Do‐Not‐Resuscitate Orders and Hospitalization of Nursing Home Residents: Trumping, Neglect, or Shared Decision‐Making at the Eleventh Hour

What should a do-not-resuscitate (DNR) order have to do with the decision to hospitalize a nursing home resident? Absolutely nothing! A DNR order is about resuscitation—not whether to hospitalize a frail nursing home resident with lower respiratory infection. Yet reported in this issue of the Journal of the American Geriatrics Society are perplexing findings that persons with DNR orders are less likely to be hospitalized when they have a lower respiratory infection.1 None of the subjects had an order to forgo hospitalization. This nonexperimental study employed appropriate steps, including the use of propensity score, to control for potential confounding. Additionally, these findings are consistent with previous studies on DNR orders and hospitalization,2 resource utilization,3 and the use of feeding tubes.4,5 Do these findings represent physician trumping autonomy, benign neglect, or shared decision-making at the eleventh hour? Simply stated, we don't know which pertains, and we should. With the aging of the population and fiscal incentives to discharge patients from hospitals quicker and sicker, nursing homes are now becoming the final safety net for a frail, vulnerable population unable to be cared for in the community—many of whom are dying. This safety net is at risk, as evidenced by high staff turnover, nursing homes filing for bankruptcy, etc. A chief concern of family members of nursing home residents is that physicians are often unavailable. I doubt that the findings by Zweig et al.1 are the result of physicians unilaterally trumping preferences regarding hospitalization. Rather, subconscious benign neglect is the most likely explanation for these findings. I suspect that many healthcare providers are not recognizing the slow decline and not raising the difficult issue of what are the goals of care. DNR orders are now out and in the open, but what is really important and matters is the decision to hospitalize a frail nursing home resident that is still not openly discussed. In some regions of the country, the default decision is to hospitalize nearly all dying persons, whereas other regions have a lower rate of hospitalization. Which rate is right is based on whether there is process of decision-making that reflects patient preferences or a deliberative process about what is in the best interest for that person. It is quite possible that these results reflect shared decision-making that occurs at the eleventh hour. In this case, the nurse or physician speaks to the family when a nursing home resident is recognized as being seriously ill, usually at 1:00 a.m., and a decision is made not to hospitalize the resident. Although such decision-making is not optimal, it is ethically defensible. We should know the answer to whether the association between DNR orders and reduced hospitalization is ethically defensible. Public policy must urgently address how to provide competent, coordinated, and compassionate care for the burgeoning population living in nursing homes and slowly dying from chronic, progressive illnesses. In 1989, fewer than one in five persons died in a nursing home. By 1997, one in four chronic disease deaths were reported in nursing homes.6 In the future, 40% of Americans will die in a nursing home.7 These statistics are an underestimation of the role of nursing homes in the end-of-life care in that many persons are transferred from a nursing home to die in a hospital. As the baby boomers begin turning 60, public policy must address how best to insure and care for a burgeoning population of frail, older persons dying of chronic progressive illnesses. The key end-of-life decision for these persons will not be resuscitation but at what point in this disease trajectory one makes a transition in the goals of care. When do you decide that the risks of hospitalization outweigh the potential benefits? Nearly three decades after the creation of the DNR order, persistent problems exist. For the frail nursing home resident, I believe that we should stop asking the question, “If your heart stops, do you want us to attempt cardiopulmonary resuscitation?” For the majority of nursing home residents, the question is not applicable, given the poor outcomes of attempting resuscitation in nursing homes.8 Rather, we should focus our efforts not on eliciting preferences for resuscitation but on understanding the goals of care based on where that individual is in their disease trajectory of living with multiple chronic illnesses. The first step in this process of advance care planning is actively listening to nursing home residents and their families. How are their diseases affecting residents' quality of life? Have they reached a critical turning point—a stage in their life where they are weary with life? After active listening, the second step is for healthcare providers to state what they heard from the patient and family and then to provide recommendations to meet their goals (e.g., “Given what you have told me Mrs. Smith, our focus is going to be on your comfort—that will include my writing an order to forgo hospitalization unless you need it for your comfort. We are going to do everything possible to support you and keep you comfortable here.”) This is greatly oversimplifying a process that can occur over weeks to months, but I believe that this vision of advance care planning is much better than asking, “Mrs. Smith, if your heart stops, do you want us to restart it?” This is the wrong question and simply not applicable to the majority of nursing home residents.