End-of-life planning too often inadequate

Conversation rarely happens in advance

The past 50 years in medicine have brought amazing advances in technology and pharmacology that have been able to defer death for many more people until much later in life, notes Nancy E. Havas, MD, FAAFP, associate professor at the Center for Bioethics and Medical Humanities at Medical College of Wisconsin in Milwaukee.

“The common occurrences of family members dying at home from a heart attack, stroke, or cancer, or even a bad infection, have become very uncommon in our modern health care experiences. This results in death being something that happens ‘later’ and something that we don’t often encounter,” says Havas. “But the reality is that it eventually happens to 100% of all patients, and is something we should plan for.”

Because dying is now such a foreign experience for so many people, it is an uncomfortable topic of discussion, says Havas. “This makes the conditions that often precede death, such as dementia, respiratory failure, and functional decline, very uncomfortable for us to discuss, too. We don’t want to imagine ourselves in that situation,” she says. “As humans, we avoid those uncomfortable conversations for as long as possible.” This means discussions often happen too late when patients can’t participate in the decisions that directly impact their futures.

Havas says that many times, patients tell providers and their families something along the lines of, “I don’t want to be living on machines. If it gets to that point, pull the plug!” “But the reality is that most people become disabled somewhere between playing golf and independent and living on a machine,” says Havas. “I see many people who are unable to talk, or swallow enough food to survive, who are no longer able to interact with their families.” These individuals cannot participate in the discussion about having a feeding tube placed to supplement nutrition, for example.

“Their families struggle terribly about making choices,” says Havas. “If this scenario would have been discussed before the patient was disabled, the family would have known that the patient would not want a feeding tube, and would be at peace with the decision making, even with the knowledge that the patient would continue to decline.”

Conversation needed earlier

Havas says that too often, discussions occur during “acute” times of decision making, such as during a hospitalization. Having these discussions early, preferably with the facilitation of a health care provider, generally results in a better discussion about goals and preferences than those made during the stress of a sudden change in condition.

Often, physicians are waiting for patients to tell them when they want to have the discussion, while families believe that the physician will initiate the discussion at the appropriate time. “So neither of them initiate the discussion until the discussion has to be held,” says Havas. “Ideally, these conversations would be held routinely every three years, every time a major life change or serious health concern warranted revisiting the conversation, such as a new diagnosis or recent hospitalization.”

When Havas becomes involved as a palliative care consultant, she finds that these conversations have rarely been held in advance. “When I work with families who need to make choices for a loved one who is unable to participate, it’s important for me to know who the patient is and what they enjoyed doing before I met them as an incapacitated person,” she says.

Havas asks questions such as “Was she active?” “Did she enjoy socializing?” “Did she like to hike in the mountains?” “Or was she more of a solitary person, reading or watching movies?” “What does she like to do?” “How has her health been during the past year, and what has she said about it?” “What does she value?” “How do her spiritual or religious beliefs help her make decisions?”

“Going back to the preferences and beliefs of the individual patient helps us provide care that is high-quality and ethically sound,” says Havas.

Then, Havas asks additional questions about how the patient would feel about her specific situation, such as, “If Susan were sitting here with us and knew that she would have to have her leg removed to recover from this infection, would she be willing to do that knowing she could have more time with her children, but also knowing she would never walk again?”

Havas says the final part of the conversation should include a discussion about code status and should include a recommendation from the treating physician if relevant to the situation, and should always be framed in the context of the overall goals. “As physicians, we do a very poor job in making recommendations on this,” she says. “We are quick to recommend medications that may have serious side effects, and yet hesitate to recommend against an intervention that has a very poor success rate.”


• Nancy E. Havas, MD, FAAFP, Associate Professor, Department of Family and Community Medicine/Associate Professor, Institute for Heath and Society, Bioethics and Medical Humanities, Medical College of Wisconsin, Milwaukee. Phone: (414) 456-4318. E-mail: nhavas@mcw.edu.