Show Summary Details
Page of

Pain and Symptom Management at the End of Life 

Pain and Symptom Management at the End of Life
Pain and Symptom Management at the End of Life

Robert C. Macauley

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 23 October 2019

Death is an inevitable aspect of the human condition. Dying badly is not.

(Jennings, Ryndse, D’Onofrio, & Baily, 2003)

Historicolegal Background

For most of the history of Western medicine, there was little attention paid to the treatment of symptoms at the end of life. There were very few ways of keeping patients alive, and thus death tended to be more precipitous, especially compared to the often drawn-out dying processes of today (p. 13). While attention to the spiritual aspects of death was quite sophisticated—and, in many cases, strictly prescribed—management of physical symptoms lagged far behind.

This is not because no treatments were available, as the pharmacologic effects of the poppy plant have been known since Neolithic times (Booth, 1998). It is unclear, however, whether Hippocrates appreciated its analgesic—as opposed to the sedative and hypnotic—effects (Prioreschi, Heaney, & Brehm, 1998). These certainly came to be recognized by the time of the Roman Empire (Tallmadge, 1946) but subsequently fell out of practice after the Empire’s fall. While Muslim physicians explored the analgesic use of opioids from the ninth century onward (Al-Mazroa & Abdel Halim, 1989), in the West they were only rediscovered at the beginning of the sixteenth century by Paracelsus, who stated that “among medicines offered by Almighty God to relieve human suffering none is so universal and effective as opium” (Baraka, 2000).

Sydenham’s laudanum was among the most common opium-containing compounds in Great Britain and the Americas in the ensuing centuries (Hamilton & Baskett, 2000), with the nineteenth century seeing deeper understanding of the mechanism of its analgesic effects (Haller, 1989) as well as the first formulation of morphine by Sertürner (Macht, Herman, & Levy, 1915). Parenteral administration followed soon thereafter (Howard-Jones, 1947), and morphine was used to treat battlefield injuries both in the American Civil War as well as other conflicts (Howard-Jones, 1947).

The risks of opioid abuse soon became apparent, however, with opium overdose representing a frequent cause of death in the West (Booth, 1998), and the Chinese opium wars occurring in the East. For this reason, many felt that opioids were too addictive to be of widespread utility, especially in end-of-life (EOL) care.

This view remained prevalent in the West through the end of World War II when a new approach formed, led by John Bonica (1953) based on his experience treating soldiers in pain and his own personal experience with chronic pain as a professional wrestler. The profound impact of pain on one’s quality of life was increasingly recognized, as Albert Schweitzer (1948) famously put it: “Pain is a more terrible lord of mankind than even death himself.”

Recognizing the obligation to help patients who “scream in pain and suffer in silence” (Reich 1989), the role of opioids came to be reconsidered, especially in EOL care. Cicely Saunders (1963) took the lead in this area, and when her groundbreaking paper on the subject was published, everything changed. There was now empiric evidence that intense pain and dyspnea at the end of life could be treated effectively and safely, and upward titration proved beneficial in managing refractory pain.

However, there remained a concern that opioids might not only provide comfort as death approached but could also hasten the timing of death. In cases of extreme pain requiring administration of high doses of opioids, the well-known side effect of respiratory depression could cause a patient to stop breathing. This is especially true when benzodiazepines—another staple of EOL care, used to treat anxiety—are simultaneously administered, given their synergistic depressive effect on respiratory drive. As the AMA’s Council on Ethical and Judicial Affairs (1992) notes, “the level of analgesia necessary to relieve the patient’s pain . . . may also have the effect of shortening the patient’s life.”

In order to avoid any possibility of hastening death, some physicians were extremely cautious in treating patients with opioids, either choosing a very low dose, giving it at unnecessarily long intervals, or titrating up too slowly. Such opiophobia—referring to an irrational fear of treating with opioids (Bennett & Carr, 2002)—can be prompted not only by clinical concerns for respiratory depression but also by medicolegal concerns if the treatment were felt to represent euthanasia. Some commentators took these concerns so far as to recommend against intensive pain treatment: “Just as a physician must learn that lack of a cure does not equate with failure, a nurse must learn that the presence of pain [in the dying patient] does not mean failure” (Pohlman, 1990).

The impact of opiophobia on patient care can hardly be overstated. The SUPPORT study—in addition to noting the degree to which patients’ wishes were either unrecognized or, if recognized, were not heeded—clearly documented the amount of needless suffering that was occurring for lack of adequate analgesia (p. 10). Half of the conscious patients in the study experienced moderate to severe pain in the last three days of their lives (Desbiens & Wu, 2000), largely for wont of effective opioid management.

This led to increased attention to pain, including the classification of pain as the “fifth vital sign” (Davis & Walsh, 2004). Physicians and nurses were expected to inquire about the patient’s pain level, document the answer, and respond appropriately. The World Health Organization (2017) devised an analgesic ladder, with the second and third steps composed primarily of opioid medications. The American Board of Anesthesiology recognized “pain medicine” as a certificate of added qualifications in 1991, followed a decade later by subspecialty certification by the American Boards of Psychiatry and Neurology and of Physical Medicine and Rehabilitation (Cope, 2010). During this period a wide variety of new opioids—administered through different routes and with varying durations of action—were developed.

Recently, however, the pendulum has begun to swing back in the other direction, but not because of the use of opioids in EOL care. Rather, the concern relates to the opioid “epidemic” in the United States. Despite representing only 5% of the world’s population, the United States consumes 80% of the world’s opioids (Express Scripts, 2014). To some degree, this has been attributed to the emphasis on lowering the patient’s self-reported pain rating, rather than focusing on functional measures and admitting that not all pain can be reduced to “zero,” at least by opioids alone (Younger, McCue, & Mackey, 2009).

A variety of steps have been taken to address the opioid epidemic. Drug manufacturers can be required to implement a risk evaluation and mitigation strategy to ensure that the benefits of a medication outweigh its risks (Food and Drug Administration [FDA], 2017). Prescription drug monitoring programs help ensure that prescriptions are not falsified and also identify patients who may be receiving opioids from multiple providers or pharmacies (Islam & McRae, 2014). “Daily milligram limits” on opioid use—generally approximately 90 milligrams of oral morphine equivalents—have been established, with greater scrutiny of higher dosing (Dowell, Haegerich, & Chou, 2016).

These are all prudent steps in response to a serious health crisis, but they can also have unintended negative consequences on the practice of palliative care. A blanket indictment of opioid prescribing falsely equates the use of opioids for chronic pain—where they may not be indicated and for which they carry significant long-term side effects (Dowell et al., 2016)—and for EOL care. To a large degree, the opioid epidemic was fueled by the failure to make this very distinction, extrapolating from the well-documented benefits of opioids in EOL care to their widespread and injudicious use in other contexts. The response to the epidemic threatens to make the same error only in the reverse direction, projecting the well-documented risks of long-term opioid use and frequent prescribing in chronic, nonmalignant contexts onto EOL care. For instance, a daily milligram limit on opioid use might make sense for chronic pain—and helpfully redirect the attention to non-opioid (and especially nonpharmacologic) pain management—but such a limit can represent a barrier to appropriate upward titration of opioids at the end of life.

