Currently, Medicare does not offer a paid benefit for advance care planning (ACP). As a result, health care providers who want to assist Medicare enrollees with ACP do so voluntarily and neither they, nor their institutions, are compensated for their time and efforts. This is not only an unfair expectation on individual practitioners or health institutions, it is also medically and ethically unsound. Fortunately, two recent events have the potential to reshape the landscape of advance care planning in the U.S.

Cultural And Policy Evolution In Advance Care Planning

On September 17, 2014, the Institute of Medicine (IOM) published Dying in America: Improving Quality and Honoring Individual Preferences Near the End of Life. The report is built on two basic premises:

  • Despite humankind’s medical advances to lengthen life, death remains the inevitable outcome for all humans, and thus all people have a stake in improving end-of-life care; and
  • Medical advances complicate the dying processes for many people in ways that medical training and health systems in America generally were not designed to address, making them ill equipped to balance complex patient needs in practice.

The IOM report finds that advance care planning is essential to improving the quality of end-of-life care. It also notes how traditional approaches to ACP—namely, having people complete advance directives on their own, such as living wills (LWs) or powers of attorney for health care (POAHCs)—have ultimately failed. Only about 25 percent of the general U.S. populace has documented its ACP preferences; even among people 60 and older, that rate only hovers around 50 percent.

According to the IOM report, there are multiple reasons why these completion rates remain relatively low, including people’s reluctance to converse with loved ones about death and dying and health systems’ failure to standardize their internal ACP processes. The report also points to the typically poor quality of ACP conversations between patients and their loved ones or doctors in which vague or value-laden claims—such as “I don’t want to die hooked up to machines”—are unexplored and taken at face value.

The IOM identifies a number of characteristics shared by most effective ACP approaches, such as understanding and treating advance care planning as a recurring discussion about someone’s goals, values, and treatment preferences instead of seeing it as a one-time activity including in ACP conversations those designated as power of attorney for health care; and accommodating diverse belief systems that can shape the treatment decisions people make at the end of life.

The second major event is the Centers for Medicare and Medicaid Services’ (CMS) apparent willingness to begin considering reimbursement for advance care planning discussions between patients and providers. On October 31, 2014, CMS issued a final rule regarding revisions to its payment policies. This document contains two new practice billing codes, 99487 and 99497, which will officially recognize the efforts of physicians or other qualified health care professionals—those with education, training, and licensure—to engage in ACP with patients, including the explanation, discussion, and completion of standard advance directive forms.

There is still no actual reimbursement associated with these two codes (similar to when CMS initiated practice codes for palliative care). It appears that the usage of these two codes will help CMS determine whether or not to eventually reimburse for ACP — and for how much. It is not uncommon for CMS to use this kind of data when there is some uncertainty about how often a service like ACP is, or will be, used.

Respecting Choices

Both the IOM report and Medicare’s new billing codes mark an evolutionary turning point for advance care planning in the U.S. Not only do these two events signify that ACP should be treated as a part of standard medical care, but the IOM report indicates what should be paid for under the new CMS practice codes. In particular, the IOM report states that the hallmark of good ACP is open, clear, respectful communication between the person doing the planning and his or her clinicians, loved ones, and health care agents. This implies that if CMS pays for ACP, the reimbursement should be based on ACP models that focus on this type of communication.

Advance care planning tools such as Physician Orders for Life Sustaining Treatment and living wills, can effectively communicate people’s preferences when they can no longer make decisions for themselves due to long-term or acute mental declines. However, as the IOM describes, these tools are only as effective as the system in which they are implemented and used. One initiative highlighted by the IOM report for its successful incorporation of advance care planning into a broader health care system, is the La Crosse, Wisconsin-based Respecting Choices (RC).

RC was developed in the early 1990s as a community-wide collaboration between the four major health organizations in La Crosse: Gundersen Clinic, Lutheran Hospital, Skemp Clinic, and Franciscan Health System. While Respecting Choices is now exclusively owned and operated by Gundersen Health System, it continues to be used in all health care facilities in La Crosse, and the entire Gundersen system, which includes six hospitals and 27 clinics in 19 counties across the Wisconsin-Iowa-Minnesota tri-state area — a patient population of approximately 560,000. At Gundersen, operating costs for advance care planning are assumed by the organization. ACP conversations between clinicians and patients are not reimbursed by any traditional payers and ACP facilitators—those who are trained to assist persons with the advance care planning process—volunteer their time.

