Printer Friendly

After the Patient Self-Determination Act: the need for empirical research on formal advance directives.

In the past decade, formal advance directives have come to be viewed by nearly all parties as having the potential to be a very important component of medical decisionmaking. For the past year, they have been the main object of a federal initiative under the Patient Self-Determination Act, intended to encourage patients to claim their rights in regard to decisionmaking. The PSDA requires that notice of the possibility of formulating advance directives be given to nearly all patients in the health care system; it also carries substantial requirements for documentation, transfer, clarification of the law, and public education.

While theoretical justifications for use of advance directives are easy to marshall,[1] to date little empirical research has been done to assess the merits of advance directives in practice.[2] For all we know, advance directives could be costly, irrelevant, or harmful. If so, encouraging their use would not be good policy. Here we will outline the issues needing to be addressed about formal advance directives and give an overview of what is now known. We will not assess what we could and should know about informal advance directives (oral communications, letters, medical records, etc.). They are both more difficult to measure and have been even less often studied. Nonetheless, we would note that these informal advance directives are probably more important to clinical practice than are formal advance directives.

The Rationale for Advance Directives

The fundamental claim of mainstream American medical ethics today is, roughly, this: the decisions made in the care of a patient should be those that are expected to deliver the best possible outcome for that patient, as judged by the patient.[3] Although beyond the scope of this article, the merits of this decisionmaking model also could be studied empirically, although virtually no such research has been done.

However, when the patient cannot make his or her own assessment of alternative outcomes it is very difficult to make decisions that reflect patient preferences. Especially since illness is now so commonly chronic and incompetence is so often the result of long-term illness in old age, clinicians can anticipate incompetence and enable patients to direct in advance the kinds of treatments and/or outcomes they would prefer. This provides a way to have the patient's nearly absolute authority endure into a period of incompetence and thereby greatly simplifies decisionmaking, both by making it more expeditious and by making it more likely to comport with the patient's own preferences. Having the patient make decisions in advance of incompetence allows others to avoid the often quite difficult and uncertain task of making the choices and provides a way for the patient to continue to bear the responsibility for decisions. Thus, various formal measures have been implemented to effectuate this endeavor. The most common are "living wills" and "durable powers of attorney."

These fundamental claims were at the heart of the Patient Self-Determination Act. However, good policy must depend upon effective implementation at least as much as sound argument. How effective has the implementation of advance directives been?

Prevalence: Awareness and Completion of Advance Directives

Most people now seem to have at least a passing acquaintance with formal advance directives.[4] Support groups for persons with AIDS appear to have been singularly effective in assuring that this population is well informed.

Descriptive research to date has shown that advance directives enjoy widespread approval but that the rate at which they are actually written is much lower and varies with the population. In 1988, a public opinion survey conducted by the American Medical Association found that 56 percent of the general population had discussed with family their treatment preferences if they were in a coma. However, only 15 percent had a living will.[5] Emanuel and colleagues found that greater than 90 percent of respondents had a positive attitude toward advance directives but fewer than one in ten had completed one.[6] In contrast, about two in ten persons with AIDS treated in established community settings have a written directive.[7] In SUPPORT (The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments), 21 percent of a population of very seriously ill hospitalized patients has been reported to have advance directives, with 7 percent living wills only, 9 percent durable powers of attorney only, and 5 percent having both.[8]

Characteristics associated with completing advance directives are known only for select populations with limited generalizability. Among persons with AIDS, white or well-educated respondents were more likely to have written a directive.[9] An association of ethnicity and intent to complete an advance directive has been reported in a population admitted to one Veteran's Administration hospital, with 91 percent of whites and 66 percent of blacks expressing an intent to complete an advance directive.[10]

One persistent concern about use of advance directives has been whether their meaning and usefulness might be linked to a rational, "middle-class" approach to health care. The very meaning of offering an advance directive might be ineluctably altered when the patient is too poor to demand health care at all.

