My 64-year-old patient with terminal cancer and less than six months to live wanted to go to Oregon. He was contemplating assisted suicide, which is legal there. “My life has been long and good,” he said. “I believe it is my right. I want the ability to say it’s too much, I can’t do it anymore. A person should have a dignified quality of life.”

Another one of my patients, an 84-year-old woman from a nursing home, had heart failure, lung failure and kidney failure. She lay in her bed on a ventilator and on a dialysis machine with little hope for survival.“We want everything done,” her daughter insisted. “It’s in God’s hands, and God can do miracles.” For weeks we continued aggressive and ultimately futile efforts to keep her alive.

We will all die, but the trajectory of our death will follow one of a few predictable patterns. According to statistics published in 2003 in the Journal of the American Medical Association, two-fifths of us will die with prolonged, dwindling illness on a slow downward slope typical of dementia or frailty. One-fifth will die with a sharp decline typical of, for example, metastatic cancer. Another fifth will die with intermittent dips, like a roller coaster, from heart or lung failure; and a small percent will die suddenly and unexpectedly, like falling off a cliff.

Nearly 80 percent of us want to die at home, polls have shown, but most die in the hospital, often strapped to a bed in the intensive care unit. And a recent study suggests that many people are subjected to surgery, even in the last week of their lives.

My experience tells me that whatever our trajectory, we are grossly ill-equipped to handle the dying process. Often I wonder how we — doctors, patients, the health-care system and society — can better prepare for death.

A few months ago, a patient of mine in his late 40s — a massively obese man, weighing more than 600 pounds — developed an overwhelming staph infection. As organ after organ failed and he became unresponsive, I convened a family meeting. In a conference room adjacent to the ICU, the nurse, my patient’s close relatives and I sat on comfortable chairs with a lone box of tissues on a side table.

I always choose my words carefully during such conversations. When I see little hope of meaningful recovery, I focus on quality, not quantity, of life. I talk about the pain and suffering that can be caused by chest compressions if the heart stops or by a tube pushed down the throat if the lungs can’t pull a breath.

My tone and demeanor are distinctly different if I feel there is hope. When the wife of a middle-aged man who had been in and out of the ICU for six months met with me, she was very discouraged, but there was hope, so I talked about “continuing with the struggle” and outlined a recovery plan.

In the case of my obese patient, we decided to back off his quadruple-strength blood-pressure medicine and not put him on the ventilator. Yet I could not convince the daughter of the 84-year-old nursing home patient to take her mother off the ventilator and make her comfortable. Ultimately, it is the patient and family who decide how the end of life will come.

In all these conversations, I felt I gave my patients and their families honest and timely advice. Yet I could have done a better job. These conversations could have occurred weeks and months in advance — at the time of admission to the hospital or the ICU, or even when the diagnosis of a terminal illness was made.

In most places in America, this rarely happens. One of the exceptions is a community in Wisconsin that has become known as one of the best places in America to die. For more than 25 years, Gundersen Lutheran Hospital system in Lacrosse has pioneered an end-of-life advance-care planning program, helping people make decisions on how they wish to die and encouraging the use of hospice and palliative-care services.

Palliative care is an interdisciplinary specialty focused on reducing symptoms and suffering, whether or not there is a cure. Hospice is a type of palliative care that is available when a doctor deems that a patient has less than six months to live.

The results from Gundersen have been impressive. Nearly 90 percent of the patients have an advance directive (twice the national average), and patients spend 13.5 days in the hospital in the last two years of their lives compared with more than 30 days in most major-city hospitals. To me, that’s a quality of life issue.

The effort has also had important financial implications: 30 percent lower spending for Medicare patients in LaCrosse compared with the national average ($18,359 vs. $25,860). Nationally, a quarter of Medicare costs occur in the last year of life, and 40 percent of those in the last month.

In a 2009 commentary in The Post, former House Speaker Newt Gingrich praised the Gundersen Lutheran Health Systems as a “best practice” model. Ironically, it was Gundersen that pushed for physicians to be compensated for end-of-life conversations with patients and families, an idea that, when included in the health-care overhaul bill in 2010, was pulled after former Alaska governor Sarah Palin and other opponents of the bill labeled these conversations “death panels.”

I am convinced that Gundersen Lutheran is taking the right approach. Patients want to talk about the end of life. In a 2005 AARP survey of Massachusetts seniors over age 50, nearly 90 percent said they wanted honest answers from doctors, but fewer than 20 percent had discussed their end-of-life wishes with their physicians.

A 2009 Archives of Internal Medicine study of patients with terminal metastatic lung cancer (the majority of whom lived only two months after they were interviewed) found that just slightly more than half had had their health-care provider discuss hospice care with them.

Many patients fear that if they choose palliative care, they will die sooner. But, in a landmark 2010 study in the New England Journal of Medicine involving patients with advanced lung cancer, those who had received early palliative care lived nearly three months longer (12 months vs. nine months) than those who had received standard cancer treatment. And the quality of life for those in the palliative-care program was significantly better.

Although 80 percent of large hospitals offer palliative care, according to a commentary accompanying the 2010 study, few doctors order such services for their patients.

Why are so many doctors reluctant to have end-of-life conversations?

Often we don’t want to be the bearer of bad news, especially when we have received little formal training in such communication. Also, we associate death with failure. Most difficult for me is that each conversation about death is emotionally draining. With each conversation, I find myself rehearsing for a conversation about my own death or that of a family member.

My grandfather, who raised me in India, is 93, and he often says without fear, “Son, I don’t have much time left.” He has sold the family house and lives with his sons, has given away his money and possesses a shelf of religious books and some clothes. Much as expectant parents prepare for a birth with an overnight bag, my grandfather is preparing for death.

In Eastern culture, death is accepted more than feared. Religious scriptures encourage a “holy death” when the body can no longer house the soul, by reducing food intake and spending time in prayer and meditation. In Western society, where youth and immortality have become the purpose of life, acceptance of death is much more difficult.

Manoj Jain is an infectious disease physician and contributor to the Washington Post, where this article originally appeared.  He can be reached at his self-titled site, Dr. Manoj Jain.

Submit a guest post and be heard on social media’s leading physician voice.