Exploring patient-defined goals of care is the first step in determining the most appropriate therapeutic interventions for a specific disease or symptom. An organized approach to goal setting can help the clinician and patient achieve clearly articulated goals. Goal setting is best accomplished through meeting with the patient and family or surrogate decision-maker (Table 126-2). Before a goal-setting meeting, the clinician should review the disease course, response to prior treatments, and potential for further disease-modifying treatments and develop a realistic short- and long-term vision for the future clinical course, including a general sense of prognosis. With this in mind, the clinician can begin to review treatment options and help the patient decide which treatments are most likely to help meet his or her specific goals. All therapeutic options should be examined in light of the question, “Does the intervention match or assist with the patient’s treatment goals?”⁶ As the burden of decision-making increases near the end of life, it is important for physicians to understand their central role in helping patients make decisions. A model of shared decision-making, in which the physician provides guidance and recommendations, generally is preferred to a paternalistic approach, or, at the opposite extreme, to one in which options are presented with no guidance.⁷

Table 126-2 Process Steps for a Goal-Setting Family Meeting

1. Determine the reasons for convening a family conference.

Clarify the goals of care and review the patient’s medical condition.

Decide on future levels of care and treatments, and resolve family conflicts.

2. Determine who should and who will be present for the conference.

Include appropriate health care providers (e.g., nurse, chaplain, social worker, physician consultants, primary care physician).

Ask the patient, or health care power of attorney, who he or she would like to participate (e.g., the designated health care power of attorney, appropriate family members, clergy, lawyer, friend).

Family and other participants should be given significant time (often 48 hours) to prepare for the meeting.

3. Determine whether the patient has decision-making capacity.

Able to understand information about diagnosis and treatment?

Able to evaluate alternatives and compare risks and benefits?

Able to communicate a choice—verbally, in writing, or with a nod or gesture?

4. Choose the proper physical setting.

5. Introduce yourself, explain your relationship to the patient, and invite all participants to do the same.

6. Identify the legal decision maker, if available.*

7. Review the goals and purpose of the meeting.

8. Establish ground rules.

Everyone will have the opportunity to talk.

No interruptions are permitted.

9. Review the patient’s current medical condition.

“What is your understanding of ____’s present condition?” or “What have you been told about ____’s condition?”

Review with the patient and family the current medical condition (e.g., expected prognosis and potential treatment plans). Avoid medical jargon.

Invite questions.

Defer treatment decision-making until all questions about the patient’s medical status have been asked to the extent possible.

10. Family discussion guidelines include the following:

When the patient can speak for himself or herself:

Ask the patient what he or she is considering.

Ask the patient what type of support he or she would like from family members and from the health care team.

Invite discussion from other family members.

11. Follow these guidelines when there is no consensus.

Ask the family to discuss the issue on their own and schedule a follow-up meeting (use time as an ally).

Ask each family member on what values his or her decision is based and how the decision will affect the patient and the other family members.

Review again the goals you are trying to reach: What would the patient say if he or she were able to speak?

Discuss other resources to support decision-making.

* Laws governing surrogate decision-making vary from state to state.

Adapted from Weissman DE, Ambuel B. Establishing treatment goals, withdrawing treatments. In: Weissman DE, Ambuel B, Hallenback J, editors. Improving End-of-Life Care. 3rd ed. Milwaukee: Medical College of Wisconsin; 2000. p 101.

PROGNOSTICATION

Physicians’ estimates of patient survival are important to physicians as well as to patients and families in all phases of a patient’s life because they influence both medical and nonmedical decisions. This is especially true at the end of life when patients and families have to make numerous personal, fiscal, and social arrangements in anticipation of impending death.

Despite the great need for accurate prognoses, prognostication remains an elusive clinical art. Numerous empirical studies have revealed a consistent optimistic bias,⁸ with most physicians overestimating anticipated life span by a factor of three. Physicians’ prognostic radars are further skewed when they are more connected with the patient. Consequently, patients who are in great need of quality palliative care are referred too late to palliative care services and thus are deprived of access to good symptom management.

There are two key tasks in prognosticating: 1) formulating the prognosis by reviewing patient-specific medical factors and disease-specific actuarial estimation of survival; and 2) communicating the prognosis to patients and families while providing ongoing support.