The stakes in this debate could not be higher, ranging from addiction and even death in the context of opioid overuse, to unnecessary and extreme suffering at the end of life for underuse. It is crucial, therefore, to develop strategies which effectively address the former while not unintentionally turning back the clock to the days when pain at the end of life was inadequately treated.


Palliative care is often associated with pain management, and this is certainly a common reason for palliative care consultation. However, there are many physical symptoms beyond pain that require treatment in patients receiving palliative care, including fatigue, weakness, loss of appetite, and loss of energy (Teunissen et al. 2007). Sometimes treatments for either the underlying disease (e.g., chemotherapy) or for the physical symptoms related to the disease (e.g., opioids for pain) can cause additional or worsened symptoms, such as nausea or pruritis. The reason for focusing on management of pain here is the potentially lethal side effect of respiratory depression due to opioids.

Pain is multifaceted, not limited to physical suffering. Patients may also experience other types of pain, including social (isolation), spiritual (wondering why this has befallen them), and psychological (demoralization and depression). These phenomena are not mutually exclusive (Cherny, Coyle, & Foley, 1994): intense physical pain can lead to isolation and metaphysical questioning, and by the same token depressed mood can heighten the perception of physical pain (Saunders, 1984). Saunders (1964) thus coined the term “total pain,” which is an accurate description of what many patients with life-threatening illness may endure.

This underscores the significance of recognizing the type of pain(s) that one is treating, because opioid analgesics are effective in some cases but not others. Spiritual pain calls for spiritual care and emotional pain for psychological support (and perhaps antidepressant or anti-anxiety treatment). Some have gone so far as to argue that certain aspects of suffering fall outside the physician’s purview altogether. For instance, Ahmedzai (1997) writes, “Ultimately, suffering from losses, lack of love, existential doubts as well as from poverty and cruelty are not medical issues, and the response to them is not necessarily the responsibility of any healthcare discipline.” A similar point is made by critics of using palliative sedation to treat existential distress (p. 265).

This seems, though, to overlook the transdisciplinary and comprehensive nature of palliative care in addressing suffering in all its forms. To be sure, opioids (or any pharmacologic modality) are not the “cure” for pain of every stripe, but the palliative care team’s role extends far beyond the prescription pad. To say that certain forms of individual suffering—acknowledging that societal ills such as poverty are broader questions—are not the “responsibility of any health care discipline” overlooks the uniqueness of palliative care, which intentionally takes up that broader mantle. To be sure, not all suffering can be “solved,” and those forms that can be may well lie outside the sphere of the palliative care team. In its most fundamental form, “Suffering is not a question that demands an answer; it is not a problem that demands a solution; it is a mystery which demands a presence” (Wyatt, 2009). But that does not mean that suffering of every form should not be taken seriously and attempts made to address (if not eradicate) it.

In terms of physical suffering, opioids are commonly used not only for pain but also for dyspnea, and initially in small doses. These are usually increased over time, both in response to disease progression but also because of the development of tolerance. This refers to the physiologic need for higher doses to achieve the same effect, as the body adapts to the current dose of opioid medication. The need for increasing doses, therefore, does not necessarily reflect either disease progression or “addiction” but may simply be the body becoming accustomed to the current dose and requiring an increase to achieve the same effect.

Ongoing use of opioids also leads to physiologic dependence, defined as having a negative reaction if the medication were to be discontinued (i.e., withdrawal). This is distinct from addiction—what some have described as “psychological dependence”—which “is characterized by inability to consistently abstain, impairment in behavioral control, craving, diminished recognition of significant problems with one’s behaviors and interpersonal relationships, and a dysfunctional emotional response” (American Society of Addiction Medicine, 2011).

The distinction between physical and psychological dependence is crucial, because sometimes undertreated pain can prompt behavior which resembles that caused by addiction. Such pseudoaddiction is characterized by reports of continued pain despite currently receiving treatment, emphatic demands for higher doses of opioids, and querying multiple physicians (so-called “doctor shopping”) when initial responses are insufficient to control the pain (Weissman & Haddox, 1989). In such cases, a physician must determine whether the patient’s pain is, indeed, being undertreated, for the proper response to pseudoaddiction is an optimized analgesic regimen. The proper response to true addiction is, of course, extremely different.

Another unique aspect of opioids is that there is no “correct dose.” For most medications, there is a standard dose, often determined by the patient’s weight or body surface area, with modifications made for impaired organ function. Sometimes there are multiple tiers of dosing, as in the case of high-dose steroids. But for the most part, if a medication does not have the desired effect at the recommended dose, it is considered a treatment failure, and another medication (or form of treatment) is tried.

Opioids for pain or dyspnea are an exception to this rule, for the “correct dose” is the one that controls the patient’s symptoms. While the basic rule is to “start slow and go slow” in titrating upward—given side effects such as sedation and respiratory depression—rapid titration is often necessary to relieve severe symptoms. The doses that are reached can be stunningly high, especially if the patient has developed tolerance over time. Standard teaching is to increase the dose by as much as 50% for moderate unrelieved pain and 75% or more for severe unrelieved pain (Schneider, Yale, & Larson, 2003). This highlights the problem with a daily milligram limit, because judicious upward titration over time—especially in the face of advancing disease—can exceed that by an order of magnitude or more.

While most of this chapter focuses on pain, other symptoms present similar clinical and ethical challenges. In particular, management of dyspnea can be quite challenging. Many of the principles identified in relation to pain management also apply to dyspnea, for opioids are a mainstay of the treatment for both. Appropriate upward dose titration to manage symptoms—in light of the risk of respiratory depression—as well as the challenge of distinguishing suffering from either pure dyspnea or pain require clinical expertise as well as ethical nuance and insight.


While simultaneously acknowledging the physiologic efficacy of opioids in treating pain and dyspnea, significant concerns have been expressed regarding potentially hastened death from the side effect of respiratory depression. Some have gone so far as to refer to high-dose opioid administration at the end of life as “indirect euthanasia” (Huang & Emanuel, 1995; Smith, Orlowski, Radey, & Scofield, 1992) or “accidental euthanasia” (Lundberg, 1988).

So what is the medical team to do, when clinical guidelines recommend upward titration of opioids which could compromise the patient’s respiratory drive and hasten her death? One potential response, of course, is to risk undertreating pain, as the SUPPORT study revealed (p. 10) (SUPPORT Principal Investigators, 1995). Opiophobia is common: nearly half of physicians and nurses in another study acknowledged the prevalence of opiophobia by agreeing that “clinicians give inadequate pain medication most often out of fear of hastening a patient’s death” (Solomon et al., 1993). The end result is unnecessary—and often extreme—patient suffering, leading Grond, Zech, Schug, Lynch, and Lehmann (1991) to conclude that “most doctors are much more aware of the side effects of opioids . . . than of the side effects of pain.” Such physicians are not unsympathetic to the plight of patients, but likely are concerned about violating professional (not to mention legal) obligations to the degree that it influences their clinical practice.

To ensure optimal pain management of patients, then, physicians need to be reassured that it is not unethical to treat pain at the end of life aggressively with opioids that may risk hastening death. Quite the contrary, it is an ethical obligation to do so, and here the Rule of Double Effect (RDE) can provide valuable guidance.

