Withholding and withdrawing life-sustaining treatment in a patient's best interests: Australian judicial deliberations

What should doctors do if they cannot reach agreement with a family about life-sustaining treatment for an adult who lacks decision-making capacity? Effective conflict-resolution strategies generally resolve these disputes and intractable conflict is rare.1-3 But when such disputes occur, they are stressful for clinicians, families and patients. Health professionals can experience moral distress, with significant personal and professional impact.1,4 In such cases, or where there are concerns about the lawfulness of proposed conduct, medical practitioners or their hospitals may need to seek a court or tribunal opinion.1,5,6 Concerned family members may themselves seek a judicial intervention.

Although all Australian states and territories have adult guardianship tribunals,7 Supreme Courts retain an important role in this field. They have jurisdiction to resolve end-of-life disputes, and their decisions provide authoritative guidance for guardianship tribunals and clinicians in their deliberations. Supreme Courts apply the test of whether the proposed treatment is in the patient's “best interests”, and this term (or analogous concepts) is also one of the criteria applied by guardianship tribunals.7 Yet, what “best interests” means is contested.8,9 In this article, we identify six key themes from the developing body of Supreme Court jurisprudence about life-sustaining treatment decisions for adults who lack capacity (Box 1).

Supreme Court case law concerning end-of-life decisions

We are aware of only 16 Supreme Court decisions in Australia that concern the withholding or withdrawal of life-sustaining treatment from an adult who lacks capacity. Of these decisions, the issue of the patient's best interests was directly relevant in eight cases (Box 2). The other eight cases focused on other issues, such as the validity of an advance directive and interpretation of guardianship legislation.

To determine trends in judicial reasoning, we thematically analyse the eight cases involving a determination of best interests. Most of these cases involved proposed withdrawal of treatment. The law generally treats withholding and withdrawing treatment as equivalent.10 However, situations involving withdrawal are possibly more likely to lead to family conflict, because decisions to stop treating, as opposed to not offering treatment, are more prone to be construed by families as more causally connected to death.11

Theme 1: Futile treatment is not in a patient's best interests

The concept of “futility” is contested.12,13 We do not add to that debate here, but instead highlight the link made in judgments between futility and best interests. In three of the eight decisions (Melo, Herrington and Messiha), the Supreme Court explicitly stated that where treatment is futile, it would not be in the patient's best interests to commence or continue with it. In Melo and Herrington, the court did not explain why the treatment was futile. Instead, the medical practitioners' assessments of futility were relied on to inform the best interests assessment. In Messiha, it was held that treatment would be futile because continuation of life-sustaining measures would confer no real prospect of recovery.

Two other cases also clarify the role of futility in these deliberations. In Justice Health, the court stated that active treatment was futile because it would “achieve no more than a short prolongation of life without quality” and therefore did not need to be provided. By contrast, in JT, the court found that treatment was not futile (the patient was not dying) and so it would not authorise non-treatment.

The courts have expressed the view that futile treatment will not be in a patient's best interests.14 But what justifies labelling treatment as futile? Courts have generally relied heavily on medical determinations of futility. Despite ongoing debate regarding this concept, medical opinion is important to judicial determinations.

Theme 2: Overly burdensome treatment is not in a patient's best interests

In the end-of-life setting, decisions may be made to commence or continue invasive forms of treatment such as assisted ventilation or to progress to palliative treatment. Supreme Courts have held that life-sustaining treatment that creates excessive burdens for a patient, relative to possible benefits, is not in the patient's best interests (Slaveski). This includes considering potential pain or indignity that the patient may suffer through receiving treatment (Herrington). Treatment may still be burdensome even where the patient is unconscious and unaware of these burdens (Messiha).

Theme 3: Quality of life is relevant

Quality-of-life considerations have not been explicitly acknowledged by the Supreme Court as relevant to patients' best interests, apart from the brief mention in Justice Health noted above. Nevertheless, in the remaining seven cases, we contend that quality of life has been relevant when assessing the patient's best interests. This has at least occurred implicitly through an examination of a person's prognosis as part of a best interests assessment.

