CAN PLANNING AHEAD SAVE HEALTH CARE COSTS?1
Can Planning Ahead Save Health Care Costs at the End of Life?
Joseph R. Yancey
Army-Baylor MHA-MBA Program
Executive Summary
Health care costs are rising at an unsustainable rate in the United States. A high proportion of health care costs occur in the last year of a person’s life. About half of these costs are incurred from inpatient medical services. There is concern that such a high proportion of these costs occurring in the inpatient venue may not be consistent with the wishes of many American seniors, who overwhelmingly favor dying at home. Advance care planning (ACP) holds the potential to improve quality of care at the end of life with improvement in health care expenditures. Managers should focus their ACP implementation efforts on setting quality standards, educating patients and clinicians, and developing health policy support. ACP programs should be transparent and must be very mindful of potential conflicts of interest that could create ethical problems.
Can Planning Ahead Save Health Care Costs at the End of Life?
Rising health care costs in the United States represent an ongoing threat to its economic security. The Congressional Budget Office (CBO) calculates that the U.S. Government spends 5.5% of its gross domestic product (GDP) on government health care programs; by 2046, this number will grow to 8.9% and will be the largest driver of a doubling of our national debt as a percentage of GDP (Congressional Budget Office [CBO], 2016). This would make the national debt the largest in U.S. history. The fractured nature of the American system features inefficiencies that make health care spending much greater (Shi & Singh, 2015). The U.S. has enjoyed somewhat slowed growth in health care costs since the passage of the Affordable Care Act, but it remains highly uncertain whether this is due to the law’s passage or a slow recovery from the Great Recession (Cox, Dunn, Rittmueller, & Whitmire, 2016).
Another driver for increasing American health care costs is a rapidly aging population. The “Baby Boomer” generation began to reach age 65 in 2011 and numbered approximately 77 million people. By the time they all reach age 65 in 2030, 60 million of them are expected to remain, with all of them eligible for Medicare (Colby & Ortman, 2014). The Medicare Trust Fund is currently projected to become insolvent in 2030(KHN Morning Briefing, 2014). The elderly are more likely than average to utilize health care services and less likely to have taxable incomes (Shi Singh, 2015). Any plan to address American health care costs will have to figure out how to mitigate those costs in this growing population.
A recent study suggests that the highest amount of health care spending for an elderly American occurs in the final year of life (Cubanski, Neuman, Griffin, & Damico, 2016). It is a common conception that most Americans believe their deaths will be a sudden and peaceful transition. In reality, however, this is rarely the case. For seven out of ten Americans, death happens at the end of a long, drawn-out battle with a chronic illness (Shi & Singh, 2015). For these types of issues, evidence-based guidelines are often varied and numerous factors can influence the medical decisions being made. Thus, choosing how aggressively to treat different points of illnesscan be scientifically challenging for clinicians and personally challenging for patients.
There are several factors thatcan exacerbate the issues of providing care during the end of life. One is simply the natural, yet uncomfortable uncertainty surrounding death. In addition, patients nearing death are frequently unable to speak for themselves, relying instead on family members and clinicians to decide a best course. The fear is that this frequently leads to aggressive medical treatment as a default, despite that the utility of such treatment for some elderly patients may be approaching futility (IOM, 2015). As an example, while 90% of elderly Americans state that they would prefer to die at home, Medicare data says that this occurs only about one-third of the time (Henry J. Kaiser Family Foundation, 2015). Therefore, this type of aggressive inpatient medical care is not only expensive, but it may not represent the patient’s best interests or wishes late in life.
Health care managers must understand the issues with end-of-life care in order to shape their organizations to be more patient-centered and sustainable. Accordingly, this paper will examine end-of-life care through the lens of rising American health care costs. It will focus on the elderly, as this population most commonly encounters end-of-life issues and is the best studied. The paper will then build further on the topic with an important overview of ethical and social considerations around the topic of death. Then a variety of interventions will be presented, including evidence of their effect on costs. Finally, the paper will suggest ways forward that may hold promise for improving patient care at the end of life while remaining ethical and yielding economic benefit for patients, clinicians, and the American health care system.
Health Care Costs at the End of Life
To better understand the depth of this problem, a review of current trends in healthcare expenditures for the elderly is necessary. Most of the available data examining health care costs at the end of life for the elderlycomes from the Medicare program. This data shows that health care expenditures increase dramatically for seniors in their last year of life. Approximately 25% of Medicare expenditures ever spent on an individual person is spent during their last year of life (Neuman, Cubanski, Huang, & Damico, 2015). In fact, the average amount of Medicare costs spent per person in the last year of life is $33,486 (see Figure 1).
This expenditure data also shows that Medicare spending varies by age. The total average Medicare spending tends to be highest for the very oldest, peaking at age 96 with an average of $16,145 per year (Neuman, et al., 2015). However, when a patient is in the last year of life, the expenditure pattern is different. In these patients, expenditures peak at an average of $42,933 at age 70, then decrease gradually and fall below the average of $33,486 after age 85 (Henry J. Kaiser Family Foundation, 2015). It is believed that those who are in the “oldest of the old” category are more likely to seek less aggressive medical treatment at the end of life (Henry J. Kaiser Family Foundation, 2015). This is likely due to the fact that they (patient and provider) have a better understanding that further medical interventions would be futile.
Figure 1. Medicare spending per beneficiary in final year of life (Henry J. Kaiser Family Foundation, 2015).
It is also important to understand that health care spending at the end of life can vary considerably by the type of care provided to the patient. For instance, about one-half of Medicare expenditures in the last year of life for elderly patients are spent on inpatient medical services (Henry J. Kaiser Family Foundation, 2015). In contrast, 10% of total expenditures are spent on hospice services, while another 10% are accounted for by skilled nursing care (Henry J. Kaiser Family Foundation, 2015). A studyby Kaplan (2011) helps provide an explanation for these differences in service utilization patterns at the end of life. His study indicates that care in Los Angeles, California, had a42%higher totalend-of-life costthan San Diego. The author suggests thedifference was attributed to relatively higher use of inpatient medical services in Los Angeles versus relatively higher use of hospice services in San Diego (Kaplan, 2011).While the findings may be hard to generalize, it does suggest that a high degree of variance in type of healthcare resources within even local areas can have a high impact on end-of-life costs.
It is important to note that while the difference in costs incurred at the end oflife are drastic, one should not infer that solving this problem would solve the problem of high costs in American health care. In a study examining the top 5% of individuals with the very highest health care expenditures in the U.S., only 11% of these people were in their final year of life (Aldridge & Kelley, 2015). Although it would likely help in a significant way, the problem of high health care expenditures in the U.S. is a pervasive one and cannot be solved by focusing on a single population or intervention.
End-of-life Interventions
In many ways, the problem of how to address care for the chronically ill at the end of life is a modern one. Increases in medical technology and in life expectancy worldwide have opened new doors for patients, but have also left us less than prepared for dealing with the frequently slow death that comes from chronic illness. The following interventions, none of which are mutually exclusive, are commonly underutilized and will be reviewed here.
The most common form of end-of-life intervention is an Advance Directive (AD). These are legal documents constructed by a person giving instructions to health care workers and family members on what to do in the event the author of the directives loses the capacity to make medical decisions. ADs became more common in the 1990s after the Patient Self-Determination Act was passed by Congress. This law required hospitals to inform patients about their basic rights, including the right to formulate ADs (Shi & Singh, 2015).
Advance directives can take different forms. For instance, Do Not Resuscitate (DNR) orders instruct medical personnel to forgo cardiopulmonary resuscitation (CPR) in the event of cardiac arrest. A living will, on the other hand,goes further to give more specific instructions on what to do in the event a patient loses decision-making capacity.For example, one could ask to avoid placement on a ventilator to maintain breathing. Finally, a durable power of attorney appoints a specific person to be the patient’s proxy for decisions in case the patient loses decision-making capacity (Shi Singh, 2015). Each is an important document in its own right.
While they are the most common end-of-life intervention, few individuals have actually documented their wishes for treatment preferences at end of life. In fact, about 40% of adults over age 65 do not have ADs (Henry J. Kaiser Family Foundation, 2015). Several factors are thought to play a role in the lack of widespread adoption, including communication difficulties with medical staff, a lack of trust, distrust of medical systems, lack of understanding of options (Henry J. Kaiser Family Foundation, 2015), difficulties in paperwork completion, and discord between family members (Glick et al., 1998).
It is also important to understand that, while advance directives are important tools in end-of-life care, many experts now believe that ADs by themselves are inadequate to maintain good quality of care for those at the end of life (IOM, 2015). Theories as to why vary, but include that they are not widely used, not easily found when needed, may not be valid or are too vague for certain situations, or that medical personnel are unwilling to honor them (IOM, 2015; Klinger, in der Schmitten, & Marckmann, 2016). In addition, there does not appear to be an association between the presence of ADs and lower health care costs (Teno, et al., 1997). Therefore, other innovative ways to organize for end-of-life care have emerged.