Rule of Double Effect

First formulated by Thomas Aquinas (2012) in relationship to killing in self-defense, the RDE is used to determine whether an action that has two effects—one good and one bad—is morally permissible. The RDE has four basic components:

  1. 1. The act itself must be, at worst, morally neutral.

  2. 2. The bad effect cannot be the means to the good effect.

  3. 3. The good effect must outweigh the bad effect (principle of proportionality).

  4. 4. The agent must only intend the good effect, although the bad effect may be foreseen (Garcia, 1994).

In a medical context, the RDE has been applied to several situations. Since it grew out of Roman Catholic theology, historically one application has been to delineate contexts where ending a pregnancy is morally acceptable. The standard “termination to save the life of the mother” would not fulfill all four requirements, since the “bad effect” (in this case, terminating the fetus) is the means to the good effect (in this case, improved health of the formerly pregnant woman). Performing a hysterectomy on a pregnant woman with uterine cancer, however, would fulfill the requirements, since the intention is to treat cancer and the end of the pregnancy is an anticipated—but not intended—side effect of hysterectomy (Shaw, 2002).

The most common modern medical application of the RDE pertains to the use of opioids at the end of life, given the potential side effect of depressing respiratory drive (AMA Council on Ethical and Judicial Affairs, 1992; Foley, 1997; Pellegrino, 1992; President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, 1983; The Hastings Center, 1987). If analgesia is the good effect and hastened death is the bad effect, the RDE is applied to opioid administration in the following way:

  1. 1. The act of administering an FDA-approved medication for a condition it is indicated for is not morally bad.

  2. 2. Pain is not relieved through the patient’s death but rather through direct action of the medication.

  3. 3. In cases where the pain is severe and the patient’s prognosis is poor, the benefit of analgesia can outweigh the harm of hastened death.

  4. 4. In appropriately titrating the opioids, the physician’s intention is to treat pain, with the foreknowledge—though not the intention—that death might be hastened.

The first component recognizes the obvious, but the latter three provide important clarifications.

The second component draws a sharp distinction between intensive analgesia and euthanasia. Whereas the latter “puts the patient out of her misery” (i.e., ends her life in order to end her pain), opioids directly treat pain or dyspnea. Relief of symptoms may well occur absent the death of the patient—indeed, that is the overriding objective—and thus the bad effect is not the means to the good effect.

The third component underscores that the discussion is only about extreme symptoms at the end of life. A patient whose symptoms are more moderate—or who could survive for a long period if she gets through an intense period of her disease—would not meet the proportionality requirement of the RDE.

The final component emphasizes the physician’s intention, which—again in contrast to euthanasia—is not to end the patient’s life. Rather, it is to effectively treat pain, with the recognition that this could possibly hasten the patient’s death.

The second and fourth conditions of the RDE, in particular, distinguish intensive symptom management from euthanasia. This is based on the presumption that euthanasia is not ethically permissible (p. 143), and thus intensive symptom management must somehow be intrinsically different for it to be acceptable. If euthanasia were permissible, there would be no need to appeal to the RDE because there would be no “bad effect” that requires justification (Sulmasy & Pellegrino, 1999).

Criticisms of the RDE

The RDE is not without controversy, however, with critics leveling several criticisms against it (Quill, Dresser, & Brock, 1997). The first pertains to its origin in a particular religious tradition, whose beliefs many people in a multicultural society do not share. Why should a Thomistic theory—applied to a modern problem—be persuasive for persons who do not share its theological presuppositions?

This, however, seems a weak accusation, since the RDE makes no explicit appeal to any religious text or set of beliefs. The fact that its historical origin lies in a religious context is irrelevant if its basic claims are reasonable and argumentation valid. The US Supreme Court seems to believe this to be the case, implicitly referencing the RDE in drawing a distinction between intensive symptom management and physician-assisted dying: “It is widely recognized that the provision of pain medication is ethically and professionally acceptable even when the treatment may hasten the patient’s death if the medication is intended to alleviate pain and severe discomfort, not to cause death” (Vacco v. Quill, 1997, emphasis added).

Even if the RDE were confined in its applicability to people who share the religious beliefs out of which it arose—thus limiting its scope—this would not negate its validity to that specific group. Physicians who have no moral qualm about potentially hastening death—viewing symptom management as an overriding obligation or perhaps even approving of euthanasia—have no need for the RDE to counteract an opiophobia from which they do not suffer. Rather, it is precisely those physicians who are concerned about hastening death—and who might well share the same spiritual heritage as the formulators of the RDE—who rely on the rule for reassurance. As Sulmasy and Pellegrino (1999) observe, for clinicians who are “fearful of unwittingly participating in euthanasia if a patient’s death is hastened, . . . the rule of double effect provides moral reassurance and thus encourages optimal care of the dying.”

A second criticism has to do with the physician’s intention, which is not always monolithic (Brody, 1993; Quill, 1993). When a patient is suffering and has expressed a wish for hastened death, can a physician honestly say that his only intention in treating pain aggressively is analgesia? If he would not be saddened to see a patient—whose quality of life may be so poor that she longs for death—die sooner, does that amount to “intending” the patient’s death and thus violate the last condition of the RDE?

Critics are surely right in questioning the purity of intention because it is possible to intend multiple things to various degrees. But this does not invalidate the RDE. Just because a physician would not be saddened to see a patient die sooner does not mean that he is acting purposefully to achieve that goal. The key here is not the purity of intention but rather the primacy of intention, which should be on relief of suffering. As Foley (1997) writes, “Saying that physicians struggle with doubts about their intentions is not the same as saying that their intention is to kill.”

Another criticism has to do with responsibility, noting that people are generally held to account not only for what they intended to do but also what could have reasonably been foreseen as a result of their actions. As the President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research (1983) put it, “People are equally responsible for all of the foreseeable effects of their actions, thereby having no need for a policy that separates ‘means’ from ‘merely foreseen consequences.’ ” The most important component, critics argue, is not the physician’s intended outcome but the proportionality of the sought-after good with the risked “bad effect” (Quill et al., 1997). The moral value of an action would not, therefore, depend on the motivation that prompts it, but rather on the consequences it predictably results in.

The RDE certainly recognizes the limits of intention, with the requirement of proportionality providing a substantial safeguard. Intending to eradicate pain is insufficient justification for placing any patient at risk of hastened death, especially when that patient could have lived for a long period after the pain eventually subsided. Only patients whose symptoms and prognosis meet the proportionality requirement are “covered” by the RDE, and the emphasis on intention ensures judicious upward titration, thus minimizing the risk of serious side effects.

Even admitting that good intentions—which, as the old saying goes, pave the road to hell (Boswell, 1791)—are no defense for an act that leads to excessive harm, this does not mean that intention does not matter. Imagine, for example, two physicians working in the intensive care unit, both caring for patients in extreme pain at the end of life. Physician A judiciously titrates opioid dosing in an attempt to control the patient’s pain, starting at 5 mg per hour and eventually reaching 50 mg per hour. At that point the patient is comfortable but breathing quite slowly and soon thereafter dies peacefully. Physician B, on the other hand, has many other responsibilities, acknowledges that the patient is going to die anyway and thus gives a 50 mg bolus of morphine to “get this over with.” Although the end result is the same, there exists a moral difference between these two physicians’ actions, especially as viewed through the lens of virtue ethics (p. 40).

