Nurses' Experiences of End-of-life Care in Long-term Care Hospitals in Japan: Balancing Improving the Quality of Life and Sustaining the Lives of Patients Dying at Hospitals.

Asian Nursing Research xxx (2017) 1e9

Contents lists available at ScienceDirect

Asian Nursing Research journal homepage: www.asian-nursingresearch.com

Original Article

Nurses' Experiences of End-of-life Care in Long-termeCare Hospitals in Japan: Balancing Improving the Quality of Life and Sustaining the Lives of Patients Dying at Hospitals Ryo Odachi, RN, MA, 1, 2, * Tomoko Tamaki, RN, MA, 2, 3 Mikiko Ito, RN, Ph.D., 4 Taketoshi Okita, Ph.D., 5 Yuri Kitamura, M.D., Ph.D., 6 Tomotaka Sobue, M.D., Ph.D. 6 1

Department of Nursing and Laboratory Science, Graduate School of Medicine, Yamaguchi University, Ube, Yamaguchi, Japan Division of Health Sciences, Graduate School of Medicine, Osaka University, Suita, Osaka, Japan Mukogawa Women's University, Department of Nursing, School of Nursing, Nishinomiya, Hyogo, Japan 4 Department of Nursing, Shiga University of Medical Science, Otsu, Shiga, Japan 5 Department of Medical Ethics, Tohoku University Graduate School of Medicine, Sendai, Miyagi, Japan 6 Division of Environmental Medicine and Population Sciences, Department of Social and Environmental Medicine, Graduate School of Medicine, Osaka University, Suita, Osaka, Japan 2 3

a r t i c l e i n f o

s u m m a r y

Article history: Received 27 February 2017 Received in revised form 3 August 2017 Accepted 3 August 2017

Purpose: In Japan, about 80% of deaths occur in hospitals, especially long-term-care beds. The purpose of this study was to clarify the nursing practices used for such older patients at the end-of-life stage in longterm-care wards via the modified grounded theory approach (M-GTA). Methods: Data were obtained through semi-structured interviews of nineteen nurses working in cooperating long-term-care wards, acute care wards, or hospice services (to allow for constant comparison between these types of wards) in western Japan in 2014. We analyzed the transcribed data using M-GTA. Results: The core category that emerged from the analysis was “Balancing enhancement of patients' daily life quality and life-sustaining care in the face of uncertainty about the patients' character.” 11 categories emerged, such as Seeking older patients' character with their family, Supporting families' decision making, Rebuilding patients' daily life in the ward, and Sustaining patients' life span through medical care. Conclusions: Nurses experienced uncertainty about the care needs of older patients, the ethical problems of Enhancing the patients' QOL by using risky care, and the evaluation criteria used to judge their own nursing care after the patients' death. All nurses had the goal of ensuring a natural death for all patients. Nurses' acceptance and evaluation of their own care was critically influenced by the patient's family's responses to their care after patients' death. Further research is necessary to develop evaluation criteria and educational programs for end-of-life nursing care of older adults. © 2017 Korean Society of Nursing Science, Published by Elsevier Korea LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: terminal care aged long-term-care qualitative research

Introduction The rapidly aging population of developed countries such as Japan [1] has made end-of-life care for older adults more relevant than ever before, particularly the place where these older adults

* Correspondence to: Ryo Odachi, RN, MA, Department of Nursing and Laboratory Science, Graduate School of Medicine, Yamaguchi University, 1-1-1, Minamikogushi, Ube, Yamaguchi, 755-8505, Japan. E-mail address: [email protected]

wish to die [2,3]. Approximately, 60% of Japanese older adults wish to die out of the hospital, and most older adults hope to die at their own home [4]. Despite this, approximately 80% of all deaths in Japan occur in hospitals [5]. Although the Japanese Ministry of Health, Labour and Welfare has aimed to improve the prevalence of home healthcare at the end of life [6], the dehospitalization of older adults has been unsuccessful. Nurses working in general wards that do not have palliative or hospice care functions tend to encounter serious problems during their older patients' end of life. For example, many are unable to provide ideal nursing care because they feel exhausted and

http://dx.doi.org/10.1016/j.anr.2017.08.004 p1976-1317 e2093-7482/© 2017 Korean Society of Nursing Science, Published by Elsevier Korea LLC. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Odachi R, et al., Nurses' Experiences of End-of-life Care in Long-termeCare Hospitals in Japan: Balancing Improving the Quality of Life and Sustaining the Lives of Patients Dying at Hospitals, Asian Nursing Research (2017), http://dx.doi.org/ 10.1016/j.anr.2017.08.004

