# Advanced Care Planning and Goals of Care

## Introduction

Advance care planning is the process through which individuals understand, reflect upon, and discuss their preferences for future medical care in anticipation of a time when they may be unable to make decisions for themselves. Despite its recognized importance, only approximately 37 percent of adults in the United States have completed any form of advance directive. Completion rates are somewhat higher among the elderly, ranging from 50 to 60 percent, but remain inadequate given the frequency with which decision-making capacity is lost in this population due to dementia, delirium, stroke, or critical illness.

The benefits of advance care planning are well-documented and extend to patients, families, and the healthcare system. ACP is associated with increased concordance between patient wishes and the care actually received, reduced unwanted aggressive interventions at the end of life, improved family satisfaction with care, and reduced caregiver distress and complicated grief after the patient's death. A critical conceptual point is that advance care planning is a process rather than a one-time event. It requires ongoing conversations that evolve as the patient's health status changes, as new diagnoses emerge, and as preferences shift in response to lived experience with illness. A document completed a decade ago may no longer reflect the patient's current values and wishes.

## Advance Directives

### Types of Advance Directives

#### Healthcare Proxy / Durable Power of Attorney for Healthcare (DPAHC)

The healthcare proxy, also known as the durable power of attorney for healthcare, is the designation of a trusted person to serve as the patient's surrogate decision-maker when the patient lacks the capacity to make healthcare decisions independently. This is widely considered the single most important advance directive because a trusted surrogate who understands the patient's values can respond to unanticipated clinical situations that no written document could have predicted. The ideal surrogate is someone who knows the patient's values intimately, can advocate effectively within the healthcare system, is reliably available, and possesses the emotional resilience to make difficult decisions under the pressure of a medical crisis. Healthcare proxy documents are legally binding in all US states, though the specific forms and requirements vary by jurisdiction.

#### Living Will

A living will is a written statement of the patient's treatment preferences for specific clinical scenarios, typically focused on end-of-life situations. Common elements include preferences regarding cardiopulmonary resuscitation, mechanical ventilation, artificial nutrition and hydration, and dialysis. Despite their widespread use, living wills have significant limitations. They cannot anticipate all possible clinical scenarios, and vague language such as "no heroic measures" or "no extraordinary treatment" creates interpretation challenges that may leave clinicians uncertain about the patient's actual intent. Living wills may also fail to reflect the patient's current wishes if they were completed years or decades earlier, as values and preferences often evolve with changing health status. A living will without a designated healthcare proxy is less effective precisely because it cannot adapt to unexpected situations.

#### POLST/MOLST (Physician/Medical Orders for Life-Sustaining Treatment)

The POLST, known as MOLST in some states, represents a fundamentally different type of document from traditional advance directives. It is a portable set of medical orders that translates patient preferences into actionable clinical directives, completed by a clinician in conversation with the patient or, when the patient lacks capacity, with the surrogate decision-maker. POLST forms typically cover three domains: CPR status (attempt resuscitation versus do not resuscitate), the scope of medical interventions (full treatment, selective treatment, or comfort measures only), and preferences regarding artificial nutrition.

POLST forms are distinctively colored, usually pink or green, and are designed to travel with the patient across care settings, ensuring that the patient's treatment preferences are immediately available to emergency medical services, hospital staff, and nursing facility personnel. POLST is appropriate for patients with serious illness, advanced frailty, or limited life expectancy, but it is not appropriate for healthy adults, for whom standard advance directives are sufficient. A critical distinction is that POLST is a physician order, whereas an advance directive is a legal document. The POLST paradigm has been associated with reduced unwanted hospitalizations, increased concordance between patient preferences and care delivered, and reduced ICU utilization at the end of life.

| Document | Legal Nature | Who Completes | When Appropriate | What It Covers | Portability |
|----------|-------------|--------------|-----------------|---------------|-------------|
| Healthcare Proxy / DPAHC | Legal document | Patient | All adults; most important directive | Designates surrogate decision-maker | Varies by state |
| Living Will | Legal document | Patient | All adults | Treatment preferences for end-of-life scenarios | Limited (often vague) |
| POLST/MOLST | Medical order | Clinician + patient/surrogate | Serious illness or limited life expectancy | CPR, scope of treatment, artificial nutrition | Travels with patient (colored form) |
| DNR/DNI | Medical order | Physician | When goals align with comfort | Cardiac/respiratory arrest only | Within institution; POLST for portability |

### DNR/DNI Orders

Do Not Resuscitate orders direct that no chest compressions or defibrillation be performed in the event of cardiac arrest. Do Not Intubate orders direct that no endotracheal intubation be performed for respiratory failure. Several important nuances govern the interpretation and application of these orders. DNR and DNI orders apply exclusively to cardiac and respiratory arrest and do not limit any other medical treatments, including antibiotics, blood transfusions, surgery, or hospital admission. The terminology "Allow Natural Death" has gained acceptance as an alternative to DNR because it emphasizes allowing the natural dying process rather than withholding care, a framing that many patients and families find more acceptable.