Few studies have examined whether the rate of advance directive use in practice can be improved. Among ambulatory patients in a residency program at a tertiary care hospital, an intervention consisting of an educational booklet and physician-initiated discussion resulted in only eight of fifty-two patients writing an advance directive.[11] In a randomized controlled trial of an educational intervention, Greg Sachs and colleagues found that a majority of patients (85%) did not implement a living will.[12] In contrast, Lawrence Markson and Knight Steel reported that forty-eight out of seventy-four homebound patients who were counseled about a durable power of attorney completed one.[13] The potential for increasing the rate of use of directives through an educational intervention is, therefore, uncertain.

Patients' Intent

We do not know the intent of patients who chose to complete a formal directive, or, for that matter, why some do not. Clearly, at least prior to the PSDA, advance directives have not figured centrally in patients' and physicians' interactions. The directives that have been reported to exist in the community at present do not seem to be based on discussion with a physician. In the AIDS population reported above, only one-half of persons with an advance directive discussed it with a physician, even after surviving a year of treatment.[14] Much anecdotal evidence favors the claim that physician involvement is uncommon.

In SUPPORT, we found that of the 735 patients who were reported (by self or surrogate) to have an advance directive, its existence was mentioned in the physician or nurse progress notes in only 45 cases. In this study, patient or surrogate preferences against resuscitation did not have a substantial association with having a living will. This suggests that patients may intend more to avoid loss of control while dying and incompetent, rather than to shape a palliative plan of care throughout the course of their terminal illness.

Since treatment directives are commonly vague and difficult to interpret, it is important to know what the patient understood and what he or she was hoping to accomplish by completing the directive. However, no study yet reported has asked patients what they thought they were doing and how confident they were that they had succeeded. Likewise, no study yet has examined a full range of possible incentives, including the obvious possibility that the patient is heavily motivated by a fear of loss of control, or by a concern for the well-being of family.

Patients might execute advance directives to keep decisionmaking out of the courts--avoiding certain kinds of judicial involvement seems likely with some advance directives. However, once there is a legal document, the opportunities for legal wrangling may escalate rather than decline. No study has yet addressed the actual effects. This may be especially important for instruction directives, which are often too vague for ready application or are clouded by uncertainty about patient understanding at the time the directive was constructed.

A flurry of articles has documented that surrogates (next of kin, usually) err substantially in predicting patient preferences.[15] This is often taken to be a strong argument for encouraging advance directives, especially treatment directives, since the major alternative strategy of relying on a (family) surrogate seems so flawed. Not only is this problematic in that patients might have a stronger interest in who makes decisions than in having a particular choice made,[16] but the finding of discordance is also inconclusive since no one has yet studied the efficacy of advance directives in evading this problem.

Do the preferences articulated during a time of health and prior to a pressing need for a decision endure into a period marked by serious illness and imminence of the impact of the choice? The few research endeavors that have examined the stability of preferences have shown only moderate agreement over time, with a clinically significant number of people changing their stated preferences (in both directions).[17] To date, this research has examined patient responses to a standardized interview, which is a limited method but still leads to the important question, What are we to do if competent patients often disagree later with what they said earlier in a formal advance directive?

Impact of Advance Directives on Medical Decisionmaking

Formal advance directives are not worth much unless they can be shown to improve decisionmaking. To date, only a few research projects have generated relevant insights, and none have done so definitively. Physicians who had used written directives felt that they were helpful.[18] However, in-depth interviews with fifty-seven physicians from California and Vermont yielded little indication that advance directives were helpful in medical decisionmaking.[19] Marion Danis and colleagues noted that one in four directives were not honored by the attending physicians.[20] This study did not address whether these failures to follow a directive were justifiable or not.

Only two studies have addressed the fundamental question of whether care differs between patients with and without advance directives. An abstract from the SUPPORT project indicates that seriously ill persons with and without advance directives tend to get about the same care patterns, even if the advance directives are treatment directives.[21] In a randomized trial, Lawrence Schneiderman and colleagues assigned subjects to be offered the opportunity to complete written directives or to receive routine care.[22] This intervention did not affect subsequent resource utilization prior to death. Nor did it have an impact upon DNR rates, CPR, survival, or well-being. However, the patients were not very ill (although they had long-term, life-threatening illnesses) and did not often have cognitive deficits as they died, and the sample was small (183 patients with 100 deaths during at least thirty-four months of follow-up).

