WEDNESDAY, Sept. 28 (HealthDay News) -- New research suggests that sick people who talk about advance directives don't die earlier than their counterparts who don't have those discussions about end-of-life care.

The study only looked at patients who were mildly or moderately ill, not those who were expected to die soon. And the study doesn't definitively prove that the discussions don't affect death rates.

Still, the findings suggest that "having a discussion or completing a directive doesn't lead to harm. We can promote autonomy, and not be worried that we're harming patients," said study author Dr. Stacy M. Fischer, an assistant professor at the University of Colorado School of Medicine.

Fischer and her colleagues launched their study in the wake of the 2009 health care debate, after a provision that allowed funding for advance directive discussions under the Medicare program came under scrutiny, according to background information in the study. Although this provision had been in place during the George W. Bush administration, the Obama administration removed the provision from the Affordable Care Act so that the issue could not cloud the larger debate about affordable health care, the study said. The controversy also introduced the term "death panels" to the public. Media reports say no such panels -- alleged to allow doctors to make decisions about who lives or dies -- actually existed in President Obama's health care reform proposal.

Ordinary advance directive discussions can be difficult. Dr. Jennifer Mack, an assistant professor at Harvard Medical School who's familiar with the findings, said patients and families sometimes don't feel positive about advance directive discussions. "They feel that the physicians who bring these issues up are not necessarily acting in their best interest, but have a particular agenda," she said.

Mack added that some people may be concerned about a "self-fulfilling prophecy" -- the idea that "if we say a patient has two weeks to live, the patient will believe that."

Fischer said she wanted to address a concern that the discussion of "death panels" implies that advance directives will "somehow lead families or health care providers into stopping providing care."

The study authors examined the medical records of 458 patients at Colorado hospitals who were asked at their bedside whether they'd taken part in discussions about advance directives. The researchers focused on 356 people who were thought to be at low (297 people) or medium (59 people) risk of dying over the next year; the study left out those who were expected to die imminently because that fact would be "driving those discussions," Fischer said.

The researchers found that 45 percent of the 356 patients said they'd had discussions about advance directives, although just 10 percent had advance directive documents included with their medical charts. Six percent had living wills, 9 percent had documents that gave someone else a durable power of attorney, and 7 percent had other types of similar documents.

About one-quarter of the patients died during the 2003-2009 follow-up period of the study. The researchers found that having a discussion about advance directives didn't boost or lower a patient's risk of death.

The researchers wrote that their study offers "limited data to counteract the misleading claims of those opposed to the advance care planning process."

Mack said the study appears to be sound, although the number of subjects is relatively small. "Trying to provide data to address some of these assumptions is probably a good thing," she said.

The study was published online Sept. 28 in the Journal of Hospital Medicine.

More information

For more about advance directives, try the U.S. National Library of Medicine.