Given that prognostication is challenging, many physicians are uncomfortable with this task and often avoid providing realistic prognostic information, or they continue to offer futile treatment options that can convey false hope and unrealistic expectations to patients and families. When pressed by patients, physicians then typically overestimate survival. If the imminence of death is not discussed honestly, patients might be more likely to accept costly, burdensome, and futile treatments.⁹

Prognostication guidelines are well established for cancer.¹⁰^,¹¹ The single best prognostic variable in cancer is performance status.¹⁰^,¹¹ For example, patients with a Karnofsky Performance Status (KPS) of 40 (disabled; requires special care and assistance) live on average less than 50 days; and patients with a KPS of 20 live an average of only 10 to 20 days (Table 126-3).¹⁰ Put another way, patients who are spending more than 50% of the day resting or in bed generally have a prognosis of three months or less. Specific symptoms provide further information: symptoms with an independent predictive value for a poor prognosis are shortness of breath, anorexia, difficulty swallowing, and weight loss.¹⁰

Table 126-3 Karnofsky Performance Scale

%	LEVEL OF FUNCTIONAL CAPACITY
100	Normal, no complaints, no evidence of disease
90	Able to carry on normal activity, minor symptoms or signs of disease
80	Normal activity with effort, some symptoms or signs of disease
70	Cares for self, unable to carry on normal activity or to do active work
60	Requires occasional assistance, but is able to care for most needs
50	Requires considerable assistance and frequent medical care
40	Disabled, requires special care and assistance
30	Severely disabled, hospitalization is indicated although death is not imminent
20	Hospitalization is necessary; patient is very sick; active supportive treatment is necessary
10	Moribund, fatal processes progressing rapidly
0	Dead

From Karnofsky, DA, Burchenal, JH. The clinical evaluation of chemotherapeutic agents in cancer. In: MacLeod CM, editor. Evaluation of Chemotherapeutic Agents. New York: Columbia University Press; 1949. p 196.

Prognostic criteria for noncancer diagnoses have been published and are especially useful to help physicians know when to refer patients for hospice services.¹² Specific to gastroenterology, for example, general criteria have been established for chronic liver disease (Table 126-4). Beyond guidelines, some clinicians have advocated a simple test to determine when hospice services are appropriate by asking, “Would I be surprised if this patient died in the next six months?”¹³^,¹⁴

Table 126-4 Hospice Eligibility Criteria for End-Stage Liver Disease

The patient is not a transplant candidate and opts for comfort care and

BUN, blood urea nitrogen; HBsAg, hepatitis B surface antigen; INR, international normalized ratio.

Adapted from Standards and Accreditation Committee. Medical Guidelines for Determining Prognosis in Selected Noncancer Diseases. 2nd ed. Arlington, Va: National Hospice Organization; 1996.

No matter what type of cancer or noncancer fatal illness a person has, a “common final clinical pathway” occurs in most patients.¹⁰ Signs and symptoms that predict a prognosis of hours to days are decreased or fluctuating levels of consciousness, a precipitous clinical decline, decreased oral intake, and inability to turn over in bed.¹¹ Patients close to death typically exhibit periods of apnea, retained oropharyngeal secretions (the death rattle), fever, and cool or mottled extremities.¹¹

Discussing prognosis with patients and families is a key skill in palliative care. Physicians are advised to start by asking patients if they previously have been given prognostic information, if they have a sense of how much time is left, and whether they would like to discuss prognosis. If a patient indicates that he or she wishes to discuss prognostic information, provide a broad estimate, a few days to a few weeks or a few weeks to a few months, rather than, “Mr. Jones, you have only three weeks to live.”⁴ Once the time frame is presented, important future goals can be determined by asking, “What do you want or need to do in the time that is left?” (e.g., important events, saying goodbye to loved ones). This allows the clinician to aim the information at the level of the patient and family.¹¹

KEY PROGNOSTIC VARIABLES AND TOOLS IN LIVER DISEASE

MODEL FOR END-STAGE LIVER DISEASE SCORES

The Model for End-Stage Liver Disease (MELD) (see Chapter 95) is a numeric scale used to prioritize patients for liver transplantation.¹⁵^–¹⁸ It is calculated by a formula using the serum bilirubin, international normalized ratio (INR), and the serum creatinine. The MELD has been validated for short-term and intermediate-term mortality in a heterogeneous group of patients with liver disease. Modifications of the MELD score using serum sodium values have been suggested; these might enhance its prognostic ability for patients with cirrhosis.