———————————-Francis Peabody, 1927

Medical ethics is playing an increasing role in the analysis and improvement of medical practices, yet its predominant principles of autonomy, beneficence, nonmaleficence, and justice sometimes miss the underlying essence of being a physician [1, 2]. Alternative ethical approaches, including casuistry, phenomenology, and feminist, cross-cultural, narrative, and virtue ethics [3-9] place more emphasis on the importance of the relationship between and the virtues of the particular physician and patient making decisions together. Such analyses have deepened our understanding of other core physician obligations such as truth telling, informed consent, and confidentiality. An additional central obligation that has been relatively underemphasized is nonabandonment [1, 2, 10, 11].

Nonabandonment reflects a continuous caring partnership between physician and patient. The relationship may begin in health or in sickness, last through potential recovery or adjustment to chronic illness, and often continue to the patient's death [12-20]. It acknowledges and reinforces the centrality of an ongoing personal commitment to caring and problem solving between physician and patient. It also captures the essential qualities whereby physicians and patients commit to mutual decision making over time, even when the course is uncertain. Many dimensions of this covenant are articulated in the virtues of caring, fidelity, altruism, and devotion [1, 9-11], but no one term captures the particular importance of a long-term, engaged presence continuing until the patient dies or recovers.

An emphasis on nonabandonment, with its focus on the human relationship and open-ended commitment over time, is particularly mandated by two aspects of modern medicine. One is the prevalence of chronic illness, for which ethics must move beyond episodically making decisions about diagnostic or therapeutic options to establishing a relationship grounded in continuity, realistic expectations, and a shared understanding of goals and values. The second aspect is the changing health care environment and the growing emphasis on managed care systems and competitive market approaches to cost containment. These forces may limit a patient's choice of physicians and may disrupt the continuity of the physician-patient relationship by requiring frequent changes of exclusive provider organizations [21]. In this environment, placing nonabandonment at the center of the essential obligations of physicians seems vitally important. We focus our presentation with two clinical experiences.

Cynthia

Cynthia was a 37-year-old graduate student of the psychology of learning; she was a practicing Buddhist who considered quality, human connection, and spirituality to be central to her life. She developed dyspeptic symptoms, and in 3 days her potential illness escalated from stress-induced gastritis to a potentially treatable gastric lymphoma to metastatic gastric adenocarcinoma.

She was devastated by the diagnosis. When her physicians recommended hospice care and promised to keep her free of pain until her death, she felt abandoned. She found no hope in their offer and needed to find a way to fight her illness. Although she understood the poor odds and potential toxic effects of experimental treatment, she believed she could beat the odds. Accepting her prognosis without a medical fight would have meant giving in to hopelessness and despair.

As Cynthia explored experimental therapy, she needed assurance that she could stop the treatment if it became too harsh or was not working. She also needed to inquire about what dying might be like. She feared severe physical pain, lingering on the verge of death without quality of life, and being kept alive without sufficient consciousness. Because death in her religious tradition was a form of rebirth, she hoped that her physician would help her find death should she enter one of these conditions, which to her seemed more terrifying than death itself.

To a physician (TQ) with a belief in the hospice philosophy and negative feelings about futile, medically invasive treatment of dying patients, Cynthia's request for an aggressive medical approach posed a significant initial challenge. Cynthia needed the slim hope that experimental medical intervention provided, but she also needed honesty about the odds and the potential toxic effects of such treatment. Armed with the reassurance that she would not be abandoned no matter what her clinical course, Cynthia eventually chose experimental therapy, despite its potential risks and burdens.

Not surprisingly, treatment was harsh and ineffective against the relentless progression of the cancer. As the burdens of her illness and its treatment increased, hospice care began to have a different meaning for her. The promise of intensive attention to symptom relief now seemed much more comforting than futilely fighting against her disease, and all efforts were subsequently directed to maximizing the quality of the time that remained. Because the tumor had invaded her entire abdomen, she depended on intravenous fluids and nutrition, which she chose to continue. A continuous intravenous morphine drip helped relieve her pain.

Cynthia went home with the support of her family and a home hospice program. She tried to find meaning in each day and simultaneously prepared for death. She married her long-time boyfriend, and her parents moved to town to be near her. The local Buddhist community regularly had group meditations at her home. She gave away many of her favorite possessions as she prepared for death. It was an intensely sad and unforgettably meaningful time for all who had the privilege to be with her.