2

R. Odachi et al. / Asian Nursing Research xxx (2017) 1e9

powerless in the face of patients' deaths. This suggests that nurses' skills and mentality can present many barriers to developing quality end-of-life care in general wards [7,8]. Problematic nursing care at the end of life is particularly serious in long-termecare wards. Long-termecare wards were created in 1992 in Japan as an effort to improve the care of older patients requiring long-term medical treatment and nursing. In 2001, longterm-care was defined by the Medical Care Act as “any hospital bed or clinical beddother than psychiatric hospital beds, infectious diseases beds, or tuberculosis care bedsdthat is primarily for the hospitalization of patients requiring long-term recuperation” [6]. Most inpatients are aged 65 years or older, and around 40% of them are discharged by death [9]. Although there might be a need for effective end-of-life care in these wards, previous studies [9,10] have reported that palliative care is not commonly provided in long-termecare wards when compared when general wards, home-visit nursing services, and long-termecare facilities. Furthermore, staff at these long-termecare wards have noted that the end-of-life care delivered in these wards is insufficient. This deficiency is likely related to two other characteristics of long-termecare wards. First, most of the geriatric patients within these wards have multiple diseases or complex conditions. Second, the wards are meant to serve a support function after patients have undergone advanced medical treatments, such as continuous monitoring and management for central venous nutrition, ventilators, chronic obstructive pulmonary disease, dialysis, delirium, and depression. These two characteristics indicate that patients in long-termecare wards have complex clinical features [11]. Another reason is that nurses' workload in these wards seems heavier and they tend to have more responsibilities. For example, the current guidelines for nursing personnel distribution in long-termecare wards, as dictated by the Medical Care Act, indicates that each nurse should care for as many as 20e25 patients, whereas general ward nurses are advised to care for only around 7e15 patients. Confounding the above situation is the lack of research relating to end-of-life care in long-termecare wards in Japan, even though these wards play an important role in providing end-of-life care and a place to die for older patients. Although providing end-of-life care for older patients in long-termecare wards has several problems, the specific reasons have not been clarified. To improve patients' end-oflife care in long-termecare wards, nursing situational practices and why nurses feel that their care is insufficient requires analysis. Consequently, this study clarified the nursing practices for terminally patients in long-termecare wards and the relationship between the nursing practices and their evaluation after patients' death. We adopted the modified grounded theory approach (MGTA), because qualitative methods are usually used when little is known about a phenomenon [12]. Furthermore, the M-GTA is a suitable method for cases with process characteristics [13].

