Difficult but necessary conversations — the case for advance care planning

Modern medicine saves many people from acute illness who then live longer with chronic illnesses associated with trajectories of declining physical and mental function over months and years, often punctuated by episodes of acute illness or decompensation. Regrettably, considerable suffering as well as dissatisfaction with and overuse of health care result from a mismatch between the needs of chronically ill patients and current practice norms. Advance care planning (ACP) provides a means of ameliorating this mismatch but is yet to be embedded in routine clinical practice or public consciousness.

ACP is a process of making decisions about future health care for patients in consultation with clinicians, family members and important others. It aims to ensure patients’ wishes are respected if they lose decisional capacity. Conceptually, it comprises five sequential phases, from pre-contemplation to action and maintenance (Appendix 1),1 which include completion of a written advance care plan (or advance health directive [AHD], also termed “living will”) and the appointment of a surrogate decisionmaker. Unfortunately, in the past, ACP has often been focused on raising completion rates of AHDs, despite there being no guarantee such documents in themselves improve end-of-life care2 or correspond with future care preferences.3 A better focus is to encourage widespread use of ACP as a process for iteratively identifying and facilitating what patients consider important for a “good death” (such as managing symptoms, avoiding prolonged dying, achieving a sense of control, relieving burdens placed on the family and strengthening relationships) and for informally communicating their future wishes.4

At least 50% of all deaths in Australia each year are clinically expected because of advanced disease.5 In their last year of life, Australians with advanced disease will average eight hospital admissions and incur a 60%–70% chance of dying in hospital.5,6

More than half may be denied adequate opportunity to discuss end-of-life care wishes7 or have them fully enacted.8 Many patients at the end of life undergo futile (of no benefit at all) or inappropriate (harms outweigh potential benefits) interventions.9 Almost a quarter of intensive care beds are occupied by patients receiving potentially inappropriate care,10 while up to a quarter of health budgets are spent on inpatient care during the last 18 months of life without any real prospects of extending overall survival or impacting on quality of life.11 Most complaints received from bereaved family members about hospital treatment relate to end-of-life care, mainly perceived failures of communication and preparedness for death.12 In one study, doctors spent a median time of only 1 minute on do-not-resuscitate discussions with patients after admission.13

Randomised trials of ACP are few and report mixed results. The sentinel trial, SUPPORT (Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments), which used nurse facilitators over 2 years to undertake discussions with seriously ill hospitalised patients and families and document their preferences, showed no improvement in patient control over their treatment.14 However, a more recent trial of ACP showed improved patient and family satisfaction and alleviation of anxiety in relatives of hospitalised patients.15 Observational studies have shown similar findings.16 Other randomised trials17 and before-and-after studies18 involving nursing home residents have shown 40% to 80% reductions in rates of hospitalisation and up to threefold increases in palliative care referrals resulting from advance care directives. Experience with ACP in the aged care sector in Victoria has shown that less than 3% of residents in aged care facilities approached about ACP declined it, and 90% of those who completed ACP died in the facility while receiving palliation rather than in hospital, compared with only 50% of those who had not completed ACP.19 Use of ACP invokes earlier initiation of more appropriate palliative care, which improves patient symptoms and mood, reduces undesired use of invasive interventions and life-sustaining treatments, lowers the likelihood of inhospital death, prolongs life of higher quality and decreases costs.20-22

Regardless of potential benefits, ACP uptake in Australia has been slow, despite widespread professional and public endorsement coupled with supporting legislation in every state and territory. However, many hospitals and nursing homes are now implementing ACP programs, most along the lines of the Respecting Patient Choices program initiated at Austin Hospital in 2002.23 Factors inhibiting universal adoption of ACP are several:

• Reluctance to acknowledge impending mortality: Reflecting societal norms, most individuals, including clinicians, find it difficult to think and talk about dying.

