Forsaking the Spirit for the Letter of Law: Advance Directives in Nursing Homes

It has been 5 years since passage of the federal Patient Self-Determination Act (PSDA) requiring hospitals, nursing homes and other healthcare institutions to advise patients of the right to execute advance directives and to assist in their completion. The goals of the PSDA were straightforward and widely acclaimed. It would encourage patients to specify their wishes regarding life-sustaining treatment in the event of future incompetence and to designate proxy decision-makers to act on their behalf. At the same time, increased use of advance directives would alleviate complexities and controversies in making decisions about life-prolonging treatment for incompetent patients whose wishes are not known. Thus, the PSDA would extend the autonomous control of patients over end-of-life decisions while diminishing the need for problematic judgments about the best interests of incompetent patients. In this issue of the Journal, Mezey and colleagues1 present the results of a valuable survey of social workers in New York City nursing homes regarding implementation of the PSDA. Some findings provide encouragement that the goals of the PSDA are being fostered. For example, social workers at 97% of the nursing homes reported that they had “face-to-face”discussions with new residents regarding their interest in executing advance directives. In addition, the social workers estimated that 24% of nursing home residents had executed proxy directives and that 20% had do-not-resuscitate orders. Social workers also indicated that 70% of the homes had policies “to honor agents' decisions to terminate treatment. “Finally, it was reported that 89% of nursing homes routinely send advance directives and DNR orders when residents are transferred to other institutions. Other results are less encouraging. In almost one-quarter of the nursing homes, less than 20% of the residents had executed advance directives. There appeared to be no uniform procedures for placing directives in the medical records of residents or for their timely review. Moreover, one-third of proprietary homes had no established policy to honor the verbal directives of residents. Much more disconcerting are the implications of this study regarding the bureaucratic procedures that have emerged for implementing the PSDA. These procedures may impair the quality of advance decision-making and limit the extent to which patients' goals and values are reflected in end-of-life decisions. First, the study clearly suggests that primary caregivers — the nurses and physicians who provide medical care for residents — do not play a prominent role in the process. Rather, social workers have assumed responsibility for advising patients of their rights to execute advance directives and for documenting their wishes. This fact raises concerns that patients' wishes will not be interpreted reliably or implemented faithfully at the bedside. If patients are not encouraged to discuss advance planning with primary caregivers, their values and goals related to life-sustaining treatment may not inform the interpretation of their wishes when clinical decisions must be made. In addition, when primary caregivers do not play an essential role in advance planning, they may not be attentive to the documented wishes of the patient or fully embrace the commitment to implement them. Indeed, numerous studies have indicated that physicians and nurses understand poorly the advance preferences of patients and are very often unaware that they have advance directives at all. See also p 43 Second, the principal role of social workers in advising nursing home residents about advance directives raises legitimate concerns about whether prospective decision-making is adequately informed. This problem is concealed somewhat in the results of the study because social workers placed emphasis on securing proxy rather than instruction directives. In making decisions about appropriate proxies, patients do not need medical information. By contrast, instruction directives require decisions about the forms of life-sustaining treatment to be withheld in the event of future incompetence. Standards of informed consent disclosure require that patients receive those items of information that a reasonable person would need to know in deciding about these interventions. Unfortunately, social workers may not possess sufficient medical knowledge to effectively elucidate the risks, benefits, and probable outcomes of life-sustaining procedures, such as tube feeding and cardiopulmonary resuscitation. Uninformed decisions by nursing home residents regarding life-sustaining treatment are not likely to reflect their individual values and goals accurately. For example, several studies have shown that uninformed patients greatly overestimate the potential efficacy of cardiopulmonary resuscitation. When accurate outcome data are supplied, the percentage of patients requesting the intervention declines precipitously. A third concern suggested by this study is whether the advance directive options presented to patients provide adequate scope for the expression of their wishes. Caregivers are most likely to suggest advance directives whose content is formulated in state statutes. For example, Mezey et al. found that social workers emphasized the execution of proxy directives in discussions with nursing home residents, partly because these are the only form recognized statutorily in New York State. Unfortunately, the emphasis on proxy directives may discourage many persons from providing explicit instructions regarding their preferences for life-sustaining treatment. Even in states with statutes recognizing instruction directives or living wills, a standard content for these directives is provided by law. When advance planning is discussed with patients, these standard living wills are normally provided for their review. However, there are important aspects of patients' values and goals regarding life-sustaining treatment that are not addressed in these documents. One important shortcoming is that they apply only to circumstances in which patients are permanently comatose or will soon die no matter what medical interventions are employed. With respect to patients in the latter category, living wills might be said to belabor the obvious regarding the sensibleness of further life-sustaining treatment. More significantly, they do not permit instructions about circumstances (other than coma) in which patients are not dying but suffer from severe, irreversible cognitive and physical disabilities. For example, many older persons suffer from syndromes involving progressive dementia. Issues about the use of life-sustaining treatment, such as tube feeding, often arise before these patients are terminally ill. Standard living wills do not allow older persons to indicate their values and goals regarding life-sustaining treatment administered under these circumstances. Another restriction imposed by statutory living wills is that they do not recognize non-medical factors as legitimate considerations in the choices of patients regarding life-sustaining treatment. However, many studies have demonstrated that older adults place heavy emphasis on familial considerations in advance planning. For example, some patients do not want life-sustaining treatment continued if they become a severe physical or emotional burden to family caregivers. Many individuals do not want their lives prolonged if it will seriously erode the family's financial resources. When options for advance planning are restricted to statutory mechanisms, these values and goals of older persons may not receive appropriate attention. These considerations underscore the need for thorough discussion with nursing home