The next criticism of the RDE is directly related to this focus on physician intention, arguing that the rule implicitly prioritizes the physician’s own professional responsibility over the patient’s autonomy, which should be the primary emphasis (Meisel & Cerminara, 2004). What if the patient in the aforementioned example preferred to be “put out of her misery,” rather than waiting for Physician A to gradually titrate the opioid dose?

This criticism takes a rather simplistic view of autonomy, however, essentially equating it to “what the patient wants, the patient gets.” As noted in chapter 2, professional codes and obligations also play an important role in the discussion, if only to assure other patients that the treatments being offered are clinically indicated and ethically defensible. The RDE is not merely a salve for the physician’s conscience; it is also a valuable guide—in applicable situations—for navigating complex situations.

A final criticism focuses on the unintended consequences of the RDE itself, which could actually impede—rather than make possible—optimal pain management. For instance, a physician might recognize that he would be relieved if a long-suffering patient died and thus be concerned that administering high doses of opioids could reflect an intention to hasten death. Rather than reassuring the physician, the RDE’s focus on intention could discourage him from effectively treating the patient’s pain, out of concern for violating the fourth requirement.

Here it is important to acknowledge the multifaceted nature of intention and focus on the primacy rather than purity of it. If a patient truly wished to have her life end, the physician’s sharing in that hope is a sign of empathy, not unethicality. As long as this is not the primary reason for upward titration of opioids—and effective symptom management is—this should not represent a barrier to optimal care at the end of life.

This reveals another unintended consequence of relying on the RDE: the perpetuation of the misperception that intensive symptomatic treatment is likely to hasten death. The purpose of the RDE (related to palliative care) is to empower physicians to optimally treat suffering at the end of life, unencumbered by ethical concerns that unintentionally hastening death constitutes euthanasia. But while this may be extremely important reassurance to those who would otherwise be burdened by this concern, routinely appealing to the RDE as justification for intensive symptom management could perpetuate the belief that high-dose opioids are likely to hasten death. Indeed, some have gone so far as to call describe intensive pain treatment as “double effect euthanasia” (Vaux, 1989).

Yet while the physiologic mechanism by which opioids depress respiratory drive—especially upon their initiation (Hill 1993)—is well-established, the probability that they will hasten death by this mechanism is actually quite low (Whittaker, 2013). Pain itself is a significant respiratory stimulant, which explains why a person hyperventilates when he strikes his thumb with a hammer. It also explains why a patient in severe pain will continue to breathe—comfortably—after receiving an opioid dose that would have led to frank apnea in a healthy patient who is free of pain.

In point of fact, rather than being likely to hasten death, intensive symptom management may actually prolong life. Opioids have been shown to prolong survival after ventilator withdrawal (Edwards, 2005), likely by decreasing oxygen demand (Bakker, Jansen, Lima, & Kompanje, 2008). Indeed, a large study of hospice patients found that opioid use was directly—rather than inversely—correlated with duration of life for terminally ill patients (Portenoy et al., 2006). Editorials in leading medical journals have put it even more succinctly: “Morphine kills the pain, not the patient” (Sykes, 2007). Thus Fohr (1998) concludes: “In the case of medication to relieve pain in the dying patient, the RDE should be rejected not on ethical grounds, but for a lack of medical reality.”

This empirical data suggests that overreliance on the RDE to justify pain management at the end of life may be counterproductive. In attempting to reassure a select few who fear stepping over the line into euthanasia that they are not doing anything wrong, a great many others may be led to believe that hastened death is inevitable, rather than rare. Ethicists may bear some responsibility for this perception. A recent study of medical school ethics educators found that one-third believed that opioids were “likely to cause significant respiratory depression that could hasten death” (Macauley, 2012). Understandably—based on this misperception—the RDE was routinely appealed to as justification for such intensive pain management.

In light of the uncertain correlation between opioids and hastened death, some have argued that appealing to the RDE to justify optimal EOL care may not only be counterproductive but also unnecessary. After all, the RDE is designed to justify actions which inevitably lead to both good and bad effects (Von Gunten, 2015), such as hysterectomy for uterine cancer in a pregnant woman. But in the case of symptomatic treatment at the end of life, the worst that can be said about opioids is that they could possibly hasten death. Where a bad effect is merely possible, it should be evaluated as a risk of treatment. In this respect, symptomatic treatment at the end of life is no different than any nearly every other medical treatment, which involves a balancing of the hoped-for benefit with the possible complications or side effects (Emanuel, Ferris, & von Gunten, 2002). And while the bad effect (i.e., death) is indeed serious, this is also a risk of many other medical procedures. As Bleich observes (1994), “[The] assumption of prudent risk is synonymous with life. It is only when the bad or immoral effect of an action is foreseeable as a matter of certainty, near certainty or strong likelihood that a moral dilemma arises.”

Given the uncertain—or even unlikely—correlation between opioid treatment at EOL and hastened death, the RDE should not provide the sole justification of the former. Of course, some of the elements of the RDE would naturally factor into one’s analysis, such as whether the benefits of the treatment outweigh the risks. And to those with heightened concern about the ethicality of potentially hastened death, the RDE could provide valuable reassurance, especially given its roots in a religious tradition that might have prompted that very concern.

But for most physicians, all four components of the RDE need not necessarily be satisfied to justify a treatment with only potential complications. Otherwise, overreliance on the RDE could actually perpetuate the very injustice it is designed to prevent. As Angell (1982) notes, “I can’t think of any other area in medicine in which such an extravagant concern for side effects so drastically limits treatment.”

Patient or Surrogate Refusal of Symptom Management

Regulatory barriers and physicians’ opiophobia are not the only impediments to optimal pain management at the end of life. Patients and their surrogates might also refuse symptomatic treatment that is offered. The proper response to such a refusal depends on correctly identifying the basis for it, which can be classified into three general categories with some degree of overlap (Figure 7.1).

Figure 7.1 Reasons for refusing symptomatic treatment

Figure 7.1 Reasons for refusing symptomatic treatment

One medical reason for refusal of opioids is the stigma associated with them. The opioid epidemic frequently makes headlines, and some medications (such as Oxycontin) have been mentioned by name (Ryan, Girion, & Glover, 2016). While hospice clinicians recognize methadone as an inexpensive, broadly effective, and exceedingly well-absorbed pain reliever (Chhabra & Bull, 2008), many patients only know it only as treatment for patients suffering from opioid addiction. If a patient is reluctant to receive opioids based on fear of stigma, the team can reiterate that the treatment is medically indicated and review the benefits of it. It may also be helpful to rotate to another medication that achieves the same goal of pain relief but might be less commonly known or carry less negative connotations (e.g., hydromorphone).

Beyond the stigma of opioids, some patients may be concerned about becoming addicted themselves, in spite of their limited life expectancy. They might also have misinterpreted a discussion of tolerance and physiologic dependence as equivalent to addiction, or view any dependence on medication as a sign of weakness. This concern might also betray an unspoken hope of exceeding the prognostic models, thus making addiction more of a relevant concern over a longer period of time.