die. According to a previous study, understanding the death of a patient is crucial for everyone involved, including the patient, their family members, and the medical staff [12]. By adopting this perspective, we can study older adults at their end of life in longtermecare wards. It should be noted that we did not discriminate between type of disease or treatment. Similar to other studies, we also did not differ between beds covered by medical insurance and long-termecare insurance [15]. Setting and participants There are sampling strategy discrepancies between the GTA and the M-GTA. For example, the strategy of the GTA is to collect and analyze data in parallel and its process proceeds by comparing participants. On the other hand, the strategy of the M-GTA is to collect data for a specific group; then, the researchers analyze this “base data.” After this process, researchers collect additional specific data as needed [12]. This study recruited participants from three hospitals located in Western Japan who had agreed to cooperate. First, we recruited from hospital A, which is in an urban city of more than 800,000 people. It is a large hospital with more than 900 beds. Almost all its beds are long-termecare beds and it has a hospice and a general ward. We started at this large hospital because it enabled us to investigate nursing practices with diverse diseases and treatments. In addition, we could compare terminal care between the longtermecare beds and the hospital's other wards. Next, we recruited participants from hospital B, which was in a rural area and had about 50 beds. The reason for choosing this hospital was to examine the influence of the community and a smaller sized nurse organization. Hospital B also comprised a longtermecare health facility, which was for older adults who were in a similar condition as the patients in long-termecare wards. Finally, we selected hospital C, which was in a satellite city and had about 200 beds. It was an acute hospital. Most of the patients had cancer and received surgical treatments. This aim was to examine terminal care in a general hospital without a palliative care unit or hospice and compare it against our devised theory. We asked the nursing directors from each hospital to help recruit research participants. We made sure to recruit one nurse in a managerial position (e.g., head or chief nurse) and one general nurse. We included nurses from various positions to capture unique care experiences. For example, nurse managers must consider the occupancy rate for their wards, which could affect their vision for individual care. There are also differences in ability and authority regarding decision making. We anticipated, for example, that nursing managers could represent other nurses at important meetings, such as when addressing patients' families. Data collection

Method Design This qualitative study used the M-GTA [12]. The M-GTA was developed by Kinoshita, and generally adheres to the tenets of the original grounded theory approach (GTA), such as a constant comparative method and grounded data [14]. It is considered “modified” because it clarifies the method of segmenting the data during the analysis. Specifically, it posits that data segmenting is based on two concepts: the analysis theme and the individuals who are the focus of the analysis. In addition, we decided that the definition of “terminal stage” referred to when nurses determined that patients were going to

We collected data via semi-structured interviews conducted from February to September 2014. Each participant was interviewed once for about 60 minutes by two or three interviewers in a private room and ensured that their comment would remain anonymous. The main interviewer proceeded through the interview using an interview guide. The sub interviewer(s) took notes about participants' voices and reactions and asked participants additional questions (e.g., “why?” and “how?”) to obtain more detailed information. Although we used the interview guide to enhance efficiency and equity, we also asked participants to talk freely. The interview guide was created through discussions between clinicians who had experience with terminal care for older adults and researchers including nurses, doctors, and a clinical



psychologist. We began each interview by telling the participants, “We would like to know about your experience of caring for older patients at the end-of-life stage. Please tell us about your experiences of particularly notable case(s).” The major questions covered in the interview guide were about cases that participants had experienced. For example, “What did you do in this case?” “Do you have any impressive episodes about communications between you and the patient, the patients' family, or other staff members?” “How did you evaluate the case?” Before interviews, we requested that participants select the case(s) along the following criteria: (1) the patient had died in their ward and (2) he/she was an older adult. The interview data were recorded with participants' consent and transcribed verbatim later. If participants refused to the recording, the subinterviewers took notes during the interviews. We continued data collection until theoretical saturation had been achieved (i.e., no new concepts or categories emerged). Data analysis There are some differences in the analytical procedures of the M-GTA and the GTA regarding the perspective toward the data. The symbolic difference concerns segmented data. In the GTA, the aim of segmentation is to keep a distance between obtained data and researchers to avoid being influenced by participants' perspectives. However, the M-GTA asks a researcher to understand the participants' perspective; specifically, the M-GTA requires a qualitative analysis to interpret meaning. Therefore, in the MGTA, one does not use segmented data by predefined units such as a word or a sentence, allowing for a deeper interpretation of meaning [12]. First, we decided the analysis theme to determine the rage of using data about actions and emotions of the nurses for analysis; then, we decided the person to focus on for analysis to determine the person as the subject doing the analytical theme. The former, comprised “the experience of end-of-life care” and “the feelings and evaluation about patients' death.” The latter, comprised “nurses caring for older adult patients during their end of life at a long-termecare ward.” In the M-GTA, the minimum analytical unit is the concept. Each concept is derived from several pieces of data known as variations. Variations are collected along the analysis theme and transcribed verbatim and analyzed repeatedly to obtain their meaning. When a concept emerged, similar or antithetical data related to the concept were examined to prevent arbitrary interpretation. The relationships between concepts were examined and categories were deduced from the related concepts. Then, the relationships between the categories were examined, and the data comparison and analysis were repeated. A core category emerged because of comparing between the categories repeatedly. A core category explained the process about its analysis theme.