residents regarding the goals and values that inform their preferences. Moreover, these informal verbal exchanges should be documented properly in their medical records. In this regard, it is notable that Mezey and her associates found an ambivalent strain in nursing home policies about honoring the verbal directives of residents. Nearly one-in-five voluntary homes and one-in-three proprietary homes do not have policies to honor these directives. Even in homes that acknowledge their validity, residents may not be encouraged to express their preferences beyond the scope of statutory documents. A fourth problem intimated in the findings of Mezey and her colleagues concerns the apparent failure to routinely include potential proxy decision-makers in conversations with nursing home residents about their values and goals related to life-sustaining treatment. There is no evidence from the study that social workers implementing the PSDA usually encourage the involvement of potential proxies. One reason for this lacuna may be that social workers emphasized the preparation of directives for appointing the proxies themselves. Nevertheless, the result is troubling. Several studies suggest that family members, much like physicians and nurses, predict the preferences of patients regarding life-sustaining treatment with a frequency that is only slightly better than chance. The accuracy of their estimates is not improved even when they express greater confidence that they know the patient's wishes. These findings engender concern about proxy judgments that are formulated without the benefit of candid discussions with patients about their goals and values. On the other hand, available studies have not explored the concordance of patients' wishes and proxy decisions after both parties have received the same medical information about life-sustaining treatment and have had the opportunity for frank discussion of the values and goals of the patient. It is reasonable to expect that such discussions could improve substantially the ability of proxies to implement accurately the wishes of patients. An additional problem in advance planning practices is more conspicuous in the results of the survey. In responding to a question regarding nursing home policies on reviewing advance directives, social workers indicated that in 23% of the homes, advance directives are reviewed only annually, and in 26% they are never reviewed or the review process is not known. Because existing directives are reviewed no more than annually in half of the nursing homes surveyed, it is probably even less common to revisit the issue with residents who do not execute directives at the time of their admission. Moreover, it is unclear who is assigned the responsibility for reviewing advance directives and whether primary caregivers are apprised of the results of such reviews. These findings suggest a process of executing and updating advance directives that is significantly disjoined from the daily provision of medical care to nursing home residents. Residents' control over end-of-life decisions might be promoted more effectively if advance planning were as much a routine part of medical management as reviewing their medications. This requires that primary caregivers familiarize themselves with the advance directives of patients regularly as an essential part of the overall care plan. Furthermore, it requires that as the medical condition of each resident undergoes significant alteration, primary caregivers should review with each patient his or her previously expressed wishes. Many older persons suffer from chronic medical conditions that worsen progressively over a long period. Their experiences with medical interventions, the burden of suffering incurred, their perception of the impact on family, and other factors may revise and refine substantially their views of how they want to be treated should they become incompetent. Several studies have revealed, for example, that a large percentage of adults take advance planning seriously only when acute medical crises occur or when a fatal condition is diagnosed. The appropriate occasions for reconsideration of their wishes are set by the progression of their illnesses and the course of their personal reflections rather than a schedule for the review of documents. When attention to the advance planning of patients is a regular part of their medical management, these issues can be addressed in a more timely and sensitive manner. A final caution relates to the portability of advance directives when nursing home residents are transferred between health care facilities. Mezey and associates found that 89% of New York City nursing homes send advance directives and DNR orders when residents are transferred. This is a positive finding, although it must be wondered why 11% of nursing homes would fail to transfer these documents. However, reservations exist about whether the simple transfer of documents will assure that the advance decisions of older patients are properly acknowledged by caregivers at the receiving institution, usually a hospital. When acutely ill older patients arrive at the hospital, the primary focus is usually placed on stabilizing their immediate medical condition. It is easy to avoid asking how acute care interventions may fit into the previously expressed preferences of the patient. Personal communication with the hospital-based physician by the primary caregiver at the nursing home may provide stronger assurance that treatment will not contravene the previously expressed wishes of the patient. More importantly, when nursing homes are able to provide a level of care consonant with the values and goals of the patient, transfer to the hospital is inappropriate. Because residents who are transferred to acute care facilities are more likely to receive aggressive life-prolonging treatment, nursing home policies that require transfer in acute medical crises may diminish substantially the likelihood that advance directives will be honored. In short, the study by Mezey and colleagues provides an excellent picture of the procedures utilized in implementing the PSDA in nursing homes. Advance planning options are discussed by social workers with new residents. Standard statutory directives are completed, entered in the medical record, and, perhaps, reviewed periodically. The documents are usually transmitted with residents to other institutions. In these respects, the implementation of the PSDA has generally satisfied the letter of the law. On the other hand, grave doubts are engendered that implementation of the law has served as a catalyst for a richer collaboration of physicians, patients, and families within the therapeutic relationship. In order to achieve this goal, primary caregivers must accept advance planning as an integral element in fashioning end-of-life care that comports with the values and goals of patients. They must assure that advance directives are informed by the medical facts about life-prolonging treatment. Informal discussion of the goals and values of patients related to end-of-life care must not be restricted to the narrow focus of standard legal directives. Patients should be encouraged to confer with families in refining and clarifying their views. Review of advance directives should be a regular part of caregiving, especially at the time of significant changes in medical condition. Finally, primary caregivers must discourage transfers to hospitals when life-sustaining interventions may contravene the preferences of residents. These collaborative efforts within the therapeutic relationship are essential if the values and goals of patients are to be reflected accurately in treatment decisions when they become incompetent. Otherwise, the spirit of the PSDA will remain unfulfilled.