Here it is important to distinguish between addiction, physiological dependence, and tolerance, stressing that the latter two are normal physiological mechanisms. With regard to physiologic dependence in particular, it may also be helpful to note that this phenomenon is not unique to opioids, as patients may become accustomed to—a more pleasant term than “dependent on”—other medications and thereby suffer negative effects from their discontinuation. For instance, a patient receiving long-term steroids will experience an adrenal crisis if they are withdrawn precipitously, but there is (properly) no stigma attached to this. Reasoning by analogy, just as one would never say that a patient with diabetes is “addicted” to insulin—even though the patient is dependent on that medication for health and to avoid negative symptoms—so also we should not say that a patient physiologically dependent on opioids is “addicted” to them.

Another reason that a patient might refuse optimal symptomatic treatment—which blurs the line between the medical and psychological—is her reliance on pain as a “marker of disease.” If at least some of the symptom remains, then resolution of it signifies the improvement of the underlying disease, while a worsening may reflect disease progression. If analgesics completely eradicate pain—so the argument goes—then improvement in the underlying condition would not be noticeable. By the same token, if the patient were receiving too much analgesia, then incremental worsening of the condition would not be identified.

In such cases, patients can be reassured that there are other ways to mark the course of the disease beyond symptomatology. This might include blood counts or radiological scans. Focusing more on the psychological impetus for treatment refusal, it is reasonable to offer psychotherapy and empathic listening to directly address the patient’s anxiety, while reassuring her that she will be fully informed of any developments in her condition.

In some cases, though, patients may still prefer to have some minor level of discomfort for the resulting “peace of mind.” In cases such, it may not be appropriate to say that the patient is “suffering” from the symptom, since she has decided that she is on the whole more content with a small level of discomfort than being entirely symptom-free. Ideally ongoing conversations can identify the optimal balance of analgesia and reassurance for the patient, addressing whatever psychological or spiritual concerns make monitoring disease progression so important.

This highlights the distinction between pain or dyspnea on the one hand and suffering on the other. Certainly these terms are related, as both pain and dyspnea can cause suffering. But neither is synonymous with suffering, as some patients who might be experiencing high levels of pain or dyspnea would not report extreme suffering, while others would report it in the context of lesser measured levels of pain or dyspnea.

Ultimately, suffering is a broader concept than pain, taking into account nonphysical considerations that surpass the ability of quantitative scales to measure. It is just as subjective as either pain or dyspnea, and even more individual. If, indeed, “pain is whatever the experiencing person says it is, existing whenever the experiencing person says it does” (McCaffery & Beebe, 1989), suffering is at least equally so, encompassing both the physical and non-physical realms. It is ultimately up to the patient to determine what will increase or ameliorate her suffering, and the analgesic regimen should be tailored to optimally achieve the latter.

A classic example of this involves the patient who values lucidity over comfort. Sedation is a well-known—and, unlike respiratory depression, inevitable—side effect of opioids. In certain cases there needs to be a “trade-off” between lucidity and analgesia, with the severity of the patient’s symptoms sometimes making it impossible to achieve both. Studies have shown that when faced with that choice, physicians generally prioritize comfort over lucidity, while patients are more likely to value lucidity even at the price of some degree of pain (Steinhauser et al., 2000).

Here it is important to clearly delineate what the primary goal of treatment is through direct discussion of this potentially tenuous balance. Palliative care should strive to minimize suffering, not merely physical pain. A patient who is pain-free could suffer greatly for wont of the opportunity to meaningfully engage with her loved ones at the end of her life. In an attempt to find the optimal balance, it may be possible to reduce the opioid dose without incurring extreme pain or to treat the sedation directly with a psychostimulant (Reissig & Rybarczyk, 2005).

Moving into the explicitly spiritual realm, another reason for refusing symptomatic treatment is a feeling of undeservedness. Some patients may have such a deep sense of past transgressions that they feel that the suffering they are experiencing is justified (Periyakoil, 2015). This is an explicitly spiritual issue which demands a spiritual response that addresses the underlying basis for this belief and identifies a way forward whereby the patient can make peace with her past while also (hopefully) receiving present-day relief of pain.

Lastly, a patient might believe that suffering is required by her religious tradition. This is not so much a question of deservedness as much as devoutness and thus merits a response rooted in that patient’s religious tradition. The response to so-called “redemptive suffering” is addressed in detail later (p. 496).

All of the reasons cited so far can be voiced by patients themselves, and hopefully with attentive listening, accurate identification of the underlying issue, and focused discussion the patient’s concerns can be addressed and allayed. If not, the clinician may be forced to accept the patient’s request for suboptimal analgesia, at least temporarily. Admittedly, it can be very difficult to watch a patient endure pain “needlessly” when effective treatments are available. But if palliative care takes seriously the notion of “total pain,” then it must recognize that increased physical discomfort may be the price that some patients are willing to pay for peace of mind or spirit. While one might wish it were not so, some patients find their balance in that place, and the job of palliative care clinicians is to simultaneously respect the patient’s choice while also thoughtfully working with the patient to identify one that might better meet her physical as well as nonphysical needs (Macauley, 2011).

Clinicians should also take the time to reflect on the obligation they feel to eradicate suffering, specifically whose suffering it is that they are trying to address: the patient’s, or their own? It is indeed a noble calling to ameliorate the suffering of another person, but when the drive to do so stems from the clinician’s own need to discharge his moral duty, it runs the risk of prioritizing the clinician’s values over those of the patient.

Up to this point, the discussion has assumed that the patient herself was refusing symptomatic treatment. But in some situations the patient might lack decision-making capacity and the refusal of symptom management is made by a surrogate decision-maker. As was the case with a patient’s direct refusal of symptomatic treatment, a surrogate’s refusal could stem from many reasons. The first step, once again, is to identify the precise reason why the surrogate believes the patient would refuse the treatment in question. Here the precise question is critical: What do you think the patient would choose? appropriately invokes substituted judgment, whereas What do you think we should do? opens the door for the surrogate to insert his own values (p. 63). Relevant factors include written advance directives, previous comments about the patient’s own health status, and overall personal, spiritual, or religious values. The way in which the patient responded to previous episodes of discomfort or pain—including prior willingness to accept analgesia—are all relevant.

But none of these may be determinative. As noted earlier (p. 72), it is often not clear what the patient would have wanted. Most patients have not completed a written advance directive, and even those who did may have expressed rather general sentiments, or perhaps only named a health care agent. The majority of patients have done neither, leaving a surrogate—either identified through a state hierarchy or informally identified by the medical team (p. 62)—to make the decision. That surrogate necessarily has to rely on inferences from what the patient has said in the past and how that patient has lived her life.

The fact that the patient named that person as her decision-maker—or that that person is deemed by statute or practice as the appropriate surrogate—reflects not only the closeness of the relationship with the patient but also likely the overlap in their values. Thus, if the surrogate declines symptomatic treatment based on a notion of redemptive suffering, this could reflect a shared spiritual belief. He might also be deciding on the basis of a patient’s longstanding aversion to taking medication or prioritizing of lucidity over pain relief.

The patient’s context may have changed, though, since she espoused those beliefs. For instance, a patient who was generally averse to medication—during a period of good health and minimal symptomatology—might well be willing to accept medications when faced with severe symptoms. Similarly, concerns for addiction may fade in the face of a limited life expectancy.