After several weeks, Cynthia's quality of life deteriorated despite everyone's best efforts. Her pain required increasing doses of morphine, and she had to make hourly tradeoffs between pain and sedation. She had unrelievable nausea and vomiting, and her wounds and ostomies had an unpleasant smell that she found inescapable and humiliating. She sadly accepted her inevitable death and said goodbye to the important people in her life. At this point, no viable avenues were available to recapture quality in her life, and further disintegration and suffering were larger enemies than death itself.

Because her physician had made a commitment not to abandon Cynthia at this critical moment, the physician's obligation was to help her find death on her own terms as much as possible. For Cynthia, this meant discontinuing central hyperalimentation treatment and accepting the sedation that came with escalating doses of the intravenous narcotics used to control her pain. She was prepared to die but did not want to be perceived as committing suicide. She and her family were reassured that this method was fully compatible with widely accepted ethical principles.

Cynthia gradually became more sedated, went into a coma, and died relatively peacefully 10 days later with her husband, parents, and friends in attendance.

Mrs. K

Mrs. K was a 93-year-old woman who had survived the Nazi death camps where most of her family perished. She had begun a new life in the United States, raised one son in her second marriage, and outlived her husband by 20 years. She lived alone in a small apartment and her social contacts were increasingly limited because of her own physical infirmities and the deaths of many friends.

She was healthy until 8 years ago, when a painful neuropathy developed in her left leg that was eventually diagnosed as reflex sympathetic dystrophy. Seeking pain relief and a way to stem the associated loss of strength, she saw an unending series of consultants. Trials of antidepressants, antiepileptics, nonsteroidal and narcotic analgesics, orthotics, and physical therapy were ineffective and often had unacceptable side effects.

She valued only two medications: flurazepam for inducing sleep and diazepam for treating the "aggravation" of the chronic pain. While seeking a geriatrician for a primary care physician, she was repeatedly told that these were the wrong medications for her because of their long half-lives, potential side effects, and potential for addiction. Mrs. K was determined to express her concerns, attributions, and experience to the physicians she saw; she would dismiss most other medical suggestions as unworkable and then proceed with her litany of problems—pain, sleeplessness, loneliness, and frustration with old age and infirmity. She was strong-willed and opinionated and traveled from physician to physician to everyone's mutual frustration.

Although I (CC) initially repeated this unproductive pattern of intervention and rejection, I eventually realized that what she needed most was to be cared about and listened to. I discovered that she did not expect medical answers to all of her problems. She wanted a confidant or advisor, someone who knew the whole picture, could help interpret what was happening to her, and could find value in her personhood. I began to listen intently to her descriptions of the suffering and frustration of her daily life. I learned about her past suffering and losses in the death camps and about how she survived. I came to understand how lonely it can be to outlive one's family and friends. Through this listening, Mrs. K became more complex to me as a person, which allowed me to empathize more meaningfully with her plight. I helped her with bouts of constipation, guided her through cataract surgery, conservatively evaluated her chest pain of uncertain cause, and allowed her to openly explore whether life was worth living in the face of loneliness and loss. I also re-initiated therapy with flurazepam and diazepam without constantly questioning her need or their efficacy, and I accepted the small risk that she could be secretly stockpiling them in case she decided to end her life.

Although she remains lonely and frightened much of the time, she does not have to face the future alone. When she must confront medical challenges and even death in the future, she will have the distinct advantage of working with a medical guide who knows and respects her.

The Obligation of Nonabandonment

As shown in these two cases, the quality of medical care can be substantially enhanced when the care is provided in the context of a caring, continuous, committed physician-patient relationship. The mutual decision making described can never be adequately covered in formal "practice guidelines," treatment algorithms, or rigidly defined contracts because they cannot possibly include either the distinct and profound specificity of the patient and physician as persons or the shared experiences and meanings that develop between them over time. Medical care can be both humanized and individualized in such relationships. Respect for and curiosity about the person are essential [12, 19], as is the desire to actively involve patients in their own care and empower them as much as possible [18]. Although patients often become more dependent and vulnerable when ill, a caring, committed relationship can respect and explore that vulnerability while simultaneously allowing as much choice and control as possible given the patient's circumstance and personality. Relationships between physicians and patients can be both personal and professional, and empathy [22] and personal connection [20] can enrich the task of facing the reality of the patient's condition together [23-25]. Intuition and emotion supplement the intellect, and there is a flow between sharing information and making decisions about the patient's condition and exploring the associated feelings and reactions.