In both of these studies, subjects had a treatment and/or a proxy directive. It will take some focused attention on the matter to discern whether advance directives that designate proxies are followed and whether the same proxy would have been identified without the directive. Likewise, the impact of instruction directives will have to be elucidated much more clearly. If they turn out to have little or no effect, then one is obliged to examine whether that is because very good decisionmaking practices apply in the case of the patients without formal advance directives, so that the differences are slight, or whether the formal advance directives are too easily ignored, even though they would have stipulated substantial differences in care. Of course, for the latter to be the case, there must be a substantial rate of incompetence near death.

In doing such research, certain methodological issues are recurrent and need careful consideration. For example, it turns out to be quite difficult to define the necessary components of a formal advance directive. Must it have met the witnessing and registration requirements of the state? Must it have been shared with some health care provider? If it is a surrogate designation, must it name someone who is actually available and appropriate? If an instruction directive, is it enough to state general sentiments about the appropriateness of dying, or must it also say something more directly applicable, such as addressing the acceptance of particular treatments or outcomes? Should an advance directive have substantial force if it is likely to be based on serious misunderstandings by the patient about his or her situation and its likely outcomes? Probably most important, is a written document ever as important as comprehensive communication about preferences and likely outcomes among patient, providers, and potential surrogates? These last policy issues would be decided much more wisely if there were more understanding of the effects of enforcing even directives that are acknowledged to be deficient as compared with the effects of alternative strategies for decisionmaking.

Whatever was the practice before the Patient Self-Determination Act may well have changed substantially since, and those changes will need to be described. In addition, other initiatives might well have promise in encouraging optimization of decisionmaking through more extensive use of advance directives. The burgeoning array of efforts to enhance use of advance directives includes values history forms,[23] better formal advance directive forms, consumer education material (including the descriptions of patients' rights under state law that are required by the PSDA), videos, interactive videodiscs, and skilled legal counseling (mostly during estate planning). The merits of these are unknown without careful assessment. Some are probably helpful. Some are probably unreasonably costly or actually misleading. Program evaluation and demonstration projects are urgently needed.

One must eventually define the outcomes of interest in assessing the merits of advance directives. It cannot be adequate to document the frequency of directives, or to show that an intervention can increase the rate of completion. One must show that decisionmaking is improved by the presence of formal advance directives, or that patient outcomes are improved. Specifying the measures for either of these is quite difficult. Patient outcomes might include survival, function, symptoms, satisfaction, and global quality of life.[24] Perhaps it is also an "outcome" to have had the treatment course that one preferred. How to measure each dimension and how to combine them is largely unresolved. Nevertheless, patient outcomes do matter, and we must develop a serviceable metric and accomplish the research necessary to assess whether advance directives improve outcomes and, if so, in which forms and which populations.

Asking the Right Questions

Patients face decisions among potential courses of care; for reasons as diverse as caregivers' desire to leave the patient responsible for the outcome and society's investment in the moral value of letting people make their own decisions, we want to have the patient's preferences dictate the choice. When the patient is adult but incompetent, our community value of honoring self-determination becomes hard to implement, as there is no process that ensures decisions are made so as to honor the patient's preferences. Making decisions in advance of incompetence seems to reconstitute the competent decisionmaker and to evade the obvious ambiguities and compromises inherent in allowing decisions to be made by physicians and families. What is not clear is that the solution works or, more precisely, that it works better than alternative strategies. Nor is it clear that we have a common conception of what it would mean to "work"; that is, What is the standard to which one would compare any policy choice?

There are a few fundamental issues about advance directives in practice that must be researched if we are to be at all confident about the appropriate role for them. First, research must elucidate the degree to which patients' preferences are stable and represent important beliefs. If patients' preferences are quite unstable, then carrying them forward into incompetence is unjustified. Likewise, some preferences might be strongly held, so that thwarting them wrongs the patient, while others are lightly held, and thwarting them is barely perceived by the patient. Such considerations must be incorporated into the implementation and evaluation of advance directives. Second, the effects of advance directives under various conditions need to be compared with various alternative strategies to guide decisionmaking for adults with incompetence. Otherwise, we will not be able to assess whether the policy of encouraging advance directives is, on the whole, an improvement over alternatives. Third, we need to understand why people do and do not write advance directives, especially the barriers to writing one when the concept is understood and is relevant to the patient. If many people do not write advance directives, for example because they do not want to have responsibility for these choices, that rationale should reshape the policy and practice.

