 year old lady with COPD, heart failure, osteoarthritis and increasing forgetfulness, who lives alone. She fractured her hip after a fall, eats a poor diet and finds mobility difficult. She wishes to stay at home but is increasingly unable to cope alone and appears to be ‘skating on thin ice’.
Perhaps then it is not unreasonable to introduce the idea of an ACP to people with common cardio-respiratory problems that do not seem ‘life-limiting’ in the most obvious sense.
The same document is explicit that:
If an advance decision includes refusal of life sustaining treatment, it must be in writing, signed and witnessed and include the statement ‘even if life is at risk’
And warns that it,
Only comes in effect if the treatment and circumstances are those specifically identified in the advance decision
These stipulations make ACPs less workable in primary care or general medicine. How can a person be expected to anticipate specific circumstances? How will they know, in advance, what the burden of those life sustaining treatments will be, in relation to the possible benefits. Again, the emphasis is on the negative. Wouldn’t it be more helpful to know what they do want, what their ultimate goals are. For instance, ‘to be able to look after myself when I get home, feed myself, to interact with my grandchildren…’ These goals may preclude extended life preserving treatment that can be reliably predicted to result in chronic debilitation or loss of independence. They may be better suited to a ‘Preferred Priorities of Care’ statement. In reality, I suspect the term ACP is used in a more general way than intended.
Do ACPs work? There is some evidence. A study by Karen M Detering et al (BMJ 2009) randomised 154 of 309 mentally competent octogenerian in-patients to the make an ACP, and found,
Of the 56 patients who died by six months, end of life wishes were much more likely to be known and followed in the intervention group (25/29, 86%) compared with the control group (8/27, 30%; P<0.001).
In the intervention group, family members of patients who died had significantly less stress (intervention 5, control 15; P<0.001), anxiety (intervention 0, control 3; P=0.02), and depression (intervention 0, control 5; P=0.002) than those of the control patients.
A proactive approach to ACP (or equivalent) is not without risk. As part of the National End of Life Care Strategy GPs were encouraged to identify those under their care who were predicted to die in the next twelve months, and to make a list. Some parts of the press did not take this well;
Thousands of patients have already been placed on ‘death registers’ which single them out to be allowed to die in comfort rather than be given life-saving treatment in hospital, it emerged last night.
Nearly 3,000 doctors have promised to draw up a list of patients they believe are likely to die within a year, Department of Health figures showed yesterday.
As part of an unpublicised campaign endorsed by ministers, GPs have been encouraged to make lists – officially known as End of Life Care Registers – of people they believe are going to die soon and should be helped to do so in comfort. (Daily Mail, 18th October 2012)
The subject became entangled with the Liverpool Care Pathway controversy. I don’t know if the registers have continued to be compiled. The taboo of death remains strong. But the idea is sound. Dying Matters have promoted the ‘Find your 1%’ approach. They request that GPs ask themselves,
“Would I be surprised if this person were to die in the next 12 months?”
…this simple question is accurate seven times out of ten.
They emphasise some general indicators of likely death within 12 months:
– Limited self-care and interest in life: in bed or a chair more than 50% of their time.
– Breathless at rest or on minimal exertion
– Progressive weight loss (>10% over last six months).
– History of recurring or persistent infections and/or pressure ulcers.
The GMC too recognises the importance of being forward about end of life care planning. Its End of Life Care booklet includes this:
If a patient in your care has a condition that will impair their capacity as it progresses, or is otherwise facing a situation in which loss or impairment of capacity is a foreseeable possibility, you should encourage them to think about what they might want for themselves should this happen, and to discuss their wishes and concerns with you and the healthcare team.
Areas for discussion include:
– the patient’s wishes, preferences or fears in relation to their future treatment and care
– interventions which may be considered or undertaken in an emergency, such as cardiopulmonary resuscitation (CPR), when it may be helpful to make decisions in advance
– the patient’s preferred place of care (and how this may affect the treatment options available)
These initiatives and guidelines are fine, but harder to achieve than they are to promote. Talking about death is never easy (for families or doctors), and it is still more common than not for patients to be admitted with no evidence, verbal or written, of their fears, priorities or goals. It falls on doctors and nurses whom the patients have never met to introduce the subject and explore those issues…or, when mental capacity has been lost, to gather clues from the family who may not have had those conversation either. Until this changes it will remain a sad part of our role to gauge how far each patient should be escalated, hoping always that our judgment is correct.
* MET score – medical emergency team; most hospitals now have an ‘early warning’ system for deteriorating patients, whereby nurses can summon the on-call team if there are dangerous blood pressure, pulse rate or respiratory changes.
** NIV = non-invasive ventilation; a tight mask is applied to the face to improve lung function, effective but often quite unpleasant as the patient is awake