Log in

BY: MARY BETH FOGLIA, RN, MN, MA, and ROBERT A. PEARLMAN, MD, MPH
Ms. Foglia, a doctoral candidate, is senior ethics consultant and ethics evaluation specialist, Ethics Evaluation Service, National Center for Ethics in Health Care, Veterans Health Administration, Seattle. Dr. Pearlman is professor, Department of Medicine, University of Washington, and director, Ethics Program, Veterans Affairs Puget Sound Health Care System, both in Seattle; and chief, Ethics Evaluation Service, National Center for Ethics in Health Care, Veterans Health Administration, Washington, DC.

In this article, we hope to illustrate important limitations in the way ethics consultants tend to frame and respond to ethics concerns in clinical care. In addition, we hope to demonstrate that a traditional, case-based approach to framing ethics issues may result in solutions that solve the immediate case but do little to prevent a recurrence of the same problem. The problem recurs because the underlying systems and processes that contribute causally to the problem remain unchanged.

Unfortunately, many present-day ethics programs operate as "silos" — drawing a bright line between the concerns of clinical ethics and those of organizational ethics. An integrated model, first proposed by C. R. Seeley and S. L. Goldberger in 1999, rejects this distinction, arguing that differentiating the concerns of clinical ethics from those of organizational ethics is like differentiating those between psyche and soma, a false dichotomy.¹

Clinical cases have organizational antecedents and sequelae, and vice versa. Hence, both macro and micro lenses must be brought to bear on ethics concerns if we wish to fully appreciate the problem and recommend solutions that prevent it from recurring.

The Autopsy Case
The case described below is based on an actual incident. However, certain aspects have been fictionalized, both to protect the privacy and confidentiality of those involved and to illustrate the "take home" messages.

A patient with a terminal illness repeatedly stated to his physicians and nurses that he did not want to be autopsied following his death. A postgraduate medical trainee and another staff member communicated the patient's preference in progress notes and again in the death summary after the patient died. The patient did not have next of kin.

Following his death, the patient's body was transported to the morgue and, in accordance with hospital policy for conducting autopsies on patients without next of kin, the pathology department requested permission from the chief of staff to proceed with the autopsy. (When next of kin is available, the policy requires him or her to sign a form consenting to an autopsy.)

The chief of staff was not informed by the pathologist of the decedent's wish to not have an autopsy. The chief of staff, considering that the hospital's educational mission and core value was to improve care for living patients through the knowledge gained from autopsy, decided that these considerations provided ample justification for proceeding with it. The chief of staff gave permission for the patient's body to be autopsied, and the autopsy was performed.

The postgraduate trainee asked the facility's ethics committee about the chief of staff's decision. The committee scheduled a meeting to discuss the trainee's concern. The committee framed the question as follows: Was it ethically justifiable to override the decedent's clear and consistent premortem preferences regarding autopsy in order to benefit future patients and advance the hospital's training mission?

The committee's chairperson recognized that many of her committee members were ill-informed about autopsy practice and policy, and so invited the chief of pathology and regional counsel to provide them with needed facts and context for analyzing the case.

Background Information
The committee was given the following information:

According to data reported by the Agency for Healthcare Research and Quality, autopsies are presently performed in fewer than 6 percent of nonforensic deaths. ² The decline in autopsies is attributable to a number of factors, including the fact that nonforensic autopsies are not reimbursed by most insurances; family members infrequently request autopsies, even though they and their progeny may benefit from knowledge derived from autopsy; and clinicians are concerned that if autopsy disproves their clinical diagnosis, they may face medical malpractice claims. ³

Despite these barriers, autopsy is regarded as the most definitive method for establishing cause of death and crucially important to improving outcomes for future patients. Autopsy contributes to improved patient outcomes because it provides data that can increase the accuracy of clinical diagnosis. Currently, discrepancies between the clinical diagnosis and autopsy findings exist in as many as 30 percent of cases. ⁴

Because of the pivotal role played by postmortem examinations in improving care, the College of American Pathologists recommends that an autopsy be requested for every death. ⁵ As a result, many medical centers require that postmortem examination services be made available to the families of all decedents. These policies especially obligate the medical staff to attempt to secure authorization for postmortem examination services in deaths that are unexpected or proximate to a medical procedure. Finally, with few exceptions, the next of kin must authorize the postmortem examination in writing.

Exceptions to this requirement include forensic cases or cases in which it is reasonable to infer the potential for significant risk to public health or safety. ⁶ For example, an exception would apply to a patient suspected of having succumbed to a highly infectious disease that could require rapid mobilization of public health measures to prevent an epidemic. Whenever possible, the clinician responsible for the care of the patient at the time of death is the proper person to request permission from the next of kin to perform a postmortem examination.

Because of autopsy's important role in improving the quality of care, many institutions have instituted performance measures that encourage the securing of next-of-kin permission. These performance measures have demonstrably increased the number of autopsies performed.

Case-Based Response to the Consultation Request
Armed with background facts about autopsy and a fuller appreciation of its potential benefits, the facility's ethics committee turned its attention to the case. Its members conducted a spirited discussion both for and against adhering to a decedent's wishes concerning autopsy.

After carefully weighing and balancing the competing arguments, interests, and values, the committee concluded that it was not ethically permissible to override the decedent's clear and consistent premortem preferences, despite the potential benefit to future patients or the furthering of the hospital's training mission. In addition, the committee agreed that since the decedent's premortem preferences carried moral weight, those wishes should have been communicated to the chief of staff in order to inform her decision.

In a written summary of the case, the committee noted that a common reason given by patients or next of kin for refusing postmortem procedures is that such procedures conflict with religious or cultural views regarding respectful treatment of the body. Although some committee members argued that a corpse cannot be harmed, others averred that harm is a socially constructed concept, and that certain religious or cultural groups believe the soul or spirit may be harmed if the body is not treated in accordance with prevailing custom. ⁷ It is the patient or family's definition of "harm" that should trump other considerations in these cases.

The case summary also noted that overriding decedents' premortem preferences regarding the use and disposition of the body seriously risks undermining public trust in the integrity of both health care professionals and health care institutions, especially when it is the preferences of the most vulnerable that are ignored (i.e., patients without next of kin). Moreover, in somewhat analogous cases — such as organ and tissue donation or use of the body for education or research — consent is required. The committee did not believe that autopsy deviated materially from organ or tissue donation or use of the body for education and research. It did believe, consequently, that similar ethical reasoning should apply in both donation and autopsy.

Finally, some clinicians claimed that when patients agree to be admitted to a teaching hospital, such agreement implies consent to having one's body used in ways that can advance the public good: training the next generation of clinicians, for example, or improving care through autopsy.

The committee took strong issue with this claim. Few patients are able to choose the institution in which they receive care. That choice is usually dictated by insurance, employment, or other factors unrelated to patient choice. And even if the patient does choose a teaching facility, he or she is not giving that facility carte blanche consent to make his or her living or dead body available for a presumed social good.

Committee members discussed their opinion with the principals in the case, including the chief of staff, chief of pathology, the resident physician, the attending physicians, and the staff nurses involved in the patient's care prior to his death. They recommended that, in the future, the chief of pathology communicate the decedent's preferences, if known, to the chief of staff. Committee members emphasized that they are available to consult with the chief regarding future cases, should they arise. The committee also volunteered to attend the next pathology service meeting to provide an in-service to attending physicians, residents, and fellows related to ethics and autopsy. The full consultation summary note was sent to medical records to be scanned as an entry in the decedent's electronic health record.