The obligation of nonabandonment emphasizes the longitudinal nature of a caring and problem-solving commitment between physician and patient [12, 13, 16, 17, 20]. Ethical analyses of clinical actions sometimes focus on one moment in time and seek generalizable rules or answers, but patients and their physicians do not have the luxury of existing in such isolation. Clinical decisions involve a series of choices over time, and the consequences of one decision may immediately lead to new choices.

Furthermore, the meaning and critical nature of any medical act cannot be understood or judged by isolating it into rules that do not acknowledge the personal histories, values, motivations, and intentions of the persons involved [1-9]. For example, it is widely accepted that Cynthia's discontinuation of life-sustaining treatment at the end of her life was morally acceptable even though it resulted in a desired death (she was "allowed to die" of her underlying disease). Easing her death under these circumstances might be viewed as fulfilling a final commitment to not abandon her to further personal disintegration, which she had begun to view as worse than death. However, if that decision stemmed in part from undertreated pain, or from her physician's personal frustration with the difficulty of her dying, then such "allowing to die" could be the worst form of abandonment. Thus, any act cannot be fully understood or judged without understanding the values and intentions of the actors.

The promise to face the future together is a central obligation of the physician-patient relationship [12-20]. The commitment to not abandon is open-ended, for neither person knows what the future will hold nor what might be asked for or required. The American Medical Association (AMA) Council on Ethical and Judicial Affairs have stated a minimal expectation: "Once having undertaken a case, the physician should not neglect the patient, nor withdraw from the case without giving notice to the patient, the relatives or responsible friends sufficiently long in advance of withdrawal to permit another medical attendant to be secured" [26]. The depth and specificity of the obligation may vary from patient to patient depending both on the patient's requests and clinical circumstances and on what the physician can and is willing to commit to. For Cynthia's physician, nonabandonment initially meant promising to work with her in a desperate fight against her disease but later evolved to mean helping her find meaning, choice, and symptom relief as she faced death. For Mrs. K's physician, nonabandonment meant a willingness to share the patient's loneliness and suffering and to accept the limits of medicine while staying alert for those opportunities when medical advice might improve her function, mood, or most recent crisis. For both physicians, it meant being there through the end of life, no matter what the clinical path.

The specific methods that these commitments eventually require cannot be known at the outset—most physicians and patients commit to a process of working together through an unknown future. The physician is committed to responding to the patient's clinical situation and requests in an open-minded way but is not obligated to violate his or her own values and beliefs in the process. When such conflict arises, the physician and patient should make every effort to find common ground and shared meaning. Mrs. K and her physician eventually worked together in a way that allowed her to share her loneliness, report symptoms without having to be constantly tested, and procure an adequate amount of benzodiazepines without having to constantly justify her need. We believe that the meaning and morality of any clinical actions largely depend on the quality of the decision making, the interpersonal relationship, and the shared meaning that they both reflect.

Nonabandonment reinforces several obligations for physicians when they encounter vulnerable persons and populations. At a societal level, physicians as professionals and as persons must help solve the problems of the lack of basic health care services for major disempowered segments of our community and the lack of a coherent primary care system [27-29]. The continuity of care that we are committed to preserving in health care reform has never been widely available to the poor and uninsured. Many such persons are being abandoned to impersonal and episodic care only when their medical or psychosocial problems become so overwhelming that the medical system has no choice but to accept them.

Nonabandonment also reflects an obligation to respond to vulnerable persons whom we contact in our daily clinical work. Just as we must commit to working with the dying or chronically ill for whom cure is impossible, we must learn how to work with persons with overwhelming psychosocial and medical problems. This commitment does not imply an ability to resolve all such problems, but it does require that a physician be willing to care about, advocate for, and ultimately not desert such persons. For many disadvantaged persons, the physician may be the only caring contact and advocate that they have.

Nonabandonment and Traditional Ethical Principles

Traditional ethical analyses often apply the principles of autonomy, beneficence, nonmaleficence, and justice to critical moments in medical decision making [1, 2]. In this section, we explore how the obligation of nonabandonment might be integrated into and help inform these four principles.

