Terminally ill patients in the U.S. these days face expensive care and prolonged declines. And many hospitals lack a designated palliative care team, which focuses on patient comfort at any point in the life cycle but becomes more critical at the end. In terms of care, the solution to our medical system's woes might be as simple as a conversation about how we die.

"Many clinicians are afraid to talk about prognosis—how long a patient may have to live," says Alexander Smith, a palliative care physician at the University of California, San Francisco. "Talking about death in America is forbidden, a taboo topic."

Voices clamor to be heard in the discussion about end-of-life care. Patients prefer to die at home and in comfort. Physicians struggle to speak to patients they have been trying to save but worry about legal issues. Politicians debate the best way to curb rising health care costs and shun "death panels." Amid this fervor, researchers in Sweden found last year that conversations about imminent death are associated with improved care and less pain and suffering. Now, a team at Harvard Medical School reveals more detail about the social and psychological factors that affect quality of death.

The researchers interviewed 396 advanced cancer patients and their caregivers, and then followed up a few months after death. Nine psychosocial factors were found to make a significant difference in the quality of patients' final days.

In surveys, patients define (pdf) a good death as one where they can manage their pain, remain lucid, feel prepared and interact with family members. In the new study, patients with a higher quality of life during their final days included those who avoided hospitalizations and stays in the intensive care unit (ICU), were not worried about death, prayed or meditated, received care at home, did not use a feeding tube, were visited by a pastor in the hospital or clinic, did not undergo chemotherapy in the final week of life, and felt an alliance with their oncologist regarding therapy. The results were published online July 9 in the Archives of Internal Medicine.

ICU stays in the final week of life mattered the most, accounting for 4.4 percent of variance in outcomes, whereas a sense of alliance between patient and physician accounted for 0.7 percent. Much of the variance remained unexplained. Other influences could include underlying cancer biology or even a patient's personality and general outlook on life, says Holly Prigerson, a psychiatry associate professor at Harvard Medical School and the study's senior author. But, these factors could give policymakers and hospitals a cost-effective road map for end-of-life care decisions in the future. Prigerson cites the frustration and helplessness that oncologists feel when their patients are beyond curative care. Withdrawal, however, apparently is not the answer. "Just being present and being there, rather than abandoning patients and withdrawing," she says, "matters a tremendous amount to patients."

But not all of the findings have clear policy implications, says Diane Meier, a professor of geriatrics and internal medicine and director of the Center to Advance Palliative Care (CAPC) at Mount Sinai School of Medicine in New York City. How much a patient worries or whether they find comfort in prayer may be hard to change. But already, physicians are trying to keep patients approaching death out of the ICU. Meier says she expects more clinics, hospitals and offices to welcome pastors. Finally, training physicians to convey a sense of commitment and alliance with their patients can help. "These are teachable skills that right now are not in the curriculum at medical schools," she says.

Medical ethicists still debate how to best communicate end-of-life options to patients. But some states, such as New York and California, mandate that physicians start the conversation and offer information on palliative care designed to improve patient comfort. Such discussions could also reduce costs. In a 2009 study, Prigerson's team found the cost of care in a patient's final week was $1,041 (35.7 percent) less, if he or she had an end-of-life discussion with doctors.

A reluctance to discuss end-of-life decisions may lead to care that is too little and too late. Half of all adults older than 65 years of age made an emergency room (ER) visit in the last month of life, according to a study in the June issue of Health Affairs. Data from the Health and Retirement Study was linked to Medicare claims to track ER use and death. Three quarters of the emergency visits led to hospitalizations and of those admitted, two thirds died there.

Hospital emergency departments focus on triage and stabilization and are not equipped to deal effectively with patients near the end of life due to illness or age, says Smith, the study's lead author. The visit comes at a time of crisis for these patients and their family members, he explains. Many are not prepared for the end of life. For instance, only 10 percent of the study participants were enrolled in hospice care. "We've come a long way from where we were 10 to 15 years ago," Smith says. But, he adds, clinicians need to develop more robust palliative care programs and focus on care outside the hospital, such as in doctor's offices and in the home.

Improvement on the way
One of the topics physicians could and often do discuss is palliative care, which offers relief and comfort, such as medicine to tame nausea. Large hospitals are most likely to have palliative care teams, finds a 2011 report from the CAPC. The report gave the nation as a whole a B grade—up from a C in 2008—observing that less than a quarter of all states still need significant improvement. However, there is only one palliative care specialist for every 1,200 people living with a serious or life-threatening illness in the U.S. compared with one oncologist for every 141 newly diagnosed cancer patients. Additionally, only 54 percent of public hospitals provide palliative care.

That access could be extended under the Patient Protection and Affordable Care Act of 2010 (pdf), recently affirmed by a Supreme Court ruling, says Kathy Cerminara, a law professor at Nova Southeastern University in Florida and an expert in law governing treatment at the end of life.

Two provisions in the act explicitly address hospice care, a common way for patients near death to receive palliative services. The first provision authorizes the secretary of the U.S. Department of Health and Human Services to fund a study determining whether patients in hospice programs would benefit from offering curative and palliative care simultaneously (For now, hospice care typically requires patients to forgo curative treatment in order to enroll.)

The second provision is likely to challenge rural health providers, an already struggling sector, according to the palliative care report card. The provision requires reevaluation for any patient remaining in hospice care after six months. Hospice programs that show a high rate of reenrollment can face review and potential fines. Medically underserved areas, including rural stretches of many states, struggle with short staffing and a lack of hospice providers already, Cerminara explains. Requiring face-to-face encounters for patients hoping to remain on hospice may prevent professionals from recommending service in the first place. "Congress has a good goal in trying to prevent fraud and abuse, but they're doing it in a rough way," Cerminara says.

Overall, a change in end-of-life care appears to be on the horizon, Prigerson says, that has more to do with patients' social and psychological needs than conventional biomedicine. "There has been a dramatic shift in the recognition of the importance of attending to the psychosocial needs of patients who are dying," she says. But patients still could benefit from more consistent dialogue about their choices, not to mention wider access to care.