• Perceived irrelevance: Many patients may perceive themselves as being “too healthy”, hope (often unrealistically) for medical technology to prolong healthy life, regard future events as a matter of fate, or consider their wishes are already known to doctors and family.24

• Issues of timing: Intuitive prognostications of clinicians can be inaccurate, with formal prediction tools offering limited assistance.25 Consequently, “curative intent” remains the focus, with initiation of ACP often occurring too late, leading to rushed care decisions and suboptimal outcomes.

• Diffusion of responsibility: The role of initiating and coordinating ACP, which must encompass all patient needs — clinical, emotional and spiritual — must be accepted by someone. Patients often wait for doctors to broach the subject, while clinicians wait for patients or family members to do so. General practitioners wait for a signal from specialists that may never come if multiple specialists are involved, with no one taking charge overall.26

• Limited clinician time, skill and remuneration: The pressure of clinical work, a focus on managing acute medical problems and the absence of suitable training and remuneration discourages clinicians from dedicating time to ACP.

• Limitations of surrogate decisionmakers: The views of proxies and patients are often discordant,27 or there may be conflicts between family members. Surrogate decisionmakers may not want to assume sole responsibility, and subsequent guilt, for terminating life-sustaining measures.

• Ambiguous, inconstant or poorly recorded wishes: Patient preferences can change over time, are often based on wrong or insufficient information, or are documented in vague, incomplete terms on forms not easily retrieved when needed. Because of this lack of clarity at the time of decision making, doctors and surrogates may disregard or feel incapable of enacting expressed wishes, especially if legal or ethical concerns persist around patients’ decisional capacity at the time ACP documents were written or altered.

• Differing perceptions of ACP: Many people feel intimidated by what they perceive as legally binding and irrevocable AHDs which arise from ACP. Others equate ACP with palliative care, which they assume indicates imminent death due to a rapidly terminal illness. Finally, some link ACP with euthanasia or assisted suicide, which has seen doctors in the United States who discuss ACP with patients being caricatured as “death panels”.28

Given these challenges, no single strategy will achieve the transition of patients and their proxies from pre-contemplation to action in ACP.29 The most effective and systematic, yet personalised, approach comprises structured, iterative conversations about values and preferences for end-of-life care led by trained, trusted and paid facilitators (who may include nurses and social workers as well as doctors). Such conversations occur over several visits, actively involve properly informed surrogates, and are supported by oral, written and videorecorded information. In these interactions, an advance care plan or a medical enduring power of attorney may serve as more versatile media for ACP than a static AHD.30

In embedding ACP into routine practice, all health care organisations (general practices, hospitals, residential care facilities) need to become “conversation ready”; that is, committed to systematically eliciting, documenting and enacting patients’ care preferences.36 System-wide processes (Box 3) are needed that will consistently:

• invite all eligible patients to consider ACP for future care relevant to their stage of illness;

• provide competent assistance by trained and accountable personnel; and

• ensure written plans (however documented) are

• accurate, relevant and understandable to all stakeholders

• stored, transferred and retrieved wherever the patient is being treated

• updated and rendered more specific as illnesses progress

• sighted and honoured at the right time.

Given their longstanding, trusted relationships with patients, GPs are probably best placed for timely ACP, but they need to be supported in this task by medical specialists, senior nurses and allied health professionals. Legal clarification is required regarding the need for advance care plans or AHDs to comply with specific forms, their transferability between jurisdictions, their scope in covering all future treatment decisions, and the enforceability of oral plans or directives. In all Australian jurisdictions, competent patients or surrogate decisionmakers cannot demand treatment that clinicians believe to be futile, including enteral or intravenous nutrition and hydration. There are no reports of Australian or UK courts overturning a carefully considered decision to withhold treatment doctors deemed to be futile.37

The goals of ACP are indisputable and its benefits are becoming evident. If ACP is to become a mainstream clinical activity, health care services and professionals must effectively educate themselves and their patients about its purpose. They must take practical steps to implement auditable ACP systems in routine practice, document ACP in accessible formats that enables patient wishes to accurately guide clinical management, review ACP decisions when clinically required, and evaluate the effects of ACP on clinical outcomes and the fulfilment of patients’ wishes.