End-of-life care preferences: A promising target for reducing racial disparities in terminal cancer care

Wednesday, September 10, 2014

A patient's race and ethnicity dramatically influence their preferences about the care they receive at the end of their lives, which in turn significantly affect the likelihood they will complete a do-not-resuscitate order (DNR), say a team of researchers led by a Weill Cornell Medical College investigator.

In a study published Aug. 21 in Cancer, the researchers found that in a group of 234 patients with terminal cancer, 45 percent of white patients signed a DNR order compared to only 25 percent of black patients and 20 percent of Hispanic patients.

This degree of racial and ethnic disparity in signing the DNR is not new — many studies have found the same rates — but identifying the reasons for it are, says Dr. Holly Prigerson, who was recruited to Weill Cornell as the Irving Sherwood Wright Professor in Geriatrics. The majority of black and Hispanic terminal cancer patients, compared to white patients, were willing to risk aggressive intervention in the hopes that it would buy them more time, the researchers found. They would not sign off on an advance directive that would limit their lives and their prospects for a miraculous cure.

Researchers teased out patients' rationales by probing their treatment preferences and asking them if they had signed a DNR order. They asked, among other things, if they would want to be kept alive by chemotherapy, ventilation, tube feeding or antibiotics, if each method would extend life by one day, one week, or one month.

Patients' choices about life-prolonging care differed dramatically by race and ethnicity. For example, 80 percent of Latino patients without a signed DNR wanted to be on a ventilator if it prolonged life for one day compared to 30 percent for non-Latino white patients; 46 percent of black patients without a DNR indicated a preference for ventilation if it would extend life for one week compared to 30 percent of non-Latino white patients.

Dr. Holly Prigerson Photo credit: Roger Tully The issue of treatment preferences is important, says Dr. Prigerson, because terminally ill patients who do not actively engage in planning their end-of-life care often receive aggressive, burdensome, costly and futile care at life's end. Many of these patients are black and Hispanic.

Aggressive care given to terminal patients impairs patients' quality of life, Dr. Prigerson says. But if terminal patients knew their prognoses and understood that interventions such as use of ventilation, feeding tubes, and chemotherapy inhibit a patients' ability to talk, eat, swallow and even breathe for themselves - amounting to poor quality of life - they might modify their original treatment preferences for end-of-life care and perhaps be more willing to sign a DNR order, she says.

Religious beliefs appeared to play a substantial role in the decision not to sign a DNR, especially among black patients, Dr. Prigerson adds. "Patients who want life-prolonging care often comment that they are holding out for a miracle and that their faith prohibits them from completing an advance directive that would withdraw or limit life support," she says. "Religion is a primary resource and comfort for many patients confronting life-threatening illnesses; it helps patients cope."

The results suggest that counseling from church representatives may help patients understand how their choices will affect the care they receive in their last days, she says. "It is not that religious patients don't trust their doctors --they do -- but their preferences are being shaped by members of their religious community more than by members of the medical community.

"Recognizing the basis for a patient's preference for treatment may help us reduce the racial disparity we see in the intensity of end-of-life care," she says.

Working with Dr. Prigerson on the study was Dr. Melissa Garrido of the James J Peters VA Medical Center and the Icahn School of Medicine at Mount Sinai, and Dr. Shannon Harrington of the Dana-Farber Cancer Institute and the University of Massachusetts in Boston.

This work was supported by a grant to Dr. Prigerson from the National Cancer Institute.