Imagine yourself the son or daughter of a grievously ill octogenarian. Your mom or dad suffered a catastrophic event and has been in an ICU, barely conscious, for two weeks. You understand organs are shutting down, but you struggle to understand the medical jargon or the purpose of tubes and machines attached to every orifice. You and your family have kept a steadfast vigil at the hospital, and today the doctors called a meeting to talk with the family.

The family is desperate to understand what is happening and learn what lies ahead. The doctor needs the family to authorize a treatment plan. Will you learn how close your loved one is to death? Will the doctor learn their end-of-life wishes? Are there options and difficult decisions to be made? The legal standard for these conversations is that the family consents to treatment after being fully informed of alternatives, risks and benefits of each. The medical standard is that critical treatment decisions are not unilateral, but shared by the patient or surrogate and the doctor.

Sadly, informed consent and shared decision-making, the twin pillars of patient-centered healthcare, aren’t the rock-solid structures we would hope for. They are subtle, fragile fragments, often half-buried and overlooked in a heap of technological rubble. It’s a doctor’s job to dig them up and erect them, and many are not so inclined. That’s the lesson of a new study in the Journal of Intensive Care Medicine.

It is well established that many dying patients endure excesses of torturous treatment in the intensive care units of America. Tubes, machines and other contrivances can take over in clear violation of previously expressed wishes. Advance directives declining such treatments routinely get ignored. Three years ago I blogged about a bloated and misguided medical-industrial complex that tortures patients, and called it a national tragedy. I’m sorry to report not much has changed since then.

At least we know more about the problem now.

Evidence of patterns and their causes is emerging. Investigators in last week’s study listened in as 54 doctors in five San Francisco ICUs talked with families of 72 patients facing likely and imminent death. These are highly emotional and difficult conversations, and the investigators wondered whether doctors routinely presented the option of substituting invasive techniques with comfort care and allowing a natural death. As the authors noted, decisions about life-sustaining and invasive therapies are not “right or wrong” decisions. They are “preference-sensitive, meaning the best choice depends on individual values …”

Here’s what they found, and it’s not good.

The first piece of bad news is that families learn about comfort care only around half the time. These patients were very sick. They had already been in the ICU an average of two weeks. They suffered from a variety of life-threatening and end-stage illnesses including metastatic cancer and multi-organ failure. Most (72%) died in the hospital. The investigators wrote frankly of their surprise at the finding that doctors talked about comfort care so infrequently.

The second piece of bad news is that medical science does not guide them. If doctors used science to decide whether to talk about the risks and benefits of comfort care, the researchers would have found a correlation with objective variables. Comfort-care conversations would vary depending on factors like the age of the patient, whether they had heart failure or metastatic cancer, or how long they had been in the ICU. ICU doctors have clinical scoring methods to predict the likelihood a patient will die during the hospitalization. These ICUs employed the APACHE II score, and in this study the families of patients with the worst scores, the sickest patients closest to death, were no more likely to hear about comfort care. This makes no scientific sense.

The final piece of bad news explains why. Only one variable predicted whether a doctor would inform families about comfort care. That was how strongly the physician, on a scale of 1 – 10, personally believed life support should be removed. If they did not personally believe life support should be removed, they did not offer the comfort-care alternative. If they personally believed life support should be removed, they informed the family of the option of comfort care. So the ability of families to honor their loved ones’ wishes and navigate a peaceful death depends entirely on whether the doctor believes in a goal of comfort at the end of life. Doctors who believe the only legitimate therapeutic goal is to prolong life don’t disclose the comfort-care option.

This is not how it should be. Doctors have a duty to reveal all available treatment options, and the risks and benefits of each. They are not supposed to withhold information about treatments they don’t personally support. 72% of the patients in this study died in the hospital. Many died enduring the torture of intensive care to the very end, their families never learning it would be OK to opt for comfort care, hold their loved one as they died, and spare them final agonies and indignities.

This study ran from January 2006 through August 2008. On January 1, 2009, Compassion & Choices’ Right to Know End-of-Life Options Act went into effect in California. The law instructs physicians to reveal all end-of-life options, including comfort care, when a terminally ill patient or surrogate asks for this information. Doctors’ groups opposed us when we worked for this law in the legislature. They said doctors already reveal all options because the standard of care for informed consent and shared decision-making requires it. The evidence is in, and our law was clearly needed. Patients shouldn’t have to ask, either. Doctors who omit information about peaceful dying from an informed-consent conference should be accountable, and disciplined.