To illustrate, in the four cases where treatment was withheld or withdrawn, the patients had profound brain injury with no prospect, or very little prospect, of neurological recovery (Slaveski, Melo, Herrington and Messiha). By contrast, the three remaining cases where life-sustaining treatment was commenced or continued involved patients in better neurological states (Northridge, JT and Astill). Although not couched in terms of quality of life, the capacity to engage meaningfully with the world seems to be relevant. It appears that, at least indirectly, judges consider how the proposed treatment will affect quality of life.

Theme 4: The views and wishes of the patient (and perhaps family) are relevant but not determinative

When assessing a patient's best interests, Supreme Courts have given some consideration to the views and wishes of the patient and family members. This was given at least some attention in three cases, although such views were not influential on the courts' final conclusions.

In Astill, a woman's “no blood” advance directive failed because it did not comply with formalities required by Queensland legislation. The court still considered her views but concluded that there was no evidence to suggest that they remained current when the case was heard. Further blood transfusions were authorised despite the previous directive. In Herrington, the court considered the views expressed by the patient (which were communicated to the court by the patient's partner), stating that she would have wanted continued treatment. Ultimately, however, treatment was universally regarded by the medical evidence as futile and was not provided. In JT, the court acknowledged the patient's views as a relevant factor, but they were disregarded because they were “the product of delusional and irrational thought”.

The cases also demonstrate several ways that the views and wishes of family members can be considered in the best interests assessment. First, family can provide information relating to the patient's values or wishes about proposed treatment (Herrington and Astill). Second, family members may have their own views about the patient's best interests. In Astill, the judge observed that he had taken into account views expressed by all family members when determining the patient's welfare. The other seven cases did not formally acknowledge the views of family members as being directly relevant to the best interests assessment; however, some made reference to family views and preferences (Slaveski, Melo, Herrington, Messiha and Northridge). Finally, family members have expressed views to the court about the patient's level of responsiveness and awareness of their surroundings. Thus, in some cases, the family challenged the medical prognosis to suggest that the patient had responded or engaged in a way that indicated an improved state. In all such cases, the court preferred the views of the medical team, concluding that continued treatment was not in the patient's best interests (Melo, Herrington and Messiha).

Theme 5: The interests of others are not relevant

Supreme Courts have not regarded the interests of others (aside from the potential relevance of the views of the family members outlined above) as being relevant to the best interests assessment. For example, in Northridge, the court observed that “the exercise of the parens patriae jurisdiction [the court's inherent jurisdiction that allows it to make decisions concerning people who lack capacity] should not be for the benefit of others … including a health care system that is intent on saving on costs”. Similarly, in JT, the potential distress to health professionals from providing forced treatment did not influence the assessment of what was in the patient's best interests.

Theme 6: Judicial deference to the medical profession

Supreme Courts have usually deferred to medical opinion when assessing best interests. In some instances (eg, Melo and Herrington), the courts did not question the conclusion reached by medical practitioners that treatment was futile; however, as emphasised in Messiha, the final assessment rests with the court. In Northridge, medical decision making was found not to be clinically justified, and the New South Wales Supreme Court was critical of the medical opinion provided.

Conclusion

Supreme Court jurisprudence in Australia on end-of-life treatment is still developing, but there is enough case law to provide useful guidance about assessing a patient's best interests. This guidance is significant both for future Supreme Court decisions and those made by guardianship tribunals. Although only a minority of cases require judicial intervention, legal considerations remain relevant for the larger group of difficult decisions that occur each day, because medical decisions are made in the “shadow of the law”.15 Our analysis highlights the themes that emerge from judicial decisions. Although every situation has unique circumstances, these factors may be useful for medical practitioners contemplating withdrawing or withholding treatment from incompetent adult patients.

A best interests assessment not to treat can be justified at law if there is a clear basis for deciding treatment is futile (despite this term's subjectivity) (theme 1), or if the patient is extremely unlikely to recover consciousness (theme 3). Treatments that are particularly invasive or burdensome relative to their benefits will also not be in a patient's best interests (theme 2). Decisions should take account of patient views, where known (theme 4), and those of family members. However, the interests of an overstretched health system are not relevant (theme 5). Finally, where a medical view concludes treatment should not be provided, this is likely to be supported by the court, but it should be corroborated, for example, with a second opinion (theme 6).