The principle of autonomy focuses on patients' rights to have sovereignty over medical matters that pertain to their own bodies. In the United States, patients have a right to consent to or refuse potentially life-sustaining treatment and can have any treatment discontinued if it no longer meets their goals [30-32]. Cynthia initially had a right to request experimental therapy and later to have treatment with intravenous fluids discontinued, but the patient's request of a treatment does not automatically obligate the physician to acquiesce. In fact, passively acceding to a patient request that defies the physician's recommendations and clinical judgement without fully informing the patient and actively exploring alternatives can be a form of abandonment. Furthermore, the language of patient rights suggests that physicians have unrestricted obligations to patients, but physicians also have autonomous values and limitations that need to be respected. Patients will ideally be able to express their requests, hopes, and goals. The physician's responsibility is to seriously consider and try to respond to them without violating his or her own values. The solution to conflict is in part the result of a negotiated agreement between autonomous partners but also a covenantal relationship grounded in mutual respect and caring about the patient's best interests [10, 33, 34]. Autonomy views an individual's self-determination as its highest value and may neglect the central importance of the relationship and interpersonal connection that were at the core of Cynthia's and Mrs. K's treatments.

The principle of beneficence ensures that a physician's actions are intended to benefit the patient; it is the principle most closely related to nonabandonment. Whether a given act is considered beneficent may depend on the nature of the act itself or on its meaning to the participants in the context of their relationship. Considered in the abstract, actively easing the death of a terminally ill patient who requests to die rather than to continue to suffer can either be considered a compassionate act [16] or a fundamental violation of medical ethics [35], depending on one's values and point of reference. Which constitutes abandonment—forcing Cynthia to continue to suffer against her will when a wished-for death is her only relief or intentionally helping her to die? How does her Buddhist philosophy, with its emphasis on quality of life, connection, consciousness, and death as a form of rebirth affect these considerations? How could her physicians have acted beneficently if she had not been receiving a life-sustaining treatment that could be discontinued or if she had not had pain that justified the increasing doses of narcotics? For Mrs. K, who suffers without the potential escape of a terminal illness, openly exploring the personal meaning of choosing death rather than continued living may be central to her search for meaning. Such explorations are essential to good medical care because they often uncover ways in which a physician can help that do not involve actively assisting death [36]. However, we do not yet know what vexing ethical decisions will need to be faced in Mrs. K's future to fulfill the commitment not to abandon her—she only knows that she will not have to face them alone.

The principle of nonmaleficence is captured in the maxim "above all, do no harm." Abandoning patients in severe need would clearly be harmful, but what constitutes abandonment has been debated. The choice that committed physicians and their dying patients face is sometimes between two competing "harms"—is it worse to directly or indirectly assist a suffering patient to achieve a wished-for death, or is it worse to force them to continue to suffer against their will? The "bright line" between "allowing to die" and "causing death" can become indistinct when physicians are faced with real patients who are suffering intolerably and to whom they have made the commitment not to abandon. Such a commitment at least calls for many creative solutions so that physicians can stay within their value structure and still help their patients in need. Two recently proposed solutions are barbiturate sedation to treat uncontrollable suffering [37] and allowing patients to stop eating and drinking to achieve a wished-for death [38, 39]. Although such solutions may stretch the boundaries of rule-based medical ethics, disintegration and humiliation before death require that we consider innovative solutions to fulfill our obligation of nonabandonment [37-41]. Because honesty and truth telling are so critical in these delicate deliberations, the clinician's intentions in these matters should be clarified as much as possible [42].

The principle of justice asks us to look beyond the individual to the aggregate good of society when considering a particular act or policy. The effects of any individual clinical act on the community and the profession must also be considered, even in the face of overwhelming individual suffering. In the health care reform movement, nonabandonment would focus our attention on the 40 million people who are currently uninsured, many of whom live in poverty. How can we justify spending hundreds of thousands of dollars of public funds on individual persons for unproven therapies when we are not providing even basic health care, nutrition, and shelter for major segments of our society? When we consider changing public policies to avoid abandoning the dying, we must consider both the individual persons who would be helped and the potential effect on other vulnerable populations. Perhaps facing death more boldly with our patients will enhance overall medical care at the end of life and add personal meaning to physicians' work, thus reclaiming the moral base of the profession [16, 40]. But it is also possible that such policies would promote our society's penchant for seeking quick fixes for complex moral and medical problems [35, 43]. Nonabandonment would always focus our attention on the most vulnerable individual persons and groups and would encourage health care professionals to struggle on their behalf.

The ethical discourse that emanates from these four principles can be abstract, analytical, and impersonal, with arguments over preset positions and with theory given prominence over the complex stories of actual persons. The obligation of nonabandonment can enrich this analysis by asking us to always consider the particular commitments and circumstances of each patient and physician who are doing their best to make decisions together in the patient's best interests given the uncertainties of the patient's condition. General policies and rules must find room for exceptional persons who are facing unique circumstances.

New Ethical Paradigms

Exciting new theoretical frameworks have emerged in the attempt to remedy the problems created by the too-rigid application of principles. These models are relevant to the issues we have raised under the rubric of nonabandonment. The glimpse at these approaches that we provide in this section does not do justice to their richness and complexity, but the interested reader can begin to explore them further using the references provided. These schools of thought emphasize the importance of the particular persons involved, their relationships with one another, the many levels of meaning and intention in their work together, and the processes of their interaction and decision making. Casuistry, for example, asks us to apply ethical principles to the understanding of particular cases and not always focus on universal or generalizable implications [6]. Analyzing particular cases from the perspective of preexisting positions tends to keep the discussion abstract, superficial, and impersonal. Casuists ask us to grapple in a more personal way with particular cases and in doing so deepen the analysis and make it better reflect clinical realities.

Clinical narratives that reflect the "lived experience" of physicians, patients, and families also help to add depth and feeling to the discussion [44, 45]. Through narratives, we often learn more about the multilayered nature of human life—that intention, emotion, and meaning are textured, contradictory, and much more complex than ordinary rule-based ethics acknowledges. Phenomenologists then remind us that there are no objective observers or reporters of these processes—the best we can do is acknowledge our biases and try to understand and be explicit about how we interpret and translate reality [7, 46]. Phenomenologists warn us against reducing lived experience to simple rules, theories, or moral maxims and exhort us to try to remain true to the particular persons grappling with the dilemma.

Feminist ethics highlights the importance of caring and of relationships—that kindness, mercy, empathy, devotion, altruism, love, and generosity have central roles in medical ethics that are often not adequately reflected in principles [8, 47]. It also underscores the importance of community and warns of the danger of cultural or sexual bias, domination, and subordination that may occur subtly or explicitly. Cross-cultural ethics teaches us that the meaning of an act may vary not only with the persons involved but also with the culture within which the persons operate [4, 5]. The "generalizable" rules derived by a particular ethicist may in fact unconsciously reflect his or her sex, culture, and preexisting biases and experiences; thus, such "objective" analysis and even the principles chosen as fundamental should be open to scrutiny and viewed from a culturally relative perspective.

According to virtue ethics, acts cannot be understood without knowledge of the inner character and motivation of the actors [9]. Like rule-based ethics, it looks to the meaning and intention of the participants. Yet, instead of trying to fit real human intentions into finely drawn, abstract ethical rules, one looks at the motivation, intention, and quality of interaction between the specific clinician and patient. Finally, hermeneutics suggests that all of medical ethics is by its nature interpretative, and therefore each new case or dilemma should be considered independently, keeping in mind the perspectives and values of the particular persons as much as possible [4, 5]. Although generalizable principles and rules may exist, there is always room for exceptional persons and circumstances if we approach each new clinical situation with an open mind and open heart.

Through each of these alternative ethical paradigms runs the assertion that principles must be applied in the context of particular relationships between persons with unique histories and commitments to one another. This context includes the potential closeness between physician and patient, strong feelings of caring and connection, a willingness to take risks on behalf of one another, and a promise not to abandon no matter what the future holds. Seemingly logical ethical distinctions sometimes lose their sharpness in the context of such relationships as commitments and obligations to one another supersede abstract principles. Nonabandonment is a central obligation of these relationships. It evokes less abstract, but no less theory-based, associations than do principles such as autonomy, beneficence, nonmaleficence, and justice.

By re-emphasizing the centrality of the physician-patient relationship to medical ethics, a balanced tension between individual decision making and more general ethical considerations can be made more explicit. For example, after Cynthia's heroic fight against her disease and ultimate acceptance of her death, the obligation to ease Cynthia's death in a manner acceptable to her and her family when her personhood began to disintegrate seemed to override the obligation to make sure that any contribution to her death be inadvertent and unintentional. It is fortunate that Cynthia had an acceptable way to die that was compatible with current ethical and legal mores. We believe that medical ethics excessively emphasizes the method of assistance rather than the quality of caring, commitment, and decision making that led to the act.