In sum, advance directives have been proposed as the answer to the problem of how to empower patients so that they maintain control of their care even when incompetent. We have not yet shown that directives will answer that need. Indeed, the question may well be better articulated as one of defining and implementing an optimal procedure to make care decisions on behalf of incompetent adults. We certainly do not know how well advance directives answer that need, especially in comparison to alternative strategies. We may have the wrong answer; in fact, we may have answered the wrong question. We must seek to do the research on these issues before substantial social resources are committed to any particular strategy.

[Joanne Lynn is professor of medicine and of community & family medicine, Center for Evaluative Clinical Sciences, Dartmouth Medical School, Hanover, N.H.; Joan M. Teno is assistan professor of community & family medicine, Center for the Evaluative Clinical Sciences, Dartmouth Medical School.]


[1.] Allen E. Buchanan and Dan Brock, Deciding for Others: The Ethics of Surrogate Decision Making (Cambridge: Cambridge University Press, 1989); The Hastings Center, Guidelines on the Termination of Life-Sustaining Treatment and the Care of the Dying (Bloomington: Indiana University Press, 1987); The Appleton International Conference, "Developing Guidelines for Decisions to Forgo Life-Prolonging Medical Treatment," supplement, Journal of Medical Ethics 18 (September 1992): 1-23.

[2.] Kent W. Davidson, Chris Hackler, Debra R. Caradine, and Ronald S. McCord, "Physicians' Attitudes on Advance Directives," JAMA 262 (1989): 2415-19; Marion Danis, Leslie I. Southerland, Joanne M. Garret, et al., "A Prospective Study of Advance Directives for Life-Sustaining Care," NEJM 324 (1991): 882-87; Joel M. Zimberg, "Decisions for the Dying: An Empirical Study of Physicians' Responses to Advance Directives," Vermont Law Review 13 (1989): 445-49; Joan M. Teno, Joanne Lynn, Donald J. Murphy, et al., "Impact of Advance Directives on Decisionmaking," Gerontologist 31 (1991): 41; Lawrence J. Schneiderman, Richard Kronick, Robert M. Kaplan, et al., "Effects of Offering Advance Directives on Medical Treatments and Costs," Annals of Internal Medicine 117 (1992): 599-606.

[3.] Buchanan and Brock, Deciding for Others; President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, Deciding to Forego Life-Sustaining Treatment (Washington, D.C.: U.S. Government Printing Office, 1983); Joanne Lynn and David DeGrazia, "An Outcomes Model of Medical Decision-Making," Theoretical Medicine 12 (1991): 325-43; Cruzan v. Director, Missouri Department of Health, 110 S. Ct. 2841 (1990).

[4.] American Medical Association, "Physician and Public Attitudes on Health Care Issues"; Linda L. Emanuel, Michelle Barry, John D. Stoeckle, et al., "Advance Directives for Medical Care: A Case for Greater Use," NEJM 324 (1991): 889-95; Joan M. Teno, John Fleishman, Dan Brock, and Vincent Mor, "The Use of Formal Prior Directives among Patients with HIV-Related Diseases," Journal of General Medicine 5 (1990): 490-94; Elizabeth R. Gamble, Penelope J. McDonald, and Peter R. Lichstein, "Knowledge, Attitudes and Behavior of Elderly Persons Regarding Living Wills," Archives of Internal Medicine 151 (1991): 277-80; Jiska Cohen-Mansfield, Beth A. Rabinovich, Steven Lipson, et al., "The Decision to Execute a Durable Power of Attorney for Health Care and Preferences Regarding Utilization of Life Sustaining Treatments in Nursing Home Residents," Archives of Internal Medicine 151 (1991): 289-94; Robert Steinbrock, Bernard Lo, Jeffrey Moulton, et al., "Preferences of Homosexual Men with AIDS for Life-Sustaining Treatment," NEJM 314 (1986): 457-60.