The perioperative management of DNR/DNI orders requires specific discussion with the surgeon and anesthesiologist. Many institutions require either a temporary suspension of DNR/DNI status during surgery or a goal-directed resuscitation approach that specifies which interventions are acceptable in the perioperative context and which are not.

<image>A comprehensive visual guide to advance care planning documents. Show four document types arranged in a comparison grid format. Column 1: Healthcare Proxy/DPAHC — show a document icon with two people, list key features (designates surrogate, most important document, adapts to any situation, legally binding). Column 2: Living Will — show a document with checkboxes, list features (states treatment preferences, limited to anticipated scenarios, may become outdated, complements proxy). Column 3: POLST/MOLST — show the distinctive colored form, list features (medical ORDER not legal document, completed by clinician with patient, portable across settings, for seriously ill patients only, covers CPR/intubation/nutrition). Column 4: DNR/DNI — show a medical order form, list features (applies ONLY to cardiac/respiratory arrest, does NOT limit other treatments, "Allow Natural Death" as preferred terminology). Below the grid, show a Venn diagram illustrating how these documents overlap and complement each other. Include a timeline showing when each document is most appropriate: healthy adults (proxy + living will), serious illness diagnosis (add POLST), actively dying (POLST guides care). Include state-specific variation notes.</image>

## Goals-of-Care Conversations

### When to Have Goals-of-Care Discussions

Goals-of-care discussions should not be reserved for moments of crisis but rather should be initiated proactively at several key junctures in the patient's care trajectory. These include the diagnosis of a serious or life-limiting illness, any significant change in disease trajectory such as a hospitalization, functional decline, or new diagnosis, annually at wellness visits for elderly patients, before major interventions including surgery, chemotherapy, or ICU admission, and whenever the patient or family raises questions about prognosis or the overall direction of care. The Surprise Question serves as a useful clinical trigger: if the clinician would not be surprised if the patient died within 12 months, advance care planning conversations should be initiated or revisited.

### Conversation Frameworks

#### Serious Illness Conversation Guide (Ariadne Labs/Bernacki)

The Serious Illness Conversation Guide, developed by Bernacki and Block at Ariadne Labs, provides a structured, evidence-based framework for conducting goals-of-care conversations. The guide proceeds through ten carefully sequenced steps. The conversation begins with setup, framing the discussion as an important part of medical care. The clinician then assesses the patient's current understanding by asking, "What is your understanding of where things stand with your health?" Information preferences are explored by asking how much the patient wishes to know about prognosis. Prognosis is shared honestly and empathically, tailored to the patient's expressed preferences, using language such as, "I'm worried that time may be limited." The clinician then explores what matters most to the patient, asking, "If your health worsens, what is most important to you?" Fears and worries are explicitly invited. Functional goals are explored with the question, "What abilities are so important that you can't imagine living without them?" Treatment trade-offs are discussed by asking, "How much are you willing to go through for the chance of more time?" The clinician then offers a recommendation that is explicitly aligned with the patient's stated values. Finally, the conversation is documented and shared with the care team, with plans for regular revisitation.

#### VitalTalk Communication Skills

The VitalTalk communication framework provides specific techniques for navigating the emotional dimensions of goals-of-care conversations. NURSE statements offer a structured approach to responding to emotion: Naming ("It sounds like you're feeling scared"), Understanding ("I can understand why you feel that way"), Respecting ("I can see you have been thinking about this a lot"), Supporting ("I will be here with you through this"), and Exploring ("Tell me more about what concerns you"). The Ask-Tell-Ask technique structures information sharing by first assessing what the patient understands, then delivering information in small portions, and then checking for comprehension and inviting questions. The Wish-Worry-Wonder framework provides language for transitioning to difficult topics: "I wish the situation were different. I worry that the cancer is growing despite treatment. I wonder if it would help to talk about what matters most to you."