Systems-Oriented Ethics Consultation
A systems perspective may not alter the ethical analysis of the case. However, a systems perspective draws attention to facets of the case that might otherwise have been ignored or neglected — thereby enabling one to view the case through the widest possible lens. Specifically, a systems approach causes the ethics committee to look routinely "upstream" for predisposing factors that contribute to the development of ethics concerns. A systems approach places a high priority on identifying strategies to prevent future recurrences of the problem, and these strategies will frequently target the organization's systems and processes.

At the same time, systems-savvy consultants will apply a healthy skepticism to proposed corrective strategies; they will ask whether the proposed strategy is likely to generate its own set of unintended negative consequences. Ethics consultants, in collaboration with quality-improvement experts, may identify simple measures to monitor the strategy's effectiveness over a period of time — before committing themselves to broadscale deployment. ⁸

An effective systems-level response does not rely on educational interventions or other interventions aimed solely or primarily at changing individual behavior. Rather, systems-level interventions aim to make the ethical response easy and virtually inevitable — regardless of whether the clinician is new to the system, tired and distracted, or even unsure of what an ethically appropriate response is. (Electronic reminders to update a patient's advance directive upon admission to acute care are an example.) Systems-oriented ethics consultants are guided by two related assumptions:

• Although not denying the importance of cultivating individual virtue as well as proficiency in health care ethics, the consultant recognizes that the staff's capacity to make sound ethical judgments and then act on those judgments is powerfully shaped by factors extrinsic to the individual — most notably, the organizational environment.
• Working with a philosophy similar to that underpinning quality management, the consultant realizes that the cause of gaps between how staff ought to act and how staff is acting should be sought first within the organization's systems and processes, not in the individuals who work within the system. ⁹

With these assumptions in mind, we turn back to the autopsy case.

Systems-Oriented Response to Autopsy Case
The systems-oriented ethics consultants in this case concurred with their colleagues' analysis that, in most circumstances, patients' premortem preferences for the use and disposition of the body should be elicited and abided by. In addition, they made several other systems-level recommendations.

Revise the Medical Center's Autopsy Policy, Making It Harmonious with Analogous Policies
In this case, the consultants recognized that the medical center's policy regarding autopsy contributed to the chief of staff's decision to override the decedent's premortem preferences; and in fact, the chief followed the existing policy precisely as written.

Unfortunately, concerning autopsies, the facility's policy was silent on the bearing decedents' premortem preferences should have, especially in cases where there are no known next of kin. The pathologist interpreted this omission in policy to mean that premortem preferences ought to have little bearing on the decision, and thus did not mention the decedent's preference to the chief of staff.

Noting this, the consultants recommended that decedents' premortem preferences be privileged in policy. However, they also anticipated that there might be cases in which the decedent's premortem preferences are at odds with those of next of kin. A revised policy would need to address this possibility in a consistent and ethically coherent manner.

The consultants also recommended that, with respect to the decedent's expressed wish concerning the use and disposition of his or her body, ethically analogous policies be harmonized with one another. Preferences concerning use of the body after death, they noted, include giving or withholding consent for autopsy, as well as organ, tissue, and eye donation, or use of the body for research or education.

Ensure that the Incentives and Rewards Associated with Performance Measures Align Practice and Behavior with Ethical Norms and Standards
The consultants were concerned that recently instituted performance measures could contribute to ethical missteps. The facility's autopsy rate was lower than the benchmarks at similar medical centers. As a result, the facility's clinical executive board had directed that a database track the volume of autopsies and had established a target goal to be met by all services. Service chiefs were rewarded for attaining or exceeding the measure.

The consultants observed that these incentives could have ethically perverse effects if meeting the target were to take precedence over other values, such as respecting and abiding by the decedent's or family's preferences. In short, they said, the process through which a performance measure is attained needs to be as ethically unimpeachable as the objective itself.

The consultants plan to begin a dialogue on this "means/end" problem at the next clinical executive board meeting. They hope to ensure that incentives are awarded only when performance targets are met in an ethically acceptable manner.