Limitations of Nonabandonment

Although we enthusiastically advocate the centrality of nonabandonment as an obligation of physicians, this commitment must be balanced by other ethical considerations [48]. Although physicians should try to respond to the needs and requests of their patients over time, they should not violate their own values in the process. We also encourage a creative tension between respect for the values and choices of unique human beings on the one hand and respect for societal traditions, precedents, and more general implications on the other. Thus, there should always be a dynamic interplay between patient and physician, individuals and society, traditional and personal values, subjective interpretations and objective analysis, emotion and intellect. Grappling with these inherent tensions should add depth and complexity to our moral decision making and thus more accurately reflect the reality of the human condition.

Nonabandonment as an obligation also has inherent limitations. Beyond a basic notion of ensuring some form of continuity as suggested by the AMA, the depth and nature of this commitment may vary for both physician and patient. The relationship is partly defined and explicit, but partly open-ended and implicit, more of a covenant than a contract [10, 33, 34]. Although we have focused primarily on the obligations of physicians, these relationships are ultimately reciprocal if not equal or symmetrical. Often, the rewards for physicians who make these commitments far exceed what is required of them, but the extent of caring and the level of personal responsibility will ultimately depend on mutual give and take over time. Sometimes the pace of coming to know one another is accelerated by a severe illness. At other times, it is limited by either party's reluctance or inability to trust or to address major medical issues (for example, substance abuse or dementia). Through shared experience over time, a comfortable and effective level of interaction that ranges from superficial to intense can usually be established with the patient or his or her surrogates. Although many physicians choose the profession of medicine because of their need to serve and care, physicians themselves must also lead balanced, healthy lives [49]. Many physicians are more skilled at recognizing and responding to the needs of others than to their own needs. If we ask physicians to provide a more caring, long-term commitment to their patients, we must also reinforce their need to set limits and take care of themselves. These limits might be expressed by limiting practice size, encouraging group practice for the support and the potential of shared coverage, and reinforcing the possibility of limited relationships and even explicit termination with some patients with whom a mutually satisfactory relationship cannot be established. There is sometimes an irresolvable tension in medicine between the commitment to the care of others and self-care.

We must also try to ensure that our ethical precepts and our health care system reinforce rather than undermine physicians' willingness to engage with patients when their problems seem insoluble or in situations in which the problems are not clearly resolved by current ethical thinking. The principle of nonabandonment is paramount and may allow physicians to take some risks on behalf of patients who have no good alternatives. However, the obligation not to abandon does not mean that physicians should violate their own moral values. These values should be explored and challenged but ultimately respected. Physicians should make every effort to find common ground with suffering patients without violating their own fundamental values. Such analysis often requires intense self-examination and consultation with trusted colleagues. We must learn to both challenge and respect our limitations.

Conclusions

Nonabandonment is fundamental to the long-term physician-patient relationship. It reinforces a longitudinal commitment that is an essential aspect of being a physician and is too often inadequately captured in ethical analyses. Patients seek physicians who will make a commitment to both care about and know them as persons and to be guides and partners in sickness and health until their deaths. In this context, physicians and patients can learn to judiciously use the power of medicine and expand the concept of healing to include working with persons with severe chronic illness and disability and those who are dying. The commitment not to abandon supplements the caring relationship because it requires that the physician and patient work together over time, even when the path is unclear. Moral challenges must be met and engaged in with the patient, not shied away from by recourse to falsely bright lines or unbending rules. There is a world of difference between facing an uncertain future alone and having a caring partner who will be present no matter what happens.

We believe that such commitments between physicians and patients are at the core of the medical profession and that they must be explicitly represented in the discourse of medical ethics. Physicians who find that their work with patients has lost its excitement and meaning would do well to consider whether they are engaged in these types of relationships. Health planners, legislators, risk managers, medical educators, and ethicists should carefully examine whether their contribution to health care tends to reinforce or obstruct such commitments. Clinical medicine is ultimately a humbling and exhilarating profession, filled with joy, sorrow, and an overabundance of uncertainty that comes with establishing a genuine long-term connection with patients. To practice medicine with a commitment to caring and to be there no matter what the future holds is to experience the richness of the human condition over and over again and to know one has made a difference. If the obligation of nonabandonment is better incorporated into medical ethics, medicine may become more humanized and more responsive to the real problems faced every day by physicians and patients.