### Specific Scenarios

#### Code Status Discussion

The framing of code status discussions profoundly influences the quality of decision-making. The question "Do you want us to do everything?" should be avoided because it creates a false binary that implies withholding care and frames the discussion in terms of effort rather than likely outcomes. A more effective approach describes CPR specifically and provides outcome data in context. For example, "If your heart were to stop, CPR involves chest compressions that often break ribs, an electric shock, and placement on a breathing machine. In someone with your medical conditions, the chance of surviving CPR and leaving the hospital is approximately X percent."

The outcomes data for CPR in elderly patients are sobering. For in-hospital cardiac arrest, overall survival to discharge is approximately 25 to 30 percent, but this drops to 10 to 15 percent for patients aged 80 and older and to less than 5 percent for those with metastatic cancer, advanced dementia, or multiorgan failure. Out-of-hospital cardiac arrest survival to discharge is 8 to 10 percent overall and worse in elderly patients. Among survivors, many experience significant neurological impairment.

#### Discussing Artificial Nutrition

Conversations about artificial nutrition should be framed in terms of the patient's goals rather than the feeding modality itself. Asking, "Will this feeding tube help your mother achieve what matters most to her?" redirects the discussion from the technology to the patient's values. In advanced dementia, the evidence is unequivocal: PEG tubes do not prevent aspiration, do not prolong life, and do not improve comfort, as affirmed by the American Geriatrics Society position statement. Careful hand-feeding provides comfort, social interaction, and outcomes that are similar to those achieved with tube feeding while avoiding the complications of PEG placement.

#### Discussing Dialysis

For elderly patients, the initiation of dialysis requires a frank discussion of expected outcomes. Average survival for elderly patients starting dialysis is 2 to 3 years for ages 75 to 79 and 1 to 2 years for those aged 80 and older. The time burden of dialysis, including travel, treatment, and recovery time, is substantial at 15 to 20 hours per week. Conservative kidney management, which involves symptom management without dialysis, represents a legitimate alternative. In some studies, the median survival difference between dialysis and conservative management in elderly patients is only 4 to 6 months, with better quality of life reported in the conservative management group.

## Surrogate Decision-Making

### Decision-Making Standards (Hierarchical)

When a patient lacks decision-making capacity, surrogate decision-making operates according to a hierarchical framework. The highest standard is the patient's own expressed wishes, also termed substituted judgment based on specific prior statements. If the patient previously articulated what they would want in the specific clinical situation at hand, whether orally or in writing, those wishes should be followed. The second standard, substituted judgment, applies when the surrogate must infer what the patient would have chosen based on knowledge of the patient's values, beliefs, and life preferences, even though the patient did not address the specific scenario directly. The third and least specific standard, the best interests standard, is employed when the patient's values are unknown, requiring an objective assessment that weighs the benefits and burdens of each treatment option.

### Challenges in Surrogate Decision-Making

The systematic review by Shalowitz and colleagues in 2006 demonstrated that surrogates predict patient treatment preferences with approximately 68 percent accuracy, a finding that underscores the importance of prior ACP conversations in improving concordance. Surrogates frequently project their own values and preferences onto the decision, further reducing accuracy. The emotional burden of surrogate decision-making is substantial, with 30 to 40 percent of surrogates experiencing PTSD-like symptoms, guilt, and complicated grief. Clinicians can reduce this burden by reframing the surrogate's role as representing the patient's values rather than making the decision themselves. The statement, "You are not deciding whether your mother lives or dies. You are telling us what she would want," acknowledges the weight of the situation while relieving the surrogate of the burden of authorship.

### Capacity Assessment

Decision-making capacity is both decision-specific and time-specific. A patient may have capacity for some decisions while lacking it for others, and capacity may fluctuate with time of day, medication effects, or the resolution of delirium. The Appelbaum criteria define four functional abilities that constitute decision-making capacity. Understanding requires that the patient comprehend relevant information about the condition and proposed treatment. Appreciation requires that the patient recognize how the information applies to their own situation. Reasoning requires that the patient weigh risks and benefits and consider alternatives. Expressing a choice requires that the patient communicate a consistent decision. Capacity is a clinical determination that can be made by any physician; competency is a legal determination made by a court. When capacity is uncertain, psychiatry or ethics consultation may clarify the assessment, and serial evaluations may be warranted, such as reassessing after delirium has resolved.

