Death and Primary Care

Death has always been a part of primary care. For most of the history of medicine, death followed an acute traumatic event or an infectious illness by days or weeks or perhaps months. In the past half-century, the face and pace of death began to change. Death now fades in slowly—over years or even decades. For some of our patients, the process moves through shades of gray without a clear end of the beginning or a beginning of the end. In fact, the pace of death has slowed so suddenly that we seem to have lost our ability to recognize it. Through a combination of technology, medicine, and culture, we and our patients are often lost in a foggy boundary between life and death. Primary care now finds itself in a position unparalleled in human history: we provide care for a large number of older adults with chronic conditions who either don't, won't, or are not allowed to die in a precipitous manner that could be recognized by all as the end. In attempting to identify dying (or the dying trajectory) for our patients, we find ourselves poorly trained, our systems poorly designed, and our patients and communities poorly equipped.

Four articles in this month's Journal of General Internal Medicine seek to teach our teachers how to help their trainees, patients, and communities to recognize the end.^1–4 Each of these four articles tackles a slightly different aspect of this complex problem and through these articles we can begin to outline the enormity of the task facing primary care. First, Carey et al. provide a clinically useful yardstick to recognize the end of the beginning of death even if we can't quite identify that one point when an older adult first begins to move faster toward death.¹ Older adults with functional impairment have already started on an accelerated glide path toward death. Regardless of the etiology of this functional decline, it does indeed mark the end of the beginning of death. In addition to demonstrating the tremendous heterogeneity in risk of mortality among community-dwelling older adults, these authors also suggest that the index can be used in clinical decision making, research applications, and public policy. This is an important tool in helping primary care physicians communicate with patients and policymakers because it quantifies complex data in an understandable metric. Such data, however, must always be coupled with individual and cultural values regarding the goals of care. We can help patients understand that more accurate prognostication may affect their decisions, but these are shared decisions and decisions influenced by myriad other factors. Better prognostication alone is only part of the process in changing medical practice.

Indeed, Lewis et al. show how difficult it is to change behavior through knowledge alone.² After at least a decade of observational data and expert opinion on the limited role of feeding tubes in the care of older terminally ill patients, these authors find no evidence of declining use. Sadly, the rate of artificial feeding may even be increasing among patients where there is no benefit. Lewis et al. correctly point to the important role of society and culture, in addition to medical knowledge, in determining the content of medical practice. They also correctly point to the apparent frequent discrepancies between providers and families regarding the goals of care. The role of culture, including economics, is pervasive and must be embraced as an important lever in changing behavior. This requires primary care physicians to leave the confines of their offices and play a larger role in their communities. Leadership in defining community standards and the goals of medical care is an appropriate and important role for primary care physicians.

Of course, community action must be coupled with the credibility that comes from excellent technical and humanistic care at the level of individual patients. As Sachs et al. recognize, most older adults will receive their care from generalist physicians rather than geriatric specialists.³ Unless some dramatic change in our health care system occurs in the next few years, this will be true even when the older adult approaches the end due to a geriatric syndrome. Sachs et al. provide a sobering account of the challenges facing primary care physicians in caring for patients with dementia at the beginning of the end. The barriers to excellent care span the spectrum of cultural values, health system design and incentives, caregiver roles, poor application of existing knowledge, and limitations in our treatment options. Alzheimer's disease and related dementias reflect the archetype for the challenges facing generalist physicians in the care of older adults—all of the barriers and problems seem to converge in this particular illness. Sachs et al. wisely avoid the simplistic admonition that primary care simply has to do more. The magnitude of this problem requires fundamental action at the level of health systems, economics, and public policy as well as at the level of individual providers.

So how are we to help the trainee or patient or family member on today's schedule? Lest we be shaken by the complexity of these issues, Kasman provides us with a practical, hands-on approach to shared decision making at the end.⁴ If we take nothing else from the uncertain footing for providers and dying patients in the 21st century, we can find solid ground in Kasman's reminder that caring for the patient is never futile. To care for a patient, we must embrace our role in leading negotiations about the goals of care. Primary care's leadership in these negotiations is important at the role of individual patients and at the level of communities. The need for primary care leadership in this area is yet another reason why we can focus our creative energy on death and primary care rather than the death of primary care.