One particular way of organizing is with advance care planning, or ACP, which represents a much broader approach to addressing issues at the end of life. While ACP frequently includes construction of ADs, it is meant to be more comprehensive and continuous, incorporating ongoing conversations between patients, physicians, and families about goals of care at the end of life. It represents an attempt to go further than simple written orders to an enhanced mutual understanding between patients and clinical staff. While ADs vary in form across different states, formal ACP programs are even more different and are often local or regional in scope. Proponents of ACP believe it represents a higher quality of interaction with patients in terms of end-of-life issues, allowing greater time and communication to take place on an uncomfortable, complicated topic. Also, ACP increases the use of palliative care interventions at the end of life, decreases family burden for decision making, and helps to prepare patients and families emotionally for death (IOM, 2015).
Significant barriers have prevented wider adoption of ACP. While ADs can be prefabricated and easily reproduced, ACP demands more time and emotional energy of the physician and the patient (Klinger, et al., 2016). In addition, much of the extra time to complete ACP has not been reimbursed to the physician. Recognizing this, the Center of Medicare and Medicaid Services (CMS) started to reimburse physicians for ACP-type services starting in 2016 (Henry J. Kaiser Family Foundation, 2015). It remains to be seen how an improved financial incentive will affect wider implementation of ACP.
In a recent systematic review, Klinger and colleagues (2016) asserted that ACPsmay reduce health care costs. For example, of seven studies included in their review, six of them showed that it reduced costs. The range of reduced costs was very wide, from $1,041 to $64,827, with relative reductions ranging from 5% to 68%. Cost reduction did seem to be highest in the sickest populations. The authors noted some limitations (i.e. that there was great variability in the calculation of costs) and that their results should be interpreted with some caution. Of note, the studies that showed cost savings were also more successful in their quality measures (Klinger, et al., 2016). Regardless, this review suggests that high-quality ACP programs could improve both quality and cost at the end of life, although more studies are needed to determine what components provide the highest value to patients.
Another common intervention at the end of life is hospice care. Hospice care represents a subset of palliative care where patients are expected to live for six months or less (Shi Singh, 2015). The focus of hospice care is usually on making patients more comfortable as opposed to extending their lives. Medicare covers hospice services with almost no cost sharing from beneficiaries. Hospice has become much more popular in recent years, with 47% of Medicare patients who died in 2013 using the service versus 23% in 2000 (Henry J. Kaiser Family Foundation, 2015).
It is well accepted that hospice services cost far less than inpatient medical care for the elderly; palliative care interventions typically cost far less than more intensive interventions focused on cures (Neuman, et al., 2015). Still, hospice services have become substantially more expensive since 2000; there is concern that the proliferation of for-profit hospice companies may partially explain this phenomenon. Moreover, data suggests that hospice spending per capita rises with age, peaking at age 104, and only exceeds inpatient medical services spending for patients per capita once they pass the age of 100 (see Figure 2; Henry J. Kaiser Family Foundation, 2015). This suggests that this type of care is more generally accepted for the “oldest of the old” population rather than the standard of care for other age groups at end of life. Again, these utilization patterns could show an opportunity to reorganize to improve patient centered care and reduce healthcare expenditures.
Figure 2. Medicare spending by age (Henry J. Kaiser Family Foundation, 2015).
Ethical Considerations
Controlling costs is of utmost importance to modern U.S. health care. However, seeking to do so by addressing end-of-life costs presents the health care manager with potential ethical concerns. For example, it is not difficult to see how health care administrators and providers responsible for controlling costs may seek to steer patients to less expensive modalities regardless of the goals of the patient. At the same time, modalities such as ACP and hospice tend to be more patient-centered, improving quality of life and reducing costs. The ethical implications of cost control at the end of lifewill be examined through the lens of the principles of biomedical ethics that are most affected by such decisions: autonomy and justice.
Autonomy
A recent systematic review of ACP by Klinger and colleagues (2016)recognized the potential ethical implications of cost control at the end of life. The authors argue that ACP in its purest form should enhance patient autonomy by enhancing communication between patients and clinicians about goals of care and better informing the patient about what is possible and what is likely about the end of life. Still, expert power rests in the hands of physicians in end-of-life discussions and the potential to abuse that power is real.
In order to avoid any impropriety that would risk patient trust, the authorssuggest that ACP programs should focus on the following: 1) transparency of objectives, 2) adequate supervision and training of personnel involved, 3) clear standards, and 4) educating facilitators about potential conflicts of interest (Klinger, et al., 2016). In a separate examination of the ethics of end-of-life planning, Donley and Danis (2011) argued strongly that transparency about costs, especially those borne by the patient, is essential to patient autonomy. For example, they suggest that a patient could decide that money spent to prolong his life minimally may be better spent on an enjoyable vacation in the near term instead (Donley & Danis, 2011). Therefore, ACP programs must focus on remaining patient-centered and patient-controlled.