Consider Including Discussion and Documentation of Postmortem Preferences for Autopsy as Part of an Advance-Care Planning Process
The consultants recognized that information regarding premortem preferences for autopsy is frequently unavailable, whether the patients involved have family or not. Unlike organ and tissue donation, no policy requires a discussion of autopsy as part of an advance-care planning discussion between patient and clinician.

Noting that advance-care planning is considered an appropriate vehicle for discussing organ and tissue donation, the consultants said it was difficult to see — on either ethical or prudential grounds — why the same process should not be utilized to elicit and document patient preferences regarding autopsy. Conversations regarding organ donation and autopsy are both undertaken to ensure that patient preferences regarding what happens to the body after death are known and adhered to as closely as possible. In addition, both organ donation and autopsy can benefit future patients, including, in some cases, the patient's own family.

Most importantly, the consultants said, utilizing advance-care planning as a vehicle for eliciting and documenting postmortem preferences will help ensure that the facility or next of kin have the information necessary to act in accordance with the decedent's premortem expression of preferences and values. The ethics committee agreed that a strategy of ensuring that patients' postmortem preferences are known should be studied further and possibly incorporated as a revision to the present advance-care planning policy.

Initiate a Quality-Improvement Strategy
Finally, the ethics committee chair asked representatives of the facility's quality-management program to help the committee develop, measure, and monitor an improvement strategy. As a starting point, the quality manager recommended collecting baseline information regarding the number of autopsies performed in the previous six months on patients without next of kin. The quality manager also offered to conduct a medical record review to identify the number of cases in which decedents' documented premortem preferences were either overridden and or simply not documented.

These data will suggest the scope of the problem and inform future planning. For example, the ethics committee may consider monitoring the reasons given for trumping premortem preferences regarding autopsy, to ensure that the indications for doing so are ethically defensible.

Building a Needed Bridge
Ethics committees are in a unique position to build a bridge between the concerns of clinical ethics and those of organizational ethics by applying a systems approach to the cases brought before them. A systems approach to ethics consultation takes into account the fact that ethics cases are imbedded in, and influenced by, a larger organizational context. Given this understanding, consultants should always ask if the case at hand has systems-level antecedents and implications. Recurrent cases, in particular, must be treated as red flags that portend future cases unless underlying systems issues are identified and addressed.Â

The opinions expressed in this article are those of the authors and do not reflect the views of the National Center for Ethics in Health Care, the Veterans Health Administration, or the Department of Veterans Affairs.

NOTES

1. C. R. Seeley and S. L. Goldberger, "Integrated Ethics: Synecdoche in Healthcare," The Journal of Clinical Ethics, vol. 10, no. 3, Fall 1999, pp. 202-209.
2. K. Shojania, E. Burton, K. McDonald, et al., The Autopsy as an Outcome and Performance Measure, Agency for Healthcare Research and Quality, Rockville, MD, 2002.
3. J. Robertson, ed., Autopsy: Life's Final Chapter, American Medical Association, 2001, available at www.ama-assn.org/ama/pub/category/ 7594.html.
4. Shojania, pp. 3-5.
5. K. A. Collins and G. M. Hutchins, ed., Autopsy Performance and Reporting, College of American Pathologists, Northfield, IL, 2003, Appendix E.
6. Robertson.
7. B. A. Koenig and P. A. Marshall, "Death: Cultural Perspectives," pp. 546-559; and T. Parsons, "Death: Western Religious Thought," pp. 578-594, in S. G. Post, ed., The Encyclopedia of Bioethics, 3rd ed., vol. 2, Macmillan Reference, New York City, 2004.
8. E. Fox, M. Bottrell, M. B. Foglia, et al., "Preventive Ethics: Addressing Health Care Ethics Issues on a Systems Level," unpublished manuscript, 2005.
9. J. W. Ross, J. W. Glaser, D. Rasinski-Gregory, et al., Health Care Ethics Committees: The Next Generation, Jossey-Bass, San Francisco, 1993.