<image>A surrogate decision-making hierarchy diagram. Show a vertical hierarchy with three levels arranged as a pyramid or inverted triangle. TOP (most authoritative): "Patient's Expressed Wishes" — direct statements, written directives, POLST → "Follow what the patient said they wanted." MIDDLE: "Substituted Judgment" — what would the patient choose based on known values? → Surrogate uses knowledge of patient's values, beliefs, and life experiences to infer preference. BOTTOM (least specific, used when values unknown): "Best Interests Standard" — objective weighing of benefits vs. burdens → Used when there is no information about what the patient would have wanted. Include arrows showing the flow from most to least authoritative. On the right side, show a practical application example: an 85-year-old with advanced dementia who develops pneumonia — show how each standard would guide the decision about antibiotic treatment and hospitalization vs. comfort care. Include a callout box: "Reassure surrogates: 'You are not making this decision — you are helping us understand what your loved one would want.'"</image>

## Cultural and Special Considerations

Cultural diversity introduces important variations into the advance care planning process. Truth-telling preferences vary across cultures, with some traditions favoring protective non-disclosure to the patient while prioritizing family-centered information sharing. Family decision-making models differ significantly, with many cultures preferring collective family deliberation rather than the individual-centered autonomy model dominant in Western bioethics. Attitudes toward life-sustaining treatment, death, and the afterlife are deeply shaped by spiritual and religious traditions. Clinicians should actively explore these dimensions by asking, "In your family, how are medical decisions usually made?" and "Are there cultural or religious traditions that are important to consider?"

Language barriers require the use of professional medical interpreters for advance care planning conversations. Family members should never serve as interpreters for these sensitive discussions due to the risks of information filtering, emotional distress, and role conflict. Health literacy must be addressed through the use of plain language, visual aids, and the teach-back method. LGBTQ+ patients may face particular legal vulnerabilities if their chosen family or decision-makers are not recognized under state law, making formal documentation of healthcare proxy designation especially critical. Patients experiencing homelessness face unique barriers to ACP completion and POLST portability that require creative problem-solving and advocacy.

## Documentation and Implementation

The documentation of advance care planning must be both thorough and accessible. ACP discussions should be recorded in the medical record with specifics about what was discussed, who was present, and what preferences were expressed. Advance directives must be accessible to all clinicians who may participate in the patient's care, requiring that they be scanned into the electronic health record, that copies be provided to the designated surrogate and primary care physician, and that registration with state advance directive registries be completed where available.

POLST forms must travel with the patient, be prominently displayed in the home (commonly on the refrigerator or at the bedside), and accompany the patient during all care transitions. Periodic review of advance care planning documents is essential, occurring at minimum annually, with any significant change in health status, or when transitioning between care settings. All documents must be internally consistent: the advance directive, POLST form, code status orders, and electronic health record documentation should align, and any discrepancies should be identified and resolved through conversation with the patient or surrogate.

## Key Clinical Pearls

- The healthcare proxy is the single most important advance directive — a trusted surrogate who knows the patient's values can navigate any situation, even one not anticipated by written documents
- ACP is a process, not a form — conversations should occur repeatedly as health status changes; a form completed 10 years ago may not reflect current wishes
- POLST is for seriously ill patients ONLY — it is NOT a replacement for advance directives in healthy older adults
- Framing matters: "Do you want us to do everything?" is coercive and misleading — instead, describe specific interventions, their likelihood of success, and what recovery looks like
- Surrogates predict patient preferences with only 68% accuracy — prior ACP conversations dramatically improve concordance
- CPR survival in elderly with serious illness is far lower than the public perceives — provide honest data and frame within the patient's specific clinical context
- Support the surrogate: decision-making burden causes PTSD-like symptoms in 30-40% — reassure them that their role is to represent the patient's values, not to bear the weight of the decision

## References
1. Sudore RL, Lum HD, You JJ, et al. Defining advance care planning for adults: a consensus definition from a multidisciplinary Delphi panel. *J Pain Symptom Manage*. 2017;53(5):821-832.
2. Bernacki RE, Block SD. Communication about serious illness care goals: a review and synthesis of best practices. *JAMA Intern Med*. 2014;174(12):1994-2003.
3. Hickman SE, Keevern E, Hammes BJ. National POLST paradigm initiative. *J Palliat Med*. 2015;18(1):6-7.
4. Shalowitz DI, Garrett-Mayer E, Wendler D. The accuracy of surrogate decision makers: a systematic review. *Arch Intern Med*. 2006;166(5):493-497.
5. Silveira MJ, Kim SYH, Langa KM. Advance directives and outcomes of surrogate decision making before death. *N Engl J Med*. 2010;362(13):1211-1218.