Talking the Talk

A new survey asking people about their wishes for end-of-life care has been getting quite a bit of publicity lately. Commissioned by “The Conversation Project,” a laudable grassroots effort to encourage families to talk to each other about the kind of medical care they would want in their final days, the poll reports some interesting observations—and raises some important questions.

Like other earlier studies, this one finds that most adults (94%) think it is important to talk about end-of-life care. An extensive California study, for example, found that 83% of the adults surveyed thought it was important to make their end-of-life wishes known, A poll of adults in Massachusetts found that 84% of adults were comfortable talking about dying. 

And like earlier studies, this one finds that though people think that talking about dying is important, many of them don’t actually talk about it (in the new study, only 27% did). In California, only 36% of adults actually had something in writing about their wishes; in Massachusetts, 51% had had a conversation with family members.

But if we look at people who are very sick or very old or both and ask whether they talked to their families about their wishes, the picture that emerges is a bit different. The Pew Research Center conducted a national poll in 2005 (with 1500 subjects) and another one in 2009 (with 2969 subjects) and found that while only 11% of people aged 18-29 had any kind of written documentation of their end of life wishes, 51% of those over 65 had such a document. Moreover, 63% of older Americans had talked to their adult children about their wishes for medical care in the event of incapacity—and among older women, 71% had had such conversations. So when the new study from the Conversation Project reports that only 27% of those polled had talked to family members about their personal wishes, is this too low? 

While I agree that it’s a good idea for everyone to designate a health care proxy—to state who will make medical decisions on their behalf if they are unable to do so themselves—I don’t think it makes sense for everyone to have “the conversation” about end-of-life care. The overwhelming majority of Americans will die in old age: people over 64 account for 12% of the population but 70% of the deaths;  I estimate that asking everyone who is between 18 and 64 to talk about their wishes for medical care at the end of life means that about 280 people will have such a conversation for every person who ends up actually dying in the next year.  It’s far more reasonable to target advance care planning to those people who are likely to get very sick in the relatively near future, which means those who are elderly or who already have a life-limiting illness such as advanced heart failure or metastatic cancer. Moreover, if 20- or 30-year-olds actually did have “the conversation,” they would likely express preferences that will change by the time they are old and dying. How much suffering you are willing to endure in exchange for a small chance of living longer or how much debility you find tolerable may be very different, depending on whether you are already 80 and have outlived most of your birth cohort and on your underlying health status at the time you develop a terminal illness. 

If pitching discussions about dying to everyone is excessive, talking only about dying isn’t enough. Once it becomes clear that a person is truly dying—and many people who are at that stage have difficulty accepting that the end is imminent, a situation made worse by doctors who have difficulty telling them where they stand—the vast majority of people don’t want more tests and treatments. The default approach to people who will surely die within a few months no matter what is done should be a focus on comfort. Conversations are needed principally to communicate that the end is near and to identify the small minority who do want trials of treatment that have only a very, very small chance of benefit. 

What we need desperately to talk about is the final phase of life, whether that is measured in months or in years, a time demarcated by a marked decline in the ability to function independently. The 85-year old with heart failure, diabetes, arthritis, and kidney problems may not be dying in any conventional sense of the term, but she is likely to develop some kind of acute illness in the near future, whether pneumonia or dehydration or an exacerbation of her heart condition, and thinking in advance about the approach to medical care that is right for her is critically important. She has real choices to make: does she want maximal medical therapy? Comfort-oriented treatment? Or something in between? As Katy Butler makes so poignantly clear in her recent book, Knocking on Heaven’s Door, this final phase can last a long time—it was 7 years between the time when her father got his pacemaker and death—but decisions made along the way dramatically shape the experience of those years.

So I’m not at all surprised that 94% of the 1067 people surveyed for The Conversation Project said that it is important to talk to your loved ones about your end-of-life care wishes but only 27% have had a discussion about “what they do and don’t want in their final days.” I applaud the Conversation Project for encouraging people to open the door to discussing difficult topics. But let’s be selective about whom we invite in and what we talk about.