[5.] American Medical Association, "Physician and Public Attitudes."

[6.] Emanuel et al., "Advance Directives for Health Care."

[7.] Teno et al., "The Use of Formal Prior Directives."

[8.] Teno et al., "Impact of Advance Directives."

[9.] Teno et al., "The Use of Formal Prior Directives."

[10.] Jeremy Sugarman, Morris Weinberger, and Greg Samsa, "Factors Associated with Veterans' Decisions about Living Wills," Archives of Internal Medicine 152 (1992): 343-47.

[11.] Jan Hare and Carrie Nelson, "Will Outpatients Complete Living Wills? A Comparison of the Two Interventions," Journal of General Internal Medicine 64 (1991): 41-46.

[12.] Greg A. Sachs, Carol E. Stocking, and Steven H. Miles, "Failure of an Intervention to Promote Discussion of Advance Directives," Journal of the American Geriatrics Society 40, no. 3 (1992): 269-73, at 269.

[13.] Lawrence Markson and Knight Steel, "Using Advance Directives in the Home-Care Setting: A Pilot Project," supplement, Generations 14 (1990): 25-29.

[14.] Joan M. Teno, Vincent Mor and John Fleishman, "Communications Regarding Formal Advance Directives," Clinical Research 39, no. 2 (1991): 532a.

[15.] Jan Hare, Clara Pratt, and Carrie Nelson, "Agreement between Patients and Their Self-Selected Surrogates on Difficult Medical Decisions," Archives of Internal Medicine 152 (1992): 1049-54; N. Zweibel and Christine Cassel, "Treatment Choices at the End of Life: A Comparison of Decisions by Older Patients and Their Physician-Selected Proxies," Gerontologist 29 (1989): 622-26; Joseph G. Outslander, Alexander J. Tymchuk, and Biba Rahbar, "Health Care Decisions among Elderly Long-Term Care Residents and Their Potential Proxies," Archives of Internal Medicine 149 (1989): 1367-72; Richard F. Uhlman, Robert A. Pearlman, and Kevin C. Cain, "Physicians' and Spouses' Predictions of Elderly Patients' Resuscitation Preferences," Journal of Gerontology 43 (1988): 115-21; Allison Seckler, Diana Meir, Michael Mulvihill, and Barbara Cammer, "Substituted Judgment: How Accurate Are Proxy Predictions?" Annals of Internal Medicine 115 (1991): 92-98.

[16.] Joanne Lynn, "Why I Don't Have a Living Will," Law, Medicine & Health Care 19 (1992): 101-4.

[17.] Maria A. Everhart and Robert A. Pearlman, "Stability of Patient Preferences Regarding Life-Sustaining Treatments," Chest 97 (1990): 159-64; Mark D. Silverstein, Carol B. Stocking, Jack P. Antel, et al., "Amyotrophic Lateral Sclerosis and Life-Sustaining Therapy: Patients' Desires for Information, Participation in Decision Making, and Life-Sustaining Therapy," Mayo Clinic Proceedings 66 (1991): 906-13; Joan M. Teno, John Fleishman, Dan W. Brock, and Vincent Mor, "Stability of Preferences among Patients with HIV-Related Illnesses," unpublished.

[18.] Davidson et al., "Physicians' Attitudes."

[19.] Zimberg, "Decisions for the Dying."

[20.] Danis et al., "A Prospective Study of Advance Directives on Decision-Making."

[21.] Teno et al., "Impact of Advance Directives on Decision-Making."

[22.] Schneiderman et al., "Effects of Offering Advance Directives on Medical Treatments and Costs."

[23.] Joan Gibson, "National Values History Project," supplement, Generations (1990): 51-64

[24.] Joanne Lynn and William A. Knaus, "Background for Support," supplement, Journal of Clinical Epidemiology 43 (1990): 1S-5S.
COPYRIGHT 1993 Hastings Center
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Giving Life to Patient Self-Determination
Author:Lynn, Joanne; Teno, Joan M.
Publication:The Hastings Center Report
Date:Jan 1, 1993
Previous Article:The Patient Self-Determination Act: an early look at implementation.
Next Article:Bioethics education: expanding the circle of participants.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters