Advance care directives are not always helpful

Universal advance care directives may not alleviate the current difficulties surrounding end-of-life discussions

There is an increasing call within this country for patients to express their desired level of care in the event of medical deterioration ahead of time.1 While this has previously been pertinent in the case of patients with chronic irreversible illness, there is an additional argument and evidence that the practice should be more widespread among elderly Australians in general.2 It has even recently been discussed at the Festival of Dangerous Ideas in Sydney, where the cardiologist David Celermajer suggested that all citizens over the age of 73 years should have a clear advance care directive (ACD) in place or lose their Medicare card.3

ACDs are documents that state the level of care that patients would want if they were to become significantly unwell, in case their ability to communicate this or their capacity was lost.1 Clearly there are potential benefits to patients possessing ACDs. When somebody approaches the end of his or her life, one of the most difficult discussions in medical practice is always that which involves the limitation of care. This may occur directly with the patient in question or, when this is not possible, with the patient's next of kin or family. There is little dispute that the presence of an ACD, legally binding or otherwise, simplifies the question of patients' desired care if they are unable to express this themselves. The expectation is that ACDs may allow a patient to communicate in writing that he or she would not desire invasive care, instead preferring withdrawal of treatment in favour of comfort measures.

All proponents of universal ACDs seem to suggest that they will make end-of-life care more straightforward and reduce the current burden of futile and inappropriate treatment so frequently taking place in intensive care units and acute medical wards. This is true in the situation where a patient expresses a desire to limit treatment, but it is possibly untrue for other patients and sets an unrealistic expectation of what may exist within the ACDs themselves. If an ACD requests full treatment, at any stage, at any cost, this in no way changes the difficulty so frequently encountered by physicians when the expectations of a patient or his or her family conflict with the reality of what treatment may actually offer.

To ask people who are well and whose only risk factor for death is their advancing age exactly what they would want if they were to become seriously ill may be an act of prudent foresight on the part of the clinician. Statistically, a large portion of these patients may require such a decision to be made relatively soon due to a rapid and unexpected decline in their health. Here, an ACD may assist in discussion with a patient's family members as to the potential benefit of the individual's comfort and dignity. Alternatively, the initial discussion with the well patient may come as a surprise to many such individuals and it is therefore not unreasonable to expect that some would document clearly that they desired all that medicine had to offer. If this was to happen, a scenario may occur where a dying patient should be palliated but treating physicians are placed in an ethical dilemma with an ACD that requests active measures, written at a time when the patient may not have been able to fully appreciate the implications of this directive.

It seems that those who call for all citizens to have an ACD really mean for them to document clearly that in the event of a medical deterioration they would prefer limited care or palliation, rather than full active treatment. This is not unreasonable, but it has not been expressed well so far to the public. It is impractical to expect patients within the wider community to all fully understand the limits of what medicine can offer in various situations. Ideally the composition of an ACD involves multifactorial and prolonged discussions in the form of advance care planning2 where, over time, a greater understanding of treatment limitation may be established. This is a dynamic process and, by its nature, should be under constant review. It remains the role and obligation of clinicians to recognise and communicate when certain treatments should be withdrawn and, most importantly, when comfort measures are more in the interest of patients as they approach the end of their lives.