While physicians play an integral role in advance care planning conversations, it is simply impractical to expect them to have the time, training, or comfort level to discuss in depth a person’s values, beliefs, and goals of care, address how that person’s medical information fits with that worldview, and then ensure that person’s medical choices are documented and communicated. Recognizing this limitation, the RC model relies on trained ACP facilitators to:

  • Uncover gaps in their understanding about their individual medical situations then working with medical providers to help bridge those gaps.
  • Engage patients about their values, beliefs, preferences, and goals.
  • Help people communicate with their chosen health care agents and other loved ones about those values and goals of care.
  • Help patients document their elected health care agents and goals of care using a tools such as a living will or powers of attorney for health care.
  • Ensure that documented plans are easily retrievable by those who may need access including health care agents, loved ones, and medical providers.

Facilitators are trained then certified by Respecting Choices to conduct ACP in three distinct stages: when adults are relatively healthy, when they are beginning to suffer the effects of a chronic or life-limiting illness, and when they are near the end-of-life. RC facilitators who work with relatively healthy adults include nurses, social workers, physicians, and community volunteers without medical training. Facilitators who conduct ACP with ill people usually have some knowledge, training, and experience working with adult patients in this population. RC facilitators are assigned to work with populations that fit their level of training. Facilitators will only engage with people who say they want to create an advance care plan.

One reason the Respecting Choices model was explicitly discussed in the IOM report is because outcomes data is available to measure the program’s impact. In 1995 and 1996, two years after the community-wide implementation of the RC model, Drs. Bernard Hammes and Brenda Rooney conducted The La Crosse Advance Directive Study (LADS 1), a retrospective study of 540 decedents in La Crosse. This study found that 85 percent of decedents had an advance directive and that 95 percent of these directives were documented in their medical records.

When this study was repeated in 2007 and 2008 (LADS II), the numbers of decedents with advance directives that were also documented in their medical records rose, respectively, to 90 percent and 99 percent. What is perhaps most impressive, and most important, about the high percentage of people in La Crosse with advance care plans are the rates at which these plans are followed by health care providers. As noted in the LADS II study, eight patients were found to receive medical treatment inconsistent with their documented preferences in LADS I. LADS II found no cases of patients receiving treatment inconsistent with their documented treatment preferences.

Another measure of the RC model’s success is its apparent economic impact on utilization and health care expenditures in the last two years of life. According to the Dartmouth Atlas for Health Care, Gundersen’s 2010 utilization score was 0.40. This score—which is a ratio of the number of days patients spent in the hospital and the number of physician encounters they experienced as inpatients during the last two years of a person’s life compared to a baseline score of 1—is lower than institutions in the National average (1.0), and even institutions in the 10th percentile (0.62). Gundersen’s average Medicare expenditures per person over the last two years of life in 2010 were $48,771. This too sits well below institutions in the 90th percentile ($102,939), the National average ($79,337), and even the 10th percentile ($58,866). This impact is attributable to reducing wasteful spending related to providing unwanted care to patients at the end of life.

Respecting Choices has been implemented in many health systems and organizations throughout the United States—including Kaiser Permanente of Northern California—and in Canada, Australia, Singapore, and Germany. Well known statewide ACP programs such as Honoring Choices Wisconsin, Honoring Choices Minnesota, Honoring Choices Virginia, and Honoring Choices Florida are also based on the RC model. Randomized controlled trials of implementations nationally and internationally by ACP researchers such as Detering, Lyon, Kirchhoff, and in der Schmitten further demonstrate the success of the RC model and the ACP facilitation process.

CMS’s willingness to consider reimbursing for advance care planning sends a strong message to medical providers that ACP should now be considered a standard of care. The IOM’s recognition that traditional modes of advance care planning have failed to effectively guide treatment decisions for incapacitated patients—and the report’s focus on innovative programs like Respecting Choices—sends a strong message to payers, including CMS, that advance care planning should be covered when it occurs as part of models that have demonstrated effectiveness in both completion and compliance of care plans.