The Purple Heart

Mr. O's face was gaunt, his voice barely audible. It had been an exhausting few days. A serious infection had stressed his failing kidneys and weak heart. Yet despite his fragility, his smile was bright and his eyes were attentive.

He arrived at our veterans affairs intensive care unit after his infection had worsened on the medical floor. We had supported his blood pressure and used antibiotics to stabilize him. Throughout this time, he had been confused and sleepy, and we had not been able to communicate with him directly about his care. A decision to make him do not resuscitate or intubate had been made with a brother who had not spoken to him in 20 years. Thus I was anxious to have this opportunity, now that his delirium had cleared, to readdress his wishes.

The conversation seemed to be going well. He was grateful to hear the summary of his medical progress and eager to understand the next steps forward. Unlike many patients, he was attuned to the fact that his body was failing him, and that he was nearing the end of his life. As we continued to talk, I felt confident that he would agree with our recommendation against aggressive interventions and transition to hospice. However, when we arrived at that moment in the conversation, his eyes widened and he used what little energy he had to lift his head off of his pillow.

I served my country and I deserve the same care as everyone else.

I agree sir, I replied, somewhat taken aback by his sudden shift in tone. I prompted him to further explain what he meant and his response was direct and simple:

You have to confront the racist inside of you.

I paused, stunned to have so easily arrived at what I felt was a looming and inevitable influence on our discussion. I am white, and he is black. And I was recommending that we limit our interventions. The historical context of this was not lost on me. I know that for centuries African Americans have faced second-rate medical care, and at times the purposeful withholding of curative therapies for the sake of research. ¹ But as we sat face to face, I felt unprepared to confront his concerns.

I acknowledged the significance and validity of his statement. I tried to reassure him that his race was not influencing my recommendation. But as I continued to think about this complicated interaction, I questioned whether I could honestly reassure him that this was true.

Among Medicare beneficiaries who died in 2012, 46% of whites, 37% of Hispanics, and 34% of African Americans were enrolled in hospice at the time of their death. ² This disparity in hospice utilization between racial groups seems to be multifactorial.

Structural barriers to hospice enrollment provide one explanation. Patients of color may have less exposure to hospice care models ³ and hospice staff are less likely to be minorities or speak a patient's first language. ⁴ Although hospice is a Medicare covered benefit accessible to people of varied incomes, home hospice also requires the provision of 24-hour care that must be either paid for or provided by a loved one, which is not feasible for many low- and middle-income families.

There is also significant data showing that minorities may be more likely to have negative experiences with end-of-life care.^5,6 One large survey reported that nonwhite hospice participants have worse quality outcomes with regard to pain control and treatment of dyspnea. ⁷ In another multicenter chart review, discussions with nonwhite patients were more likely to involve discord between family members and clinicians. ⁸ And another study showed that wishes expressed in end-of-life conversations with patients of color were not as likely to be respected later in medical management. ⁹

After a long discussion, Mr. O finally ended up deciding against any further aggressive measures. He agreed that he didn't want to spend any of his remaining time connected to machines that would not make him better.

Throughout his time under our care, I spoke to his brother periodically. He was eager to share with me that Mr. O had been injured in the Vietnam War, but had never received the Purple Heart he felt he had deserved. His brother had even visited the benefits office to plead his case, but had not been successful. There was great sadness in his voice as he described Mr. O's sense of abandonment by an institution he had risked his life for. It resonated so powerfully with his current fear that the same institution might not give him the care afforded to others.

I continued to ponder how my own conscious and unconscious biases could be impacting my care of Mr. O and other patients. I wondered whether providers played a role in the racial disparities seen in hospice enrollment. I was surprised to find that that although minorities are less likely to enroll in hospice, we are not less likely to make the referral.

In one study of patients with metastatic cancer, physicians referred African Americans for inpatient palliative care consultation sooner and more frequently than their white patients. This study also showed that after controlling for illness severity, black patients had a longer time between time of consult and death. ¹⁰ A survey of a large Florida hospice network showed that both African Americans and Hispanics tend to have longer enrollments in hospice services than whites. ¹¹ And in another multicenter study, end-of-life conversations with nonwhite patients were more likely to include a recommendation to withdraw aggressive care. ⁸

These data concern me. Longer stays in hospice are an encouraging finding from a palliative care perspective, since later referrals often prevent patients from receiving the full benefits of hospice services. However, that there is a racial disparity in time to referral makes me fearful that even determinations of prognosis are not immune to the effects of bias. I began to wonder whether Mr. O's concerns were not just the result of his mistrust. What if we are giving up sooner on our patients of color?

It has been over a year since I met Mr. O, and I have been reading about his progress ever since. Each time I opened up his chart, I feared the warning banner that pops up when a patient has passed. But it never did. And so just last month I went to visit Mr. O in hospice. I shared these written words with him and asked his permission for their publication. It was a privilege to see his smile again, and a humbling reminder of just how bad doctors are at guessing how long someone has left to live.

Mr. O articulated what many of my patients may be too reluctant to disclose—a fear that skin color will impact care decisions, including those surrounding how and when a person may die. To equitably serve our patients of color, we must first educate ourselves to the historical context that feeds these fears as well as the continued evidence that racial inequities in healthcare access, quality, and outcomes are far from eliminated. ¹² We must also critically assess our own capacity for bias, and the discomforting possibility that it may influence the important decisions we make at the end of a patient's life.

I will always admire Mr. O's bravery—both in the sacrifices he made for his country and the risk he took to honestly confront his doctor with his concerns. He taught me that our patient's trust is built from more than our clinical judgment and communication skills. It also comes from our curiosity toward our patient's stories, our willingness to hear their fears, our vigilance to our own biases, and our dedication to fight for each life with an equal fervor.