The team, therefore, must also consider whether the surrogate could be acting based on his own values, which the patient may not share. A classic example is the stigma associated with opioids, especially if the surrogate has had personal experience with addiction. Similarly, when a surrogate prioritizes lucidity over analgesia, it can be challenging to determine whether he is doing so because that is what the patient would have done or because he himself is seeking something from the patient—such as forgiveness or simply additional moments of meaning—even at the price of the patient’s increased suffering.

If this is the case, the general methods noted above for helping surrogate decision-makers navigate complexity and avoid excessive burden (p. 77) can be helpful. The surrogate may feel torn about advocating a course of action that violates his own personal beliefs, so reassurance that he is not really making the decision—rather, he is simply communicating what the patient would have chosen—can be helpful. Sometimes a surrogate may need to recuse himself if the sense of “moral complicity” is too great. In other situations, though, the surrogate may hold his ground and effectively demand that his instructions be followed, even though this would cause increased suffering to the patient.

Seeking additional input from others who know the patient may be helpful in such cases, not only from an informational point of view but also in terms of advocacy, as these additional stakeholders may have more sway with the primary surrogate than the clinical team does. If the impasse remains, however—and there is no clear evidence the surrogate is using appropriate substituted judgment—the medical team may have to prioritize beneficence by ensuring adequate symptomatic treatment. For while patients have the right to make “bad choices” for themselves, surrogates do not have the right to inflict suffering on patients based on their own value system.

There are several ways to accomplish this. The team could request an ethics consultation, in order to better delineate the surrogate’s rights and the team’s obligation. There may be other avenues to engage an impartial person to mediate the dispute. In extreme cases, an alternate surrogate may need to be identified. If the original surrogate is not specifically named in an advance directive as the primary agent, most states have a hierarchy of surrogates whereby others with some standing can challenge the decisions of the person at the top of the hierarchy (p. 62). In the few states without a hierarchy, there is nothing to prevent the clinical team from turning to someone else for medical decisions, unless the original surrogate opts to pursue guardianship in court. In that case, hopefully the court would take into account clinical concerns about undertreatment of suffering and appoint as guardian someone who is able to best advocate for the patient’s comfort, as well as her values. This would also be the requisite course of action if the original surrogate decision-maker were the health care agent, thus requiring judicial appeal to displace him from that role.

Justice Concerns

True “opiophobia” inappropriately influences how a physician treats all his patients, and an individual patient’s refusal of analgesia is her own choice. (Granted, one that might stem from misperceptions and certainly merits further exploration.) However, there are situations where inadequate analgesia may reflect personal or systemic discrimination against a certain subgroup of patients who are seeking relief of suffering. If all patients in pain are not treated the same, this raises profound justice concerns.

That this occurs in medical practice is irrefutable. Studies have shown that racial minorities, the poor, the elderly, women, and patients with HIV/AIDS are all less likely than white males to receive adequate chronic pain management (Nguyen, Ugarte, Fuller, Haas, & Portenoy, 2005). This could be due to a variety of factors, including an assumption that certain groups are more likely to be addicted to opioids and thus to seek them for secondary gain. Some groups are thought to have different “pain thresholds,” suggesting that requests for opioid medications may be unwarranted by the level of pain experienced. Some physicians have even posited significant biological differences between groups. For instance, the majority of medical students and residents in one recent study believed there was a biological difference—such as altered sensitivity of nerve endings or skin thickness—between white and African American patients which influenced how they would treat pain in the latter (Hoffman, Trawalter, Axt, & Oliver, 2016).

Many of the ethical issues addressed in this textbook are complex and multilayered, requiring a balancing of competing obligations and identification of what in many cases is the “least bad option.” This is not such an issue, for the ethics are clear: it is unjust to treat similar patients differently, based on nonclinical indicators such as the ones noted here.

It may be tempting to write off such discrimination as evidence of frank prejudice, but it is not as simple as that. There may exist genuine clinical misunderstanding which leads a physician to undertreat pain in a certain instance (such as by misdiagnosing undertreated pain as “addiction,” thus leading to pseudoaddictive behavior). Or perhaps this a reflection of cognitive bias—especially attributive bias—whereby certain traits are attributed to specific groups (Saposnik, Redelmeier, Ruff, & Tobler, 2016).

The solution seems obvious, namely, to avoid such discrimination. But this is easier said than done, given the multifaceted nature of pain, the shifting focus from pain as “the fifth vital sign” to daily milligram limits meant to combat the opioid epidemic, and the lingering effects of opiophobia. Implicit bias, as the name suggests, is unrecognized by the physician, who may believe that he is not prejudiced in any way. Studies have shown, however, that implicit bias frequently impacts the clinical decisions physicians make (Blair, Steiner, & Havranek, 2011). Hopefully, raising the implications of this bias to consciousness can help prevent it from impacting patient care (Devine, Forscher, Austin, & Cox, 2012), as can establishing best practice standards for treating pain. Absent such recognition and direct engagement, however, there is no reason to believe that unjust disparities in pain treatment will not continue.

Summary Points

  • For most of the history of Western medicine, relatively little attention was paid to the treatment of symptoms at the end of life. This changed following World War II, and especially after the mid-century pioneering work of Cicely Saunders.

  • Pain is multifaceted, going beyond physical discomfort to include emotional, psychological, and spiritual domains.

  • Suffering is a deeper concept which recognizes varying pain tolerances as well as patient goals. In some cases, eradicating pain could cause greater suffering, if the patient’s other goals are not respected.

  • The primary justification for intensive treatment of pain and dyspnea at the end of life is the Rule of Double Effect.

    • Some have criticized the rule because of its theological roots, emphasis on physician intention and supposed deprioritizing of patient autonomy.

    • None of these criticisms is compelling. The rule can be used to justify intensive symptomatic treatment but it is largely unnecessary. Instead of hastening death, opioids frequently prolong life when used appropriately.

  • Patients may refuse symptomatic treatment for a variety of medical, psychological, and spiritual reasons. Unless the true reason is identified, it is impossible to effectively address the patient’s concerns.

  • Surrogate refusal of symptomatic treatment is more concerning, as it may not reflect accurate substituted judgment. In such cases, it may be ethically requisite to override the refusal in order to minimize patient suffering.

  • There exist profound justice-based concerns about inequitable treatment of pain across disease, racial, and ethnic groups. These can be caused either by overt or implicit bias or clinical misunderstandings.


Ahmedzai, S. (1997). Five years, five threads. Prog Pall Care,5(6), 235–237.Find this resource:

    Al-Mazroa, A. A., & Abdel Halim, R. E. (1989). Anesthesia 1000 years ago. Royal Soc Med Intl Congress Symp Series, 134, 46–47.Find this resource:

      AMA Council on Ethical and Judicial Affairs. (1992). Decisions near the end of life. JAMA, 267(16), 2229–2233.Find this resource:

      American Society of Addiction Medicine. (2011). Definition of addiction. Retrieved from

      Angell, M. (1982). The quality of mercy. N Engl J Med, 306(2), 98–99. doi:10.1056/NEJM 198201143060210.Find this resource:

      Aquinas, T. (2012). Summa theologiae secunda secundae, 1-91: The complete works of Saint Aquinas. Lander, WY: Aquinas Institute for the Study of Sacred Doctrine.Find this resource:

        Bakker, J., Jansen, T. C., Lima, A., & Kompanje, E. J. (2008). Why opioids and sedatives may prolong life rather than hasten death after ventilator withdrawal in critically ill patients. Am J Hosp Palliat Care, 25(2), 152–154. doi:10.1177/1049909108315511Find this resource:

        Baraka, A. (2000). Historical aspects of opium. 1982. Middle East J Anaesthesiol, 15(4), 423–436.Find this resource:

        Bennett, D. S., & Carr, D. B. (2002). Opiophobia as a barrier to the treatment of pain. J Pain Palliat Care Pharamcother, 16(1), 105–109.Find this resource:

        Blair, I. V., Steiner, J. F., & Havranek, E. P. (2011). Unconscious (implicit) bias and health disparities: Where do we go from here? Perm J, 15(2), 71–78.Find this resource:

        Bleich, J. D. 1994. On the ethics of pain management. Cancer Invest,12(3), 362–363.Find this resource:

        Bonica, John J. (1953). The management of pain; with special emphasis on the use of analgesic block in diagnosis, prognosis, and therapy. Philadelphia: Lea & Febiger.Find this resource:

          Booth, M. (1998). Opium: A history (1st US ed.). New York: St. Martin’s Press.Find this resource:

            Boswell, J. (1791). The life of Samuel Johnson, LL.D., comprehending an account of his studies and numerous works in chronological order . . . the whole exhibiting a view of literature and literary men in Great-Britain for near half a century during which he flourished. London: H. Baldwin.Find this resource:

              Brody, H. (1993). Causing, intending, and assisting death. J Clin Ethics,4(2), 112–117.Find this resource:

              Chhabra, S., & Bull, J. (2008). Methadone. Am J Hosp Palliat Care, 25(2), 146–150. doi:10.1177/1049909107312597Find this resource:

              Cherny, N. I., Coyle, N., & Foley, K. M. (1994). Suffering in the advanced cancer patient: a definition and taxonomy. J Palliat Care, 10(2), 57–70.Find this resource:

              Cope, D. K. (2010). Intellectual milestones in our understanding and treatment of pain. In S. Fishman, J. Ballantyne, J. P. Rathmell, & J. J. Bonica (Eds.), Bonica’s management of pain (4th ed., pp. 1–12). Baltimore, MD: Lippincott, Williams & Wilkins.Find this resource:

                Council on Ethical and Judicial Affairs, American Medical Association. (1992). Decisions near the end of life. JAMA, 267(16), 2229–2233.Find this resource:

                Davis, M. P., & Walsh, D. (2004). Cancer pain: How to measure the fifth vital sign. Cleve Clin J Med, 71(8), 625–632.Find this resource:

                Desbiens, N. A., & Wu, A. W. (2000). Pain and suffering in seriously ill hospitalized patients. J Amer Geratr Soc, 48(5 Suppl.), S183–S186.Find this resource:

                Devine, P. G., Forscher, P. S., Austin, A. J., & Cox, W. T. (2012). Long-term reduction in implicit race bias: A prejudice habit-breaking intervention. J Exp Soc Psychol, 48(6), 1267–1278. doi:10.1016/j.jesp.2012.06.003Find this resource:

                Dowell, D., Haegerich, T. M., & Chou, R. (2016). CDC guideline for prescribing opioids for chronic pain—United States, 2016. JAMA, 315(15), 1624–1645. doi:10.1001/jama.2016.1464Find this resource:

                Edwards, M. J. (2005). Opioids and benzodiazepines appear paradoxically to delay inevitable death after ventilator withdrawal. J Palliat Care, 21(4), 299–302.Find this resource:

                Emanuel, L. L., Ferris, F. D., & von Gunten, C. F. (2002). EPEC. Education for physicians on end-of-life care. Am J Hosp Palliat Care, 19(1), 17; author reply 17–18.Find this resource:

                Express Scripts. (2014). A nation in pain. Retrieved from

                Fohr, S. A. (1998). The double effect of pain medication: Separating myth from reality. J Palliat Med, 1(4), 315–328. doi:10.1089/jpm.1998.1.315Find this resource:

                Foley, K. M. (1997). Competent care for the dying instead of physician-assisted suicide. N Engl J Med, 336(1), 54–58. doi:10.1056/NEJM199701023360109Find this resource:

                Food and Drug Administration. (2017). A brief overview of risk evaluation and mitigation strategies. Retrieved from

                Garcia, J. L. (1994). Double effect. In W. T. Reich (Ed.), Encyclopedia of bioethics (pp. 636–641). New York: Simon & Schuster.Find this resource:

                  Grond, S., Zech, D., Schug, S. A., Lynch, J., & Lehmann, K. A. (1991). Validation of World Health Organization guidelines for cancer pain relief during the last days and hours of life. J Pain Symptom Manage, 6(7), 411–422.Find this resource:

                  Haller, J. S. Jr. (1989). Opium usage in nineteenth century therapeutics. Bull N Y Acad Med, 65(5), 591–607.Find this resource:

                  Hamilton, G. R., & Baskett, T. F. (2000). In the arms of Morpheus: The development of morphine for postoperative pain relief. Can J Anaesth, 47(4), 367–374. doi:10.1007/BF03020955Find this resource:

                  Hill, C. S. Jr. (1993). The barriers to adequate pain management with opioid analgesics. Semin Oncol, 20(2 Suppl. 1), 1–5.Find this resource:

                  Hoffman, K. M., Trawalter, S., Axt, J. R., & Oliver, M. N. (2016). Racial bias in pain assessment and treatment recommendations, and false beliefs about biological differences between blacks and whites. Proc Natl Acad Sci U S A, 113(16), 4296–4301. doi:10.1073/pnas.1516047113Find this resource:

                  Howard-Jones, N. (1947). A critical study of the origins and early development of hypodermic medication. J Hist Med Allied Sci, 2(2), 201–249.Find this resource:

                  Huang, F. Y., & Emanuel, L. L. (1995). Physician aid in dying and the relief of patients’ suffering: Physicians’ attitudes regarding patients’ suffering and end-of-life decisions. J Clin Ethics, 6(1), 62–67.Find this resource:

                  Islam, M. M., & McRae, I. S. (2014). An inevitable wave of prescription drug monitoring programs in the context of prescription opioids: Pros, cons and tensions. BMC Pharmacol Toxicol, 15, 46. doi:10.1186/2050-6511-15-46Find this resource:

                  Jennings, B., Ryndse, T., D’Onofrio, C., & Baily, M. A. (2003). Access to hospice care: Expanding boundaries, overcoming barriers. Hastings Cent Rep, Suppl., S3–S7, S9–S13, S15–S21 passim.Find this resource:

                  Lundberg, G. D. (1988). “It’s over, Debbie” and the euthanasia debate. JAMA,259(14), 2142–2143.Find this resource:

                  Macauley, R. (2011). Patients who make “wrong” choices. J Palliat Med, 14(1), 13–16. doi:10.1089/jpm.2010.0318Find this resource:

                  Macauley, R. (2012). The role of the principle of double effect in ethics education at US medical schools and its potential impact on pain management at the end of life. J Med Ethics, 38(3), 174–178. doi:10.1136/medethics-2011-100105Find this resource:

                  Macht, D. I., Herman, N. B., & Levy, C. S. (1915). A quantitative study of cutaneous analgesia produced by various opium alkaloids. Proc Natl Acad Sci U S A, 1(12), 582–585.Find this resource:

                  McCaffery, M., & Beebe, A. (1989). Pain: Clinical manual for nursing practice. St. Louis, MO: Mosby.Find this resource:

                    Meisel, A., & Cerminara, K. L. (2004). The right to die: The law of end-of-life decisionmaking (3rd ed.). New York: Aspen.Find this resource:

                      Nguyen, M., Ugarte, C., Fuller, I., Haas, G., & Portenoy, R. K. (2005). Access to care for chronic pain: Racial and ethnic differences. J Pain, 6(5), 301–14. doi:10.1016/j.jpain.2004.12.008Find this resource:

                      Pellegrino, E. D. (1992). Doctors must not kill. J Clin Ethics, 3, 95–102.Find this resource:

                      Periyakoil, V. S. (2015). A never-ending battle. N Engl J Med, 373(25), 2399–2401. doi:10.1056/NEJMp1505976Find this resource:

                      Pohlman, K. J. (1990). Pain control: Euthanasia or criminal act? Focus Crit Care, 17 (3), 260–261.Find this resource:

                      Portenoy, R. K., Sibirceva, U., Smout, R., Horn, S., Connor, S., Blum, R. H., . . . Fine, P. G. (2006). Opioid use and survival at the end of life: A survey of a hospice population. J Pain Symptom Manage,32(6), 532–540. doi:10.1016/j.jpainsymman.2006.08.003Find this resource:

                      President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. (1983). Deciding to forego life-sustaining treatment: A report on the ethical, medical, and legal issues in treatment decisions. Washington, DC: U.S. Government Printing Office.Find this resource:

                        Prioreschi, P., Heaney, R. P., & Brehm, E. (1998). A quantitative assessment of ancient therapeutics: Poppy and pain in the Hippocratic Corpus. Med Hypotheses, 51(4), 325–331.Find this resource:

                        Quill, T. E. (1993). The ambiguity of clinical intentions. N Engl J Med, 329(14), 1039–1040. doi:10.1056/NEJM199309303291414Find this resource:

                        Quill, T. E., Dresser, R., & Brock, D. W. (1997). The Rule of Double Effect—a critique of its role in end-of-life decision making. N Engl J Med, 337(24), 1768–1771. doi:10.1056/NEJM199712113372413Find this resource:

                        Reich, W. T. (1989). Speaking of suffering: A moral account of compassion. Soundings, 72(1), 83–108.Find this resource:

                        Reissig, J. E., & Rybarczyk, A. M. (2005). Pharmacologic treatment of opioid-induced sedation in chronic pain. Ann Pharmacother, 39(4), 727–731. doi:10.1345/aph.1E309Find this resource:

                        Ryan, H., Girion, L., & Glover, S. (2016, May 5). “You want a description of hell?” OxyContin’s 12-hour problem. Los Angeles Times. Retrieved from

                        Saposnik, G., Redelmeier, D., Ruff, C. C., & Tobler, P. N. (2016). Cognitive biases associated with medical decisions: A systematic review. BMC Med Inform Decis Mak, 16(1), 138. doi:10.1186/s12911-016-0377-1Find this resource:

                        Saunders, C. (1963). The treatment of intractable pain in terminal cancer. Proceed Royal Soc Med, 56, 195–197.Find this resource:

                        Saunders, C. (1964). The symptomatic treatment of incurable malignant disease. Prescribers J, 4(4), 68–73.Find this resource:

                          Saunders C. M. (Ed.). (1984). The management of terminal malignant disease. London: Hodder Arnold.Find this resource:

                            Schneider, C., Yale, S. H., & Larson, M. (2003). Principles of pain management. Clin Med Res, 1(4), 337–340.Find this resource:

                            Schweitzer, A. (1948). On the edge of the primeval forest & More from the primeval forest. Experiences and observations of a doctor in equatorial Africa. New York: Macmillan.Find this resource:

                              Shaw, A. B. (2002). Two challenges to the double effect doctrine: Euthanasia and abortion. J Med Ethics, 28(2), 102–104.Find this resource:

                              Smith, M. L., Orlowski, J., Radey, C., & Scofield, G. (1992). A good death: Is euthanasia the answer? Cleveland Clin J Med,59(1), 99–109.Find this resource:

                              Solomon, M. Z., O’Donnell, L., Jennings, B., Guilfoy, V., Wolf, S. M., Nolan, K., . . . Donnelley, S. (1993). Decisions near the end of life: Professional views on life-sustaining treatments. Am J Public Health, 83(1), 14–23.Find this resource:

                              Steinhauser, K. E., Voils, C. I., Clipp, E. C., Bosworth, H. B., Christakis, N. A., Tulsky, J. A. (2000). Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA, 284(19), 2476–2482. doi:10.1001/jama.284.19.2476Find this resource:

                              Sulmasy, D. P., & E. D. Pellegrino. (1999). The Rule of Double Effect: Clearing up the double talk. Arch Intern Med, 159(6), 545–550.Find this resource:

                              SUPPORT Principal Investigators. (1995). A controlled trial to improve care for seriously ill hospitalized patients: The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). JAMA,274(20), 1591–1598.Find this resource:

                                Sykes, N. P. (2007). Morphine kills the pain, not the patient. Lancet, 369(9570), 1325–1326. doi:10.1016/S0140-6736(07)60611-4Find this resource:

                                Tallmadge, G. K. (1946). Some anesthetics of antiquity. J Hist Med Allied Sci, 1(4), 515–520.Find this resource:

                                Teunissen, S. C., Wesker, W., Kruitwagen, C., de Haes, H. C., Voest, E. E., & de Graeff, A. (2007). Symptom prevalence in patients with incurable cancer: A systematic review. J Pain Symptom Manage,34(1), 94–104. doi:10.1016/j.jpainsymman.2006.10.015Find this resource:

                                The Hastings Center. (1987). Guidelines on the termination of life sustaining treatment and the care of the dying. Bloomington: Indiana University Press.Find this resource:

                                  Vacco v. Quill. (1997). U.S. 521 793.Find this resource:

                                    Vaux, K. L. (1989). Mercy, murder, & morality: Perspectives on euthanasia. The theological ethics of euthanasia. Hastings Cent Rep,19(1 Suppl.), 19–22.Find this resource:

                                    Von Gunten, C. F. (2015). Morphine and hastened death. Retrieved from

                                    Weissman, D. E., & Haddox, J. D. (1989). Opioid pseudoaddiction—an iatrogenic syndrome. Pain, 36(3), 363–366.Find this resource:

                                    Whittaker, M. R. (2013). Opioid use and the risk of respiratory depression and death in the pediatric population. J Pediatr Pharmacol Ther, 18(4), 269–276. doi:10.5863/1551-6776-18.4.269Find this resource:

                                    World Health Organization. (2017). WHO’s cancer pain ladder for adults. Retrieved from

                                    Wyatt, J. (2009). Matters of life & death: Human dilemmas in light of the Christian faith. Nottingham, England: Inter-Varsity Press.Find this resource:

                                      Younger, J., McCue, R., & Mackey, S. (2009). Pain outcomes: A brief review of instruments and techniques. Curr Pain Headache Rep, 13(1), 39–43.Find this resource: