Palliative and end-of-life (EOL) care, once associated exclusively with hospice settings, is now becoming an integral component of acute care (Goldberg, 2004; Manfredi et al., 2000). Early recognition of those patients for whom aggressive treatment may not lead to a cure or an acceptable quality of life provides an opportunity for discussion with the patient and family about advance directives and EOL care options. Timely initiation of the discussion and decision making regarding the intensity of care (e.g., change in code status) can prevent needless treatment and suffering for both the patient and the family. Within a nurturing and supportive environment, patients and families can maintain control and optimize physical and emotional comfort and dignity. Because most patients in the neuroscience intensive care unit (NICU) are unconscious, the nurse communicates with family members who are the surrogate decision makers for the unconscious or cognitively compromised patient.
Studies examining the involvement of critical care nurses in the process of EOL decision making (e.g., withholding or withdrawing mechanical ventilation, forgoing cardiopulmonary resuscitation) are few (Baggs & Schmitt, 2000). A disappointing finding in the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment (SUPPORT) was that intensive care unit (ICU) nurses were not very involved in decision making about the level of treatment for patients and that the nurses expressed frustration about their limited role in this decision-making process (Kirchhoff & Beckstrand, 2000; SUPPORT Principal Investigators, 1995). Baggs and Schmitt (1995) found ICU nurses to be less satisfied than medical residents with decisions regarding the level of treatment aggressiveness. Twenty-eight nurses working in a medical ICU expressed being morally distressed when they provided aggressive care to patients whose treatment appeared to be futile (Elpern, Covert, & Kleinpell, 2005). Six critical care nurses said that their primary responsibilities at the end of a patient's life are to support, coordinate, and educate (Jezuit, 2000). A study of 12 nurses who cared for chronically ill ventilator-dependent patients revealed the nurses' vulnerability, suffering, and moral angst (Ray, 1998). Poor communication about prognosis, resuscitation, and patient preferences were important components of the problem (SUPPORT Principal Investigators). Some nurses reported that they were required to carry out a treatment plan that they believed was harmful or inappropriate and that the quality of care negatively affected the patients' well-being (Gaul, 1995). Critical care nurses described the transition from curative to palliative care as awkward (Kirchhoff et al., 2000) and discordant (Halcomb, Daly, Jackson, & Davidson, 2004; Jezuit; Kirchhoff et al.).
Few studies address this critically complex phenomenon. During the National Institutes of Health Stateof-the-Science Conference on Improving End-of-Life Care in December 2004, "little evidence was provided regarding the experiences of professional caregivers at the end of life" (National Institutes of Health, 2004, p. 7) because of a lack of research in this area.
Nurses express their moral distress and discomfort when faced with EOL issues in their daily practice. The investigators designed this study as a formal investigation into NICU nurses' perceptions regarding their roles and responsibilities in the decisionmaking process during the change in intensity of care from aggressive care to EOL care for patients.
Method
With this qualitative descriptive study, the investigators did not intend to generate a theory; however, this study contains overtones consistent with grounded theory methodology (Sandelowski, 2000). For instance, the authors' experiences with grounded theory resulted in codes consisting primarily of gerunds (i.e., words ending with "ing"), which is the coding scheme consistent with grounded theory. The product of a qualitative descriptive study is a complete, organized, descriptive summary of the data (Sandelowski). Institutional review board approval was obtained prior to the start of the study, and approved flyers were posted in the NICU inviting the 30 registered nurses to participate.
Sample and Data Collection
The first 12 registered nurses who volunteered to participate made up the sample for this study (Table 1). Private audiotaped interviews were conducted with NICU nurses employed at one hospital. One researcher obtained the nurses' informed consent and conducted all first interviews. Participants were asked interview questions that matched the purpose of the study. These moderately structured questions (Sandelowski, 2000) were generated by the investigators (Fig 1). Consistent with qualitative methodology, the term end of life was not defined by the investigators; rather, the participants spoke freely about what it meant to them. The interviews lasted approximately 1 hour. Checks with participants were used to establish the credibility of the findings (Lincoln & Guba, 1985). For example, during a second private audiotaped interview, each participant read and reacted to a one-page summary of the findings. Participants endorsed the synopsis or made clarifications and offered additional examples to support the existing data. Field notes primarily about the content and tone of the interview were documented after each interview for the purpose of contributing to an audit trail, thereby enhancing the trustworthiness of the data (Lincoln & Guba).
Data Analysis
The recorded interviews were transcribed verbatim by a transcriptionist, and the transcripts were then verified by the investigators for accuracy. Transcript data were content analyzed line by line, and codes (key conceptual words or phrases) were noted in the margins, creating a label for each sentence (Glaser, 1978; Lincoln & Guba, 1985; Sandelowski, 2000). As codes emerged, they were compared with codes from earlier interviews to determine similarities and differences. By the 10th interview, no new codes emerged, so saturation or redundancy among the data was reached (Glaser & Strauss, 1967). The last two interviews verified the data saturation. The three investigators met on five occasions to discuss the codes until consensus was reached. A word-processing program was used to manage the data. The codes were grouped and labeled using broader categories, yet the terms remained consistent with those used by the nurse participants (Sandelowski). The data were then organized to describe the nurses' perceptions about EOL care in the NICU.
Findings
Nurses described their perceptions related to caring for patients who are approaching the end of their lives, a time when the nurses communicate primarily with family members and physicians. The communication the nurses described involved providing guidance while being positioned in the middle of the communication process and feeling emotions. Emotions covered the gamut from confusion, helplessness, and frustration to feeling emotionally drained, being overcome by very mixed feelings, as well as feeling privileged.
Providing Guidance
The NICU nurse participants guided the patient and family members as part of the educator role. To best guide family members, the nurses artfully elicited the family members' thoughts. The nurses trusted that family members knew the patients, and the nurses listened as family members weighed the opinions of the physicians.
One nurse said the following:
Sometimes what I will do is talk to the family and try to feel them out. I get them to talk by asking, "What did the doctor tell you?" and "What do you think?" I had a patient's family ask, "What do you think I should do?" I said, "I can't make that decision for you, but have you [previously] discussed this with your husband (or your brother or your sister)?" I had patients' family [members] say [of the patients], "They never want to do this. They never want to be on this." I asked, "When you say they don't want this, what do you mean?" I try to see where they are coming from. When [the family member] said that the patient doesn't want to be on a ventilator or doesn't want a breathing machine, I ask, "Did you talk to the doctor about it when [the patient] became ill?"
Another nurse said,
Listen to what the [the family members] have to say and just back off and try to let them have space. Let them open up to you, and you [may find out] just why they are so angry and frustrated. That also helps.
During the EOL decision-making process the nurse guided the family by pushing or urging them to move forward with a decision. For example, one nurse said, "We talk to the family about what they think the patient wanted, and if I have a feeling that this is not what the patient wants, then I really try to push to change the code status[horizontal ellipsis]. We work on them [family members] until they get to where we think they need to be." Another nurse stated, "I will push and educate and talk with the family until I can feel resistance." Similarly, another nurse (without using the word "push") said, "I explain to them what each code status means, and sometimes I'll be a little more dramatic than I probably need to be if I think that they [family members] are kind of on the edge." One nurse used the word "shoving" in her description.
People get angry with you. People don't want to hear what you have to say [about the patient approaching the end of life]. They don't want to believe it. Sometimes you feel [as though you are] shoving reality in their faces, which is difficult. Trying to be the third force intermediary is not a pleasant place to be, although, when it works out, it's a good thing because it opens up the line of communications.
Amid the "pushing" and "shoving," the nurses communicated with the patients' loved ones, and connections were made while they were caring for patients and guiding family members. One nurse said, "When you are dealing with these issues, you get closer. You become like family among them." The nurse also said,
If I don't get to experience some of the networking with the family [horizontal ellipsis] then I don't feel that close connection with the family and with the patient. When the [patient] dies, you do feel a sense of loss. I enjoyed being a part of the process[horizontal ellipsis]. You need and you want to be part of that experience[horizontal ellipsis]. I'm not so connected that [I] cry over every single patient that dies, but just connecting [brings] a sense of respect, caring, and compassion.
When family members sorted out their feelings, and perhaps second-guessed their decisions, the nurses were there to answer questions and provide guidance. One nurse said,
You become a confidant. You become a wall for the family to lean on. I think sometimes the [family members] look up to you to give them reassurance [that] whatever decision they make is okay. I think they want [the nurse's] approval about whether to withdraw care or to [continue] the care.
The nurses learned firsthand from family members whether they were all in agreement with treatment decisions and whether they were ready for a change in the intensity of the patients' plans of care. The nurses found out about the families' hope for the patients' recovery and considered it important to let the families "come to terms" with status changes.
Being Positioned in the Middle of the Communication Process
"Being in the middle" meant that the nurses were positioned at the hub of the communication process. The nurses became the intermediaries, "the communication person in between," or translators for the patients' family members and physicians. One nurse said, "Sometimes families are afraid to talk to the physician. Sometimes they don't know what to say to the physicians. They might not understand exactly what the physician is saying, they didn't want a translator, and they feel like you're the safe person to talk to, and [you can] explain it a little bit better."
Another nurse described being a translator of healthcare information.
Translate what really happened with the patient-the tests they did and what the tests mean. [Physicians] have such little time with family. [Nurses tell family members] what's happening and what to expect. A lot of times [family members] just beg you for information, and you're the one that relates information in a humane way.
The nurses' goal was to facilitate communication between the patient's family members and physicians so that the patient's plan of care would be understandable and acceptable to all. The nurses advocated for the patients. As one nurse stated, "We [nurses] are the eyes and ears for the patient." The nurses orchestrated the patients' care and suggested patient care conferences to enhance communication. The nurses generally liked to be involved in the process. One nurse said,
Patient care conferences usually seem to work, especially if you can be there and put in your two cents and say, "What about this?" It seems to go pretty well when [the nurse] can coordinate [the conference].
The nurses' conversations with the patients' families consisted of clarifying and verifying patient wishes for EOL care. One of the greatest challenges for the nurse in the middle of the communication process was trying to reconcile a patient's written advance directives or previously expressed wishes and the family's wishes. These conversations between nurses and family members were often guarded. The following quotes exemplify the nurses' cautious interactions while being in the middle of the communication process with families and physicians:
[I] have to be really careful and sometimes not offer information unless I see that it's absolutely necessary, or if they come and ask me. I don't want to take the [family members'] hope away, but at the same time, I don't want to be unrealistic.
As a nurse, there's only so far that I can go. [I am] an advocate, but as an employee of the hospital, I can't say to the family, "You don't need this." You have to watch out how you handle that kind of situation. About all I can do is make suggestions to the physicians if they're approachable. [As far as the] family, you just have to feel them out because sometimes those kinds of things can backfire.
The nurse's conversations with physicians were often guarded as well.
I'm not at that comfort level where the [physicians] think of me as their peer who can help. I will keep my mouth shut around here for maybe another few years, and then [it will be] okay. It's funny how your coworkers also give you input. I know a few coworkers who are very confident in [giving input], but others are more [inclined to] refrain from giving input because they're afraid that the family will say something to the doctor. If [the suggestion is] not something that the physician is also thinking, he may say, "Why did you say that?" [Nurses] are afraid of the consequences of giving their input.
Usually, I don't confront the physician that is so intent on keeping [the patient alive]. [If I am] in a circumstance where the [patient's] arm moved [in response to nail-bed pressure], I'll talk to the physician with the family. I won't say anything to make [the physician] defensive, but I'll say, "We've given the patient 3 weeks, and now he moved his hand when you pinched him. Does that mean that he's going to be ready to go to rehab next week, or are we still going to need the trach and the PEG?" I make [the physician] try to think about the whole picture [concerning] when it's time to let go, [when] enough's enough.
One nurse communicated observations about a patient's failing organ systems to the patient's family members and physicians and then suggested a change in code status. Another nurse initiated a discussion about a change in code status when she felt that they were "not making any progress anymore with the patient's care. We're doing everything and [the patient is not] getting better, and it's been days. The prognosis is not good, and we're just doing futile care rather than comfort care."
One nurse saw signs of impending death and pointed them out to family members. Another nurse told the daughter of a patient,
If she [the patient] is not getting better and not waking up, then that's a sign that her brain is not functioning, that it's been squashed too long, and that should tell you that you need to take Mom home and love her and make those last days of her life more comfortable than being poked and prodded.
The nurse recognized the responsibility of the physician as the final decision maker regarding EOL care and upheld the physician's decisions. The nurse also knew the importance of conveying a consistent message from healthcare providers to the family.
It helps when all of the doctors are on the same page. It helps a lot if everyone has the same idea, and they are telling the families the same thing. It's difficult when they don't. [There is a] mix of emotions, and families don't want to give up if one doctor is saying [the patients] are getting better even though [the physicians] don't feel that [the patients] are better. It helps when the doctors are willing to approach code status because it is a lot more authoritative than when it comes from [nurses]. We can plant the seed, but it's basically up to [the physicians] to close the deal on what's going to happen. It helps if they have talked about [whether the patient] has a living will, if the family members don't have any bad dynamics among themselves, and if they all feel the same way about the quality of life.
The nurses in the middle of the communication process appreciated having physicians be straightforward with them.
The ones who are more helpful are the ones who are direct and say, "This is where we are with this patient, and this is what I feel might happen." Sometimes if they are clear with what they think, it helps the family make a quicker decision. [Physicians] give the families a little bit of hope, and the [family members] will grab onto any hope because that is really what they are looking for at that particular moment. The [physicians] can be very helpful if they are very direct. [They should] give the family the information they need to help them understand.
Nurses who cared for the NICU patient approaching the end of life provided guidance from the middle of the communication process. Communication, although challenging, was paramount at this time. Nurses experienced a range of emotions and personal feelings that were associated with the role.
Feeling Emotions
Not all of the nurse participants expressed the same feelings about caring for patients at the end of life. One nurse found caring for a patient at the end of life challenging yet embraced the responsibility. Comfort levels in caring for patients at the end of life differed from nurse to nurse, as did the emotional connection to the patients and families. The nurses were aware that the end of life for patients is steeped in uncertainty. The unpredictability of a patient's course was a source of confusion and a feeling of helplessness as well as an emotional drain for the nurse and family members. The nurse in the following discussion acknowledged the tremendous amount of energy expended when a nurse chooses to be involved in the EOL decision-making process.
Some [nurses] just do what [they] have to do, are task oriented, and leave the main decision making to the physician. I think that's wrong. Nurses are hesitant to initiate a [change in code status] because then you have to get the family [to talk], and the family may not be cohesive in their decision. I think when you make these EOL code status change decisions, it's a lot of emotional turmoil. I think nurses sometimes don't want to deal with it because it's not that simple.
When the patients were at the brink of death and the loss of personhood was obvious to the nurses, family members were often a source of frustration to the nurses. One nurse said, "Why can't they just get it? I tend to become frustrated with families that push and push and push in a helpless case." One nurse wanted to ask family members, "How much further do you want to push for your mom, dad, or husband?" Keeping the patients alive for the sake of the families and disregarding the patients' wishes was disturbing to the nurses. When nurses' unspoken opinions about the plan of care for patients differed from family members' expressed wishes, the nurses masked their frustration. For example, one nurse stated, "But I know it is that person's family, not mine, and so I just deal with it."
The NICU nurse works in a hospital environment where the expectation of patients, family members, and healthcare providers is that medical treatments will be optimized and that patients will be cured. Of course, not every patient can be cured, and a patient's ultimate demise may become inevitable. The nurses in this study had very mixed feelings about this realization. On one hand, they had hoped to see the patient recover, but on the other hand, the nurses recognized that allowing death with dignity would spare the patients more pain and suffering. The nurses expressed the stress and difficulty of making cognitive, emotional, and physical changes when moving from the curative mode of care to the comfort mode. The patients' situations prompted in the nurses emotions of sadness and thoughts of their own life and death experiences, but the nurses were not able to give much attention to how they themselves were coping with the experience, primarily because of a lack of time. For example, one nurse said, "You are dealing with a whole lot of grief [horizontal ellipsis] and at the same time asking, 'How quickly can I move this along so that I can get the next patient in?'"
The nurses witnessed the patients' and family members' pain and suffering at the end of life, and the nurses, too, felt pain. They described the emotional pain and what it felt like to watch families grieve and make EOL decisions as "breaking my heart," "feeling the hurt," and "heart-wrenching."
[Nurses] cry a lot. You're pushed in different [directions], and you [ask yourself], "Is this what that person wanted?" You want to do the right thing for that patient. Sometimes we have strife between the family members. Sometimes they ask you, "If it were your family member, what would you do?"
One nurse expressed the powerlessness of everyone-family members, healthcare providers, clergy-facing a patient's impending death in one situation: "At some point, no matter what we do, nature will finally take over, and that was exactly what happened. There wasn't anything we could have done to make [the situation] any different." Acceptance of the inevitable was essential for the NICU nurse.
Despite the challenges and difficulties of caring for patients and their families at the end of life, the nurses expressed feeling worthwhile and privileged. Caring for a patient approaching death made the nurses think, "What if this were me?" One nurse stated that the self-reflection prompted by being with a patient near death "enriches my life." Another nurse said, "When somebody is dying and you help him die with dignity, dyspnea- and pain-free, that is very gratifying."
When you've expedited the family's ability to make a decision, you feel good, because you facilitated some sort of forwardness or closure. As a nurse, you do that. You make a difference in the life of that patient.
Discussion
The findings of this research contribute to the understanding of NICU nurses' roles and responsibilities regarding the EOL decision-making process for patients, and these findings primarily support extant EOL care literature. Family members (Steinhauser et al., 2000) and physicians (Li, 1998) perceive themselves to be patient advocates, as did the nurses in this study. Rushton (1995) defined advocacy as "acting to safeguard and advance the interests of another" (p. 388). This general definition is consistent with the way the nurses in this study spoke about advocacy. The nurse's advocacy role is unique as compared to those of family members and physicians because nurses are nonrecipients of care who are situated in the middle of the communication process. Family members are advocates, but they also receive care from the nurse, so they may have competing needs or loyalties that strain their relationship with the nurse. Physicians are nonrecipients of care, but because of their sporadic presence at the bedside, they are buffered from the ongoing advocacy issues.
Nurses in this study expressed the need to push family members forward to make a decision. Norton and Bowers (2001) found that nurses attempt to "shift" family members' understanding of the patient's overall poor condition to fit the nurses' understanding (p. 263). In other words, nurses who worked with patients who were approaching the end of life tried to change family members' choices for curative treatment to choices for palliative treatment (Norton & Bowers). Like nurses in this study, other nurses have reported that they sometimes find it necessary to explain in simple terms what physicians attempted to convey to patients and family members (Kirchhoff et al., 2000; Oberle & Hughes, 2001). Nurses also reported grappling with contradictory information from physicians (Kirchhoff et al.), and nurses in this study voiced the importance of conveying a consistent message. When surveyed regarding activities that help in providing EOL care, critical care nurses ranked highest the item "all physicians agreeing about direction of care" (Kirchhoff & Beckstrand, 2000, p. 100), a sentiment echoed by nurses in this study. In another study (Curtis et al., 2001, p. 44), nurses ranked highest two physician skills that are crucial to patient care at the end of life: communication with patients (e.g., speaks to patients honestly and straightforwardly) and team communication (e.g., assists family in obtaining "consistent information" from all team members).
Pochard et al. (2001) found that 69% of family members visiting critically ill patients had symptoms of anxiety and 35% had symptoms of depression. Nurses who work with those who experience such anguish, as in this study, reported pain associated with witnessing the suffering of family members whose loved one approached the end of life.
Critical care nurses have described their own suffering surrounding EOL decisions (Jezuit, 2000; Ray, 1998). Like nurses in our study, nurses felt "frustration," "sadness," and "helplessness" (Jezuit, p. 147). Other nurses had moral distress related to their structure-imposed powerlessness in easing patient suffering (Oberle & Hughes, 2001). Further, Varcoe et al. (2004) described nurses' "conflicting loyalties" and feelings of "being caught in-between family members and patients [horizontal ellipsis] [and] between patients and other health care providers" (p. 321). Critical care nurses in another study described feeling "tension," "stress," and "grief" when caring for patients at the end of life (Halcomb et al., 2004, pp. 217, 219).
The NICU nurses in this study described their frustration with family members. In a similar way, other critical care nurses ranked highly family members' lack of understanding of "life-saving measures" and lack of acceptance of the patient situation as "obstacles to providing end-of-life care" (Kirchhoff & Beckstrand, 2000, p. 99). Other nurses found that the shift from curative to comfort care was difficult (Kirchhoff et al., 2000), as did the nurses in this study. Critical care nurses also spoke about feeling "a sense of satisfaction and reward" (Kirchhoff et al., p. 41), similar to the nurses in this study who felt privileged.
Nurses in this study used the word "push" when they described how they guide family members, which may unfortunately suggest pushy or manipulative behavior. They also used the phrases "withdraw care" and "futile care" instead of "withdraw treatment" and "futile treatment," which would have been more precise and less open to misinterpretation. Despite the nurses' word choices, the investigators gleaned from the interview data that nurses interact with family members in a nonaggressive, peaceful, and caring manner.
We surmised that the interviews themselves were perhaps therapeutic for the nurses; they felt comfortable expressing their feelings. Nurses have no time to come to some level of comfort when dealing with EOL patient situations. They mask their feelings so that they can address the matter at hand, which leaves them with unprocessed emotions. Nathaniel (2006) suggested that the facilitation of moral reckoning by nurses (e.g., recounting their stories, investigating conflicts) provides a means of minimizing moral distress.
A limitation of this study is that only NICU nurses at one hospital who were motivated to share their perceptions participated. We were unable to describe perceptions of nurses who chose not to participate. Another limitation of this qualitative descriptive study is that the findings cannot be generalized.
Implications
This study increases awareness of the NICU nurse's unique, central role in the communication process surrounding the patient's end of life. The study builds the small body of evidence regarding experiences of healthcare providers at the end of life (National Institutes of Health, 2004). Providing guidance and being in the middle of the communication process can be a lonely, challenging, yet rewarding position. What the nurse experiences in this role can be a source of distress. Results of this study can be used to support NICU nurses who care for patients at the end of life (e.g., by acknowledging nurses' distress and allowing them time away from the unit). Nurses should be afforded the opportunity to debrief after performing EOL care. Nurses may want to assert themselves in designing a model of interdisciplinary practice that will promote the best possible care for dying patients and their families and will mitigate the tremendous frustrations they experience. Nurses may also design interventions to foster better physician-nurse communication, regular patient care conferences, and family conferences.
Ambiguity surrounds the nurse's role at the end of life in the NICU, and future investigations might explore the processes and people involved. Future research should include elucidation of physicians' and family members' perceptions of their roles in the EOL decision-making process because they are also part of the communication process. Investigations that aim to understand perceptual incongruence between family members, nurses, and physicians at the end of life are needed (National Institutes of Health, 2004).
Providing guidance and being in the middle of the communication process can be a lonely, challenging, yet rewarding position.
Enhanced communication is vital to the provision of superior EOL care. The first step in bolstering communication is to identify and make the nurses' perceptions known, as we did in this study. After nurses' perceptions about the problem have been descriptively summarized, time and resources for nurses to provide their best EOL care can be aptly allocated.
This study helps to create a greater awareness of the nurse's choice of words. For example, use of the word "push" may serve as a reminder to nurses to avoid crossing the line and becoming pushy, authoritative, or manipulative as they communicate with patients and family members at the end of life. Also, the use of phrases such as "withdraw care" or "futile care" conveys an unintended message of abandonment and disrespect of patients. Nurses should instead employ more precise phrases such as "withdraw treatment" or "futile treatment" when discussing compassionate care for patients at the end of life.
Summary
This descriptive summary of NICU nurses' perceptions about caring for patients at the end of life revealed that the NICU nurse provides guidance, is positioned in the middle of the communication process, and feels an array of emotions. Limited empirical support exists regarding the perceptions of critical care nurses caring for patients at the end of life. This research confirms and contributes to building that body of knowledge. This study is unique because it is specific to NICU nurses, although the data are not able to be generalized. In its discussion of the concept of pushing forward, the study makes a unique contribution to extant literature and warrants future investigation. This study's findings may be used to increase awareness of the NICU nurses' experiences with patients at the end of life and garner support for the nurses.
Acknowledgments
We gratefully acknowledge financial support for this research from St. Luke's Episcopal Hospital, Houston, TX. We also thank all the dedicated nurses who participated in this study and express our gratitude for the editorial assistance provided by Sandra Hanneman, PhD RN FAAN, Jerold B. Katz Distinguished Professor for Nursing Research, at the School of Nursing at University of Texas Health Science Center, Houston, TX.
References