Dealing With Conflict in Caring for the Seriously Ill: Title and subTitle Break“It Was Just Out of the Question”

Mrs B was an 84-year-old woman with advanced dementia who developed an aspiration pneumonia requiring an acute care hospital admission to the intensive care unit. During her recovery, she had difficulty during an informal swallowing study. To provide nutritional support, her family agreed to the temporary placement of a nasogastric tube, even though she pulled it out twice. Physical restraints were used to prevent further episodes of tube dislodgement. The hospital medical team recommended placement of a percutaneous gastrostomy for feeding. However, based on the patient’s previous wish not to end up in the same state as a sister-in-law with Alzheimer dementia, her husband and the rest of her family believed that she would not have wanted a long-term feeding tube, and she was transferred to a nursing home without a feeding tube. The family’s hope for her nursing home stay was for her to regain sufficient strength that she could resume oral intake. They were uncertain if rehospitalization would be appropriate at this point in her illness.

A Perspectives editor interviewed Mrs B’s husband, Mr B, in December 2003, and the medical director of Mrs B’s nursing home, Dr Q, in January 2004.

MR B: My wife has been bedridden for the past 2 years and during that time has needed special care and caregivers since she has lost the use of both her arms and her legs. In March of 2003, she developed pneumonia, and we took her to the hospital. The doctors repeatedly suggested . . . inserting a gastric tube for nutrition. [This reminded me of] the story of my sister, who had had a gastric tube installed. She died a few years ago, and the quality of her life for years was just nonexistent. She was not really living, and the tube was all that kept her going. We were a little disturbed by the frequency of the calls for the gastric tube. [The doctors were] quite insistent. We indicated that it was just out of the question.

DR Q: [When Mrs B was admitted], my understanding was that she was on a do-not-resuscitate status. But, if need be, [she] could be transferred to an acute hospital. [Her family] told me that she did not want any invasive measures. The more I talked to the family, the more it became obvious that the patient’s advance directives did state no IVs and no artificial feeding.

The medical, logistical, and emotional issues surrounding a dying patient are frequently complicated, and differences in perspectives of the patient’s loved ones and clinicians can easily result in conflict. Physicians usually assume that conflict is undesirable and destructive—which is true if the conflict is handled poorly. Yet conflict handled well can be productive, and the clarity that results can lead to improved decision making.

Although some empirical literature exists discussing conflict in medical settings,^{1 - 10} only one study empirically tests a specific approach to conflict resolution.² In this article, we draw on evidence-based approaches used in other fields^{11 - 12} and make recommendations for mediation in medical settings.^{13 - 17} We review the course of Mrs B’s care after her transfer to the nursing home and identify missed opportunities for openly discussing and resolving conflicts that arose between her physicians and family. We describe a typology of conflicts relevant to palliative care, pitfalls to avoid, useful communication tools, and a step-wise approach to conflict resolution. Finally, we discuss the use of treatment trials as a strategy to address conflicts about the use of life-sustaining medical interventions.

Conflict in medical settings has been defined as “a dispute, disagreement, or difference of opinion related to the management of a patient involving more than one individual and requiring some decision or action.”³ Common sources of conflict are outlined in Table 1. This typology is based on the available empirical studies of conflict, all of which involve intensive care units.^{1 ,3} Although we acknowledge the limited potential for physician-patient conflict in this setting—in which patients are frequently unable to communicate—our experience suggests that the typology is accurate for palliative care.

Conflict occurs frequently. Breen et al¹ examined conflict associated with decisions to limit life-sustaining treatment in 6 intensive care units at an academic medical center and found that at least one clinician described a conflict in 78% of 102 consecutive cases. Clinician-family conflicts were identified in 48% of cases, and clinician-clinician conflicts in 48% of cases. Studdard et al⁴ examined conflicts in 7 medical and surgical intensive care units and identified conflicts involving 32% of patients with long stays (defined as >85th percentile for that unit). One informal poll of physician executives indicated that this group spent “at least 20% of their time” dealing with conflict.⁹ In usual practice, conflict is probably underreported. For example, a resident physician may believe an attending physician’s treatment plan is inappropriate, but given the power differential between them, the resident will not articulate the conflict.¹⁸

Mr B’s remark about feeling “disturbed” indicates that he felt some degree of conflict with the hospital physician about the need for a feeding tube. The physician’s behavior—to make repeated telephone calls asking for consent—gave Mr B the sense of being in an adversarial situation, and neither Mr B nor the physician got a better understanding of each other’s views. Table 2 describes ineffective approaches that we have observed. The physician may have been reluctant to negotiate because of a lack of self-confidence or skill in negotiation, constraints of time or interest, or a belief that the way to preserve relationships is to avoid conflicts or to suppress anger when conflicts do arise. In addition, common physician personality attributes, especially perfectionism, compulsiveness, and values about scientific evidence, may contribute to reluctance to negotiate and an avoidance of conflict.^{19 - 20} Yet habitual conflict avoidance can increase stress over time rather than decrease it,²¹ and a study of physicians responding to complaints indicates that conflict carries strong emotional repercussions.²² Another study of nonphysicians indicates that replacing avoidance behaviors with open communication is better for relationships and for health,²³ a finding that likely generalizes to physicians.

The most important pitfall to avoid is trying to begin negotiations when the physician is angry or upset enough to feel “flooded,” that is, consumed with defensive thoughts and unable to focus on what the other person is saying. The amount of time required to recover from this, derived from research on marital fighting, is about 20 minutes.^{24 - 25} In this situation, it may be best to identify the disagreement, avoid being drawn into a discussion, take time to observe one’s own emotional reaction, and identify negotiation techniques.

The immediate issue facing Dr Q is how to understand the conflict between the family and previous health care professionals without becoming embroiled in it. Useful communication strategies are outlined in Table 3. We recommend that Dr Q leave aside the issue of whether a feeding tube was appropriate to suggest for a patient with advanced dementia²⁶ and instead explore Mr B’s perceptions of the issue from a nonjudgmental stance. For example: “It sounds like there was some discussion about the feeding tube while your wife was in the hospital. I’d like to know your perspective so that we can develop a way to work together.” This opening invites Mr B to discuss the issue, frames the issue in a nonjudgmental way, enables Dr Q to understand and explore Mr B’s perspective, and articulates an interest in building a therapeutic alliance with Mr B. If the opening is judgmental, “I heard that you didn’t want to follow the hospital doctor’s recommendations,” Mr B is likely to respond defensively.

DR Q: Initially, she was doing fine. She had a daily caretaker provided by the family. Then the patient had an infection, went to an acute hospital, and upon her return, started deteriorating gradually. Her oral intake [decreased to] less than 1000 [kcal/d]. In our discussions with the family, they knew that she was not to be fed [artificially long-term], but [asked] if we would “just try to bridge this phase.” I think that everybody rationalized by saying that it would be just for a very short period of time.

Mrs B’s family had previously declined a feeding tube, but this request marks a change in their thinking and an apparent conflict with the patient’s advance directive and Dr Q’s initial understanding of the family’s wishes. How should Dr Q discuss this with the family? A key issue for Dr Q is deciding what would constitute success. One definition of success would be convincing Mrs B’s family that Dr Q understands Mrs B’s true preference with regard to a feeding tube. This definition creates a “them vs me” mentality that focuses on who is correct, resulting in winners and losers. A better alternative might be for Dr Q to focus on interests rather than positions and define success as developing a working relationship with Mrs B’s family in which they could collaborate to identify solutions that meet Mrs B’s values. The shared interest in this case is protecting Mrs B’s well-being and preferences for care. Fisher et al¹² would describe this as “separating the people from the problem.” Negotiating a solution to a shared problem is easier and more constructive than trying to convince Mr B that he is misinterpreting his wife’s advance directive or debating who knew Mrs B’s wishes better. In this way, Dr Q can align himself with the family, show that he understands their situation, and is on their side.

A step-wise approach to dealing with conflicts, adapted from Stone et al,¹¹ is presented in Table 4. Step 1 is to notice that there may be a conflict. Step 2 is to prepare for negotiation, which involves examining 3 types of stories embedded in the conflict: the “what happened?” story, the “feelings” story, and the “identity” story. In this case, the “what happened?” story is that Mrs B’s condition has deteriorated, and her family is wondering if a new intervention (short-term tube feeding) could stabilize her. A mistake that physicians sometimes make in the “what happened?” story is to assume that they know the truth and that the goal is to convince the family of their wisdom. If the family remains unconvinced, physicians often view them as a “problem” and impute bad intentions or psychological problems to explain why they don’t get it.²⁷ Stone et al recommend approaching the situation with curiosity about the family’s “what happened?” story and seeking a new understanding that embraces both parties’ views.

The affective aspects of a conflict are as important as the cognitive analysis. Feelings of abandonment, guilt, or sadness can be as important as the facts. If unacknowledged, these emotions make it difficult to listen to the other party and have a tendency to leak into the conversation. Thus, mediation experts recommend that rather than ignoring feelings, one should attempt to recognize and acknowledge both disputants’ feelings. From Dr Q’s perspective, the feelings story is about his feeling frustrated that this issue is being reopened (he remarked “strange things influence the course of treatment”). For Mrs B’s family, the feelings story is about their sadness over her decline, guilt that they are not “doing everything,” and frustration that the physicians are not listening. If Dr Q can articulate an empathetic understanding of the family’s emotions, he often would be able to defuse a conflict, build trust, and facilitate negotiation.

The identity story focuses on how conflicts threaten our personal and professional identity. The identity story for Dr Q is that he thinks of himself as a good communicator and strong patient advocate, and the idea of ordering a feeding tube, which is contrary to Mrs B’s advance directive, threatens his identity as an advocate. On the other side, the clinical situation has prompted Mrs B’s family members to question their identity as a loving family, and Dr Q’s hesitation to place the tube is evidence that Dr Q does not understand them. Stone et al¹¹ recommend that participants in a conflict try to identify how the conflict may threaten their identity and to acknowledge that both parties are competent; that each may have contributed to a misunderstanding that caused the conflict. For the physician, a helpful self-assessment question is: “Why do I think the family is not hearing (or agreeing with) what I am saying, and why am I—a good, caring clinician—not hearing (or agreeing with) what the family is saying?”

Negotiating agreement is not the same as giving in or being nice.²⁸ Negotiation requires that the physician take a role as mediator, identify shared understandings, invite the family to collaborate, and acknowledge feelings without judgment, blame, or attribution (Steps 3-6 in Table 4). The objective should be to find a solution that satisfies the needs of both parties rather than to win the battle. Ideally, Dr Q would achieve 2 goals in his discussions with Mrs B’s family: first, all parties should feel that their perspectives have been heard and respected; and, second, that medical decisions are based on sound evidence, reasonable clinical judgment, and respect for patient and family values.

DR Q: She had the NG [nasogastric] tube placed and the oral intake improved to the point where we removed the NG tube. After that, we said we would not place the NG tube back in. She did relatively well. She would look at me and answer very simple things—was she in pain, or not, things like that. Her daughters told me that they had meaningful conversations with her. I was never present [for these], so I really don’t know.

The distinctive feature of conflicts between physicians and family members, in our experience, involves surrogate decision making. The ability of families to report what a patient wants is only slightly better than chance,²⁹ and their decision making may be influenced by their emotions. Complicating the situation, family members often complain that health care institutions do not respect their role as caregivers,³⁰ that they do not get the information they need,³¹ and that their cultural values are ignored.^{32 - 33}

Consequently, resolving conflicts between physicians and family members often requires that the physician move from trying to convince the family toward seeking to understand why the family holds a particular view.³⁴ Family misunderstandings about medical facts are most likely due to inadequate or ineffective communication.^{35 - 37} For example, Dr Q could further explore the family’s sense of what a meaningful conversation had been to deepen his understanding of the family’s view of Mrs B’s quality of life; then he could reframe the issue: “In thinking about this issue, I’d like to ask, ‘What would Mrs B say she wanted if she could think clearly?’” One study showed that when families are asked in this way, the surrogate decision maker was more likely to accurately represent patient wishes.³⁸ Finally, although it is tempting for Dr Q to try to fix longstanding family dynamics, he should recognize that he is not a family therapist.

Suppose that Dr Z, a partner of Dr Q, stopped him in the hallway and said: “I was covering over the weekend and I have to tell you that it’s just ridiculous that Mrs B got a feeding tube. Her advance directive clearly states that she didn’t want it!”

Distinctive features of conflicts between physicians include different understandings of patients and family members, especially since clinicians may focus on different aspects of illness; different domains of medical knowledge about efficacy and outcomes of different treatments; incomplete or inaccurate insight about the physician’s own views, values, and identity issues; and the ability to acknowledge the professional identities and values of others.^{39 - 40}

The Dr Z scenario illustrates how a physician may defuse such an attack.¹¹ Although the normal response to an angry attack is “fight or flight,” this often escalates the fight or leads to avoidance. It is possible to reframe a personal attack as an attack on the problem. Dr Q might say: “Well, I can tell you that dealing with Mrs B’s family has not been a simple process and it has been frustrating for me as well.” Another strategy would be to reframe the attack as a request for advice: “It hasn’t been easy and if you have constructive suggestions I would love to hear them.” A third strategy would be to offer to explain: “The family and I have spent a great deal of time considering the right course of action. Perhaps when you have time I can fill you in on what happened.”

Similarly, Mrs B’s nurse might disagree with Dr Q’s position on the feeding tube. Studies indicate that physicians do not always recognize nurses’ perspectives on conflict.^{41 - 47} In a study of conflict in intensive care units, nurses identified nearly twice as many conflicts as were identified by both the physician and the nurse.³ In a large survey, nurses rated “physician value and respect for nurse input and collaboration” significantly lower than did physicians, suggesting that physicians’ perceptions of the degree of collaboration are markedly different from the impressions of their nursing colleagues. Finally, although another large survey indicates that nurses and physicians agree that their working relationships are essentially positive, when asked about the most important element of a good working relationship, physicians most often mentioned “willingness to help,” whereas nurses emphasized “mutual respect and trust.”⁴³

Although the same issues described above for conflicts between physicians may be present in physician-nurse conflicts, there are often additional issues related to gender and communication style.⁴⁸ Tannen and Guarnashelli⁴⁹ have described how women tend to use conversation as a way to connect by describing and responding to feelings while men tend to use conversation to achieve or maintain social status.

In addition, physicians and nurses are trained to attend to different aspects of patient care. The bedside nurse may appeal to her specialized knowledge to convince Dr Q to change his mind (“She just lies there and moans. I can’t believe we are going to prolong her suffering. What are you thinking?”). Although it would be easy for Dr Q to hear this as a criticism of his expertise (rather than a comment about her moral distress), he could redirect the conversation to focus on the impact on the nurse. For example, Dr Q might respond, “I’m gathering that this decision to place a feeding tube in Mrs B has had some impact on you. I’d like to understand your opinion and have you participate in the next family meeting.” Dr Q’s objective would not be to provide an outlet for the nurse to ventilate her feelings but rather to create a more collaborative relationship with the nurse.

In our experience, physician-patient conflict occurs less frequently than the other types of conflict although this may be related to patient reluctance to disagree openly with physicians. When physicians do perceive conflict, we find that they frequently raise concerns about patient competence. It is often easier to question a patient’s competence and ability to understand a complex situation than to elicit and understand the patient’s story.⁵⁰ Yet a deeper understanding of the patient’s story may reveal other issues; in a study of the reasons patients left the hospital against medical advice, the majority of explanations offered by patients involved personal issues.⁵¹

DR Q: [Mrs B’s] oral intake became less and less. At that point, we discussed the possibility of temporarily hydrating her. My idea at that time was to maybe give her a liter or two, because she was clinically dehydrated. Her oral intake improved and her mentation improved. After the family saw that she improved, the next time her oral intake declined, they asked for IVs again. So that same scenario was repeated a few times. Toward the end, maybe a month or two before she died, I was at the point where I felt that she should not be getting any more IVs. I think the way I put it was, “I don’t think that this is helpful anymore.” I think that [the family] realized that themselves, too. When it came down to really saying that it was time to let her go, the family needed more time to settle in, and feel that everybody was ready for that. In a way, they were thankful for us. Thankful for us not pushing them to say, “Okay, no more feeding, no more IVs, nothing like that.”

Treatment trials can be used to assess whether a particular intervention is meeting the goals of care. Thus, in Mrs B’s case, the physician and family agreed that if she was not mentally awake that medical treatments should be limited; the question was whether it was “worth it” to try intravenous hydration. Rather than engaging in a theoretical debate over risks and benefits, a time-limited treatment trial was proposed. In our experience, treatment trials work best if they are explicitly structured (Table 5). The process of developing agreement about the goals and structure of the trial builds trust. Dr Q essentially conducted a series of informal treatment trials of artificial nutrition and hydration with Mrs B, and his practice reflects studies indicating that physicians who use a process of deliberation should draw on the patient’s medical condition and the family’s readiness to withdraw.^{52 - 53} In our experience, explicit use of treatment trials allows physicians to initiate discussions about medical interventions of unclear value and it functions as a mechanism to guide withdrawal of those interventions in a dying patient. The value of structured communication about life-sustaining treatments has been demonstrated in case series reports in an intensive care unit.^{54 - 55}

DR Q: Once I did agree to leave the NG tube in, I think I could not go back and say, “I’m taking the tube out.” Once I said okay, the rest had to be negotiations. If I had said early on that we had to respect Mrs B’s wishes, if I had given [as] a firm base [the] end of NG tube and end of hydration treatments, she would have succumbed to her disease much sooner. But I think that I just could not go back once I said, “Yes, let’s try.”

Dr Q’s willingness to take responsibility for his role improves his ability to negotiate. While Dr Q might be tempted to project all the responsibility onto the family, this stance would subtly contribute to an attitude that “if only the family would change, my life would be easier.” Some ways that physicians contribute to conflicts are so embedded in the health care system that they are easy to miss. Examples include: avoidance of end-of-life issues until late in the patient’s course, changes in care plans related to rotating schedules of physicians, and paternalistic physician attitudes that undervalue collaboration.³⁹ Building self-awareness about these issues can enable physicians to develop a greater capacity for negotiation.^{56 - 57}

The approach we outline will not enable physicians to resolve every conflict. Bringing in an outside person may help defuse tension, depersonalize the issues, and identify new solutions. Thus, step 7 in Table 4 involves seeking negotiation assistance from ethics, risk management, or palliative care consultants.⁵⁸ Empirical studies indicate that ethics consultations can help deal with difficult conflicts and the use of futile medical treatments,^{59 - 60} and even then some conflicts will persist.⁶¹ If differences prove to be irreconcilable, it can be appropriate for the physician to transfer care.⁶² Here again, physician empathy for the experience of patients and families will help improve an inevitably difficult situation. While irreconcilable differences invariably raise physician concerns about lawsuits, it is worth keeping in mind that physicians are more often sued over relationship and communication issues than competence issues.⁶³

In the end, we think that the intravenous hydration Mrs B received supported the goals that Mrs B and her family had for her medical care. A willingness to acknowledge conflict and the use of a step-wise approach to conflict could have helped the acute care physician (who made the repeated calls) negotiate more effectively. For Dr Q, framing the hydration as a treatment trial could have engaged family members in the decision-making process to a greater degree, making the final decision to stop hydration more of a shared decision.

Dealing with conflict is a critical skill for physicians. Recognizing and dealing with conflict can improve relationships, shed light on complicated clinical situations, and help guide family members, patients, and other clinicians through difficult decisions. The rewards for physicians who acquire and use these skills are also evident: a grateful family and a personal sense of satisfaction.