Log in

BY:  FR. JOHN TUOHEY, Ph.D.

Fr. Tuohey is chair, applied health care ethics, Providence St. Vincent Medical Center, and director, Providence Center for Health Care Ethics, Portland, OR.

The Portland Service Area (PSA) of the Providence Health System Oregon Region, where I work, comprises three hospitals: Providence St. Vincent Medical Center and Providence Portland Medical Center, both of which have internal residency programs, and Providence Milwaukie Hospital, which has a family practice residency program. Each has an ethics committee.

Historically, one responsibility of an ethics committee has been to provide a resource or forum for ethics consultations. Six years ago, the PSA committees handled case consultations mostly on an ad hoc basis. In one hospital, a member of the pastoral care team functioned as the facility's "unofficial ethicist." At another, the director of the internal medicine residency program played a similar role, though for the most part with members of the house staff. Occasionally a case would find its way to the whole committee for discussion.

That this approach was not completely adequate can be seen in the types of consults the PSA saw in 2005: More than a third required the time necessary for a care or family conference, and almost a third were urgent cases. Something more structured was needed to respond to these scenarios.

In 1999, the region established an endowed chair in applied health care ethics, the position I hold. At the time, some staff members had the idea that all of these consultations could be done by the chair. There was also, however, a great deal of interest in establishing a consultation service with specially trained consultation teams, rather than leaving it to one person. Many questions accompanied this interest: Should the committees be trained? Should there be separate teams? How should we get word out so that more people would use the service? When should a consult be initiated? What qualifications do ethics people need and what kind of training should they have?

As holder of the endowed chair, I had a preference for special teams, and we ultimately decided in favor of such teams as an extension of the committees' work. The region's newly established Providence Center for Health Care Ethics, under my direction, would ultimately be responsible for the consultation service. This made sense for two reasons, both of which are perhaps peculiar to our situation:

• The center was already an integral part of residency education at our teaching hospitals; and, through residency education, it was integrated into the broader opportunities for physician education.
• The center itself was to be the principal resource for all staff ethics education in the region. This would allow us to create different teams in different hospitals with effective quality measures, a single set of competencies, and a single model for decision making and standard tracking. It would also, as it turned out, allow us to use our ethics programs to build an educational and information infrastructure, which we have found to be a very effective way to attract requests for consults because it provides a broad understanding of and confidence in our approach.

We have also found, as will be seen below, that ethics teams are more effective in their work when there is a broad understanding of what ethics is and how it can be helpful.

Challenges and Questions
The first challenge was to choose a model for case consultation and then develop an initial team around use of that model. PSA internal medicine residency faculties were already familiar with the four-quadrant model employed in Clinical Ethics¹ so we began there. Beginning with a small, interdisciplinary group of physicians, nurses, case managers, and chaplains chosen to constitute the first teams, we created a 30-hour education program around this model. Our program covered the knowledge and skill competencies suggested by the 1998 report of an American Society for Bioethics and Humanities task force, ² modified somewhat to include knowledge competencies related to the Ethical and Religious Directives for Catholic Health Care Services. ³ This was followed by a 40-hour mentoring program that included weekly case discussions, continued education, and discussions of relevant journal articles.*

*This one-hour weekly meeting of the consult team continues as a requirement for continued participation on the team, in addition to participation in the CCIII programs described later in this article.

Having created an education program, we went public with our consultation teams, publishing pamphlets that explained what the teams did and how to access them, and began tracking the cases. What we found was interesting.

In "marketing" the consult team, we discovered that many of the consults done in the past had been rooted more in familiarity with individuals involved than with ethics per se and the consultation process. Many staff perceived involvement in an ethics consultation as an indication that something was wrong. Not a few cases could have been more easily addressed if the consult had been called earlier, but this concern about error or wrongdoing made some staff hesitant, if not defensive. We needed to more broadly address the issue of the purpose of an ethics consultation.

A frequent question was: "How do I know when an ethics consult is needed or helpful?" People wanted a concrete list of situations in which a consult should be called. Some even wanted us to develop "triggers" so that certain clinical indications would immediately initiate a consult, even if no one had requested — or wanted — the consultation. Our tracking showed us that the consult team could not be successful unless we created a broad educational infrastructure to support it.

Problems with the Model
We soon discovered that the model we were using did not respond well to the kinds of cases we were seeing and to the concerns we were bringing to them. By working cases through this model (at least the way we were using it), we were failing to adequately incorporate certain ethical considerations that we felt were important. In some cases, using the model brought us to one conclusion, whereas we could see that if we had brought other ethical considerations to bear on the case, we would have reached a different conclusion. We found ourselves adapting the model in each case to more explicitly address our needs as well as to give us the confidence that the model itself would adequately organize our reasoning.

To help organize an ethics case, Clinical Ethics offers a model with four quadrants: "Indications for Medical Intervention," "Preferences of Patients," "Quality of Life," and "Contextual Features." My point here is not to critique this model, for it has been well used in a variety of settings; I do, however, want here to illustrate how our experience challenged us to make some major adaptations.

For example, one issue with which we frequently struggle has to do with the timely discharge of patients, particularly homeless patients and patients with special needs that make their placement difficult. Sometimes the prevailing attitude has been that our mission prohibits us from discharging people to the streets, or to an environment, such as a skilled nursing facility (SNF) or nursing home, to which the patient does not want to go. Our mission, it is argued, requires us to provide "hospitality" for such patients. Our desire to have good patient satisfaction scores also informs this approach, because we do not want to force a placement issue on reluctant patients or families.

In such cases, "patient preferences" are clear: The patient wants to stay in the hospital, or, at least, does not want to go anywhere but home. The "medical indications" are such that discharge of some kind is appropriate. This would seem to be the dilemma: to discharge or not discharge — how do we decide? There are, however, more than two sides to this dilemma.

What is helpful for us in such cases is explicit recognition that the dilemma is playing itself out in an acute care facility. As such, our staff, while highly competent in their acute care fields, is not necessarily competent or available to provide the kind of care provided in an SNF or nursing home. Our professional integrity is called into question by our keeping patients who do not need acute care. There are also risks for the patient. Studies show that long-term stays in an acute care setting put patients at risk for infection. ⁴ It is dangerous to keep patients who do not need to be hospitalized. The principle of nonmaleficence, our duty to protect patients, as well as staff, needed more explicit integration into the conversation.

There is also a public order dimension to the problem, which can be illustrated by a recent emergency grand rounds I attended at Providence Health and St. Paul's Hospital in Vancouver, British Columbia, where I was attending an ethics conference.

In a conference presentation, Grant Innes, MD, chair of the Department of Emergency Medicine, cited several international studies that show overcrowding and increased risk of mortality in the emergency department (ED) to correlate more closely with overcrowding in the hospital itself than to non-urgent use of the ED. Several studies he cited show a discernable risk in overcrowding and mortality in the ED when a hospital reaches more than 85 percent occupancy and a crisis state when occupancy is more than 90 percent.⁵ The community is at risk if a hospital is overcrowded, and this risk is difficult to justify if some of those patients who are part of the overcrowding do not need those beds. Cases like this had us appealing to arguments not easily rooted in the model we were using.

Consider, also, the case of patients with mental illness or dementia. Our PSA cares for a great many such patients. These illnesses can compromise a patient's decision-making capacity. The illness does not, however, always take away the patient's desire to express his or her own preferences. Initially, we found ourselves viewing mental illness or dementia as part of the clinical picture, and lack of decision-making capacity as requiring a surrogate or advanced directive. The model did not seem to factor in the strong will of the patient, which, when accompanied by physical strength, can make care plans dangerous for both the patient and the staff. Our experience is that a patient's coping mechanisms, regardless of his or her capacity to make decisions, are critical to quality-of-life assessments. Patient preferences are better addressed within a context of autonomy that allows for respect for both patient choices in the absence of decision-making capacity and patient decisions made when such capacity is present.

Reworking the Model
In following the Clinical Ethics four-quadrant model, we discovered that we needed another — a model that, first, could better address the kinds of cases we see; and, second, enable us to bring our ethical values to bear more explicitly on the decisions we make.

Presenting our new model in detail would be beyond the scope of this article. Suffice it to say that we kept the four-quadrant idea, but amended some of the quadrants.

Clinical Integrity
For the Indications for Medical Intervention quadrant, we substituted Clinical Integrity, thereby allowing us to deal more explicitly with such professionalism issues as truth-telling, disclosure, and conflict of interest.

Autonomy
For the Preferences of Patients quadrant, we substituted Autonomy, which we broadened to more explicitly include patients' coping skills and the influence of faith, culture, race, ethnic background, and life experience.

Beneficience
For the Quality of Life quadrant, we substituted Beneficence, which we narrowed to refer to the quality of the patient's life activities, thereby excluding any potential psychological/emotional benefit — such as, for example, the "benefit" to a patient of being kept alive until a court can rule on an unsigned will establishing a trust for surviving children in the midst of a bitter divorce.

Justice and Nonmaleficence
For Contextual Features, we substituted considerations of Justice, such as the public-order concerns mentioned above, and issues of Nonmaleficence, such as a perceived duty to protect some third party (a staff member, for example) from harm, or to protect the patient himself or herself from some harm.

Educational Structure
Having reworked our model, we then proceeded to develop an educational structure to support it, so that staff members could better appreciate the model's role in ethics decisions and have confidence in its use. We called this structure our Core Curriculum (CC) I, IIA, IIB, and III programs. They serve as the basis of our residency and physician education programs, and are used as well by the general staff.

CCI and II together constitute a 24-hour program with three modules.

Core Curriculum I
In CCI, we present an overview of ethics and our model. Key in this eight-hour presentation is the idea that ethics cannot be reduced to the knowledge of what is and is not allowed. Ethics is richer, having to do with understanding the nature of human relationships and the obligations, responsibilities, and duties that flow from them.

For example, we are not content with knowing that the Health Insurance Portability and Accountability Act of 1996 does not allow one to share, with a minor child, important genetic information concerning a parent, without first obtaining the parent's consent, even if the child's health is thereby put at risk because the provider, lacking that information, is unable to pursue appropriate diagnostic or prophylactic interventions.

Even to begin to understand our ethical obligations in such a case, we need to consider the myriad relationships involved and the duties that fall on each party. This discussion helps us weigh just how much "pressure," for example, we can ethically impose on the patient involved to reveal such information. It gives us guidance in discovering the line between persuasion and coercion, so that we don't have to yield absolutely to the "you cannot reveal" of the law.

In presenting our model, our goal is to also help our staff understand and have confidence in an ethics consult by giving them exposure to what goes on in a consult and how it works. We do not expect everyone who completes CCI to be able to lead a consult; we do expect him or her to know enough about how it works to respect its role in the care plan.

Another key objective is to introduce staff to what we call "indicators" for a consult. Indicators are not triggers that automatically engage the service. We want our staff to understand the goals and process of the consult so that they can recognize situations or circumstances that indicate a consult might be helpful. These indicators might include:

• Failing to follow standards of care to keep the family happy
• Imposing a surrogate's decision on a patient fully capable of expressing his or her own choices with regard to a feeding tube even though he or she does not have the capacity to make the decision
• Prolonging a patient's last days on the assumption that he or she would find "emotional benefit" in the knowledge that his or her family was present
• Being asked to inform the police of the discharge plan of a patient in an inpatient psychological unit, against the patient's wishes, because he or she is a "person of interest" in a crime

None of these instances is necessarily a clear-cut "right/wrong" case. Each case involves circumstances indicating that a consult might be helpful because, in each of them, neither the relationships among the players nor the responsibilities and duties of each player are immediately clear.

CCI sessions have been critical for us, not so much in "getting out the word" that we have a consult service, but, rather, in making people aware of the variety of things one should pay attention to in the care setting, and in giving them confidence that there is a model for decision making.

Core Curriculum IIA and IIB
These sessions reflect what our tracking has told us are the ethical issues our facilities see most frequently. Hence they deal with the kinds of indicators to which we want our staff to be most alert. One four-hour module is based on patient decision making, which includes noting the fact that decision making can often lead to conflict between the care team and the patient or family. Another four hours, devoted to professionalism and organizational ethics, addresses the kinds of issues that can challenge professional judgment and behavior. Such issues might include:

• Conflicts of interest
• Organizational pressure to focus on increasing patient/family satisfaction, rather than on providing care
• Pressure to avoid lawsuits by yielding to questionable care demands

CIIB is an eight-hour curriculum that addresses ethical issues related to palliative care and the end of life.

Core Curriculum III
CCIII programs address special topics that arise. We usually hold four or five such programs every year.

For example, one CCIII program dealt with the 2004 papal allocution on mechanically administered nutrition and hydration. Another program addressed the role of deception in care. Still another explained to employees the ethical rationale for the allocation of limited flu vaccine during the 2004-2005 flu season. Some of these 90-minute programs are available to a wider audience through video conferencing.

In the end, our efforts to create an effective consultation service have resulted in an annual educational program of more than 400 hours for the PSA staff.* Thanks to a grant from the Robert Wood Johnson Foundation, we are presently working with nurse researchers from the University of Portland to study the impact of the program on patient care.

*The 400 hours include time spent on physician and resident education, as well as regional and other service-area programs.

We know that having this infrastructure built on our decision-making model has led to some dramatic practice changes. In one hospital, physician requests for a consult have jumped from a minority of cases to over 55 percent of them. Fifty-seven percent of those requests were for recommendations concerning care decisions, and 20 percent were requests for moral support in difficult situations.

The program has enabled us to make some interesting observations. More than half of consults, for example, involve questions about who is to make the decision, or what authority he or she has to actually make certain decisions. Such discussions demonstrate that people are growing in awareness of the indicators for a consult related to autonomy; they also give us direction for our CCIIA program. Another 17 percent of the consults are related to a conflict between the care team and the patient or the patient's surrogates about referral to comfort care.)

We have also found that 90 percent of those engaged in a consult found it helpful to them. ⁶ We attribute this not only to quality but also, as a result of our education, to reasonable expectations of what a consult can and cannot do. We have recently received approval from our institutional review board to conduct research aimed at better identifying when conflicts in care are likely to take place⁷ and which interventions (ventilators, CPR, tracheotomies, mechanical nourishment) are likely to be the most problematic.

Responding to Ethical Needs
In closing, we have moved ethics consultation out of ethics committee work per se and created specially trained teams to conduct it. As part of that transition, we have developed a model based on our own experience, an educational infrastructure to support that model, and a disciplined tracking of consult data. All this has helped us to identify and respond to some of the hospital ethical needs that ethics committees are charged to address.   

NOTES

1. R. Jonsen, M. Siegler, and W. J. Winslade, Clinical Ethics: A Practical Approach to Ethical Designs in Clinical Medicine, 4th ed., McGraw-Hill, New York City, 1998.
2. Task Force on Standards for Bioethics Consultation, Core Competencies for Health Care Ethics Consultation, American Society of Bioethics and Humanities, Glenview, IL, 1998.
3. Ethical and Religious Directives for Catholic Health Care Services, 4th ed., U.S. Conference of Catholic Bishops, Washington, DC, 2001.
4. See for, example, Nicholas Graves, "Economics and Preventing Hospital-Acquired Infection," Emerging Infectious Diseases, vol. 10, no. 4, April 2004, pp. 561-566.
5. Among the studies Dr. Innes cited was A. J. Forster, et al., "The Effect of Hospital Occupancy on Emergency Department Length of Stay and Patient Disposition," Academy of Emergency Medicine, vol. 10, no. 2, February 2003, pp. 127-133.
6. This figure compares favorably with those drawn from evaluations of ethics consults at other sites. See L. J. Schneiderman, et al., "Effect of Ethics Consultations on Nonbeneficial Life-Sustaining Treatments in the Intensive Care Setting: A Randomized Controlled Trial," JAMA, vol. 290, no. 9, September 3, 2003, pp.1,166-1,172.
7. One study showed that in 76 percent of the cases reviewed, the care team wanted a more aggressive approach than the family did, whereas in 24 percent it was the family that wanted the aggressive care. See C. M. Breen, et al., "Conflict Associated with Decisions to Limit Life-Sustaining Treatment in Intensive Care Units," Journal of General Internal Medicine, vol. 16, no. 5, May 2001, pp. 283-289.