Ethical considerations We informed participants about the study and assured that the nature of participation was entirely voluntary and that the collected data would be anonymous. We also informed participants that they could withdraw at any time or refuse to have their data used in the research, even after interviews had been completed. This study was approved by the Osaka University Division of Health Sciences Ethics Committee (Approval no. 278). Results Participants' characteristics Nineteen nurses were recruited and interviewed (Table 1). All nurses were women and there were twelve nurses working in longterm-care wards. Although most of nurses were in their 30s, there were also nurses in their 50s. Fourteen participants had more than ten years of experiences working as nurses. In addition, general nurses were eight and the nurses in managerial positions were eleven. Characteristics of nursing practice in Japanese long-termecare beds Core category The core category was “Balancing enhancement of patients' daily life quality and life-sustaining care in the face of uncertainty about the patients' character.” This balance refers to nurses having to reconcile “Rebuilding patients' daily life in the ward” and “Sustaining the patient's life span through medical care”; however, it is difficult for them because of the uncertainly regarding patients' wishes and the evaluation of their practices after patients' death. Many of the nurses considered a “calm death” as the main criterion for a good death and it was important for them to clarify the patients' individuality and have interest in the patients' daily lives rather than sustaining their lives. Rebuilding patients' daily lives and managing risky environments were also reported. However, many patients in the wards had conditions such as somnolentia or dementia, which made them unable to express their wishes. In addition, a climate existed where nurses entrusted serious decisions to others, excluding the patients even when they could

Table 1 Participants' Characteristics (N ¼ 19). Characteristics Location

Ward

Rigor The criteria of trustworthiness, credibility, dependability, and transferability were used to ensure rigor [16]. All recording data were transcribed verbatim and researchers shared them. Because multiple researchers performed all interviews, transcripts were checked multiple times. The first author was the only one involved in the coding process, but the analysis sessions were conducted in conjunction with the remaining researcher. The discussions were held until all researchers reached agreement on the codes. We also showed informants the preliminary results of our analysis to obtain their feedback.

3

Age (yr)

Experience (yr)

Position

Classification

n

Rural Satellite city Urban Long-term-care Hospice Acute care General Long-termecare facility 20s 30s 40s 50s 20 General nurse Chief nurse Head nurse Certified nurse

3 3 13 12 2 3 1 1 1 8 4 6 2 3 6 8 8 3 7 1


4


express their will or opinion. Therefore, patients' individuality was uncertain in the wards. The remaining categories that emerged indicated that nurses tried to clarify the uncertainty and find a balance between the above two practices. There was additional uncertainty concerning the process results. For example, the family might not have a chance to share evaluations of nurses' practices after the death. Therefore, the evaluation is limited as it only concerns medical staff members, which also may affect nurses' sense of balance. We present the categories and concepts that make up this core category in Table 2, and the relationships between categories are illustrated in Figure 1. Seeking older patients' character with their family. These practices referred to collecting information about older patients to understand their character and estimate their needs through daily communication with their families on admission to the hospital. Through this, nurses built a comprehensive image of each older patient to provide the best possible care. Most patients can't talk. I get information from his family or visiting persons about what he used to be like. I care about information not only from the present, but also from the past and future, in a time series. (Chief, 30s) Nurses “extracted the patient's story from the family” to deduce patients' character and background. They also attempted to grasp the relationship between the patient and the family by considering the patient's life history. When the patient became unable to express his or her will because of impending death, nurses began “estimating older patient's needs with the family.” This involved discussing appropriate patient care, such as whether to provide life-sustaining treatment. Nurses often tried to match the family's care intentions with those of the patient; however, only when the patient could accurately communicate his or her wishes. Seeking older patients' character without their family. Nurses engaged in practices without family members to deduce patients' character. These practices involved supplementing information about a patient's character by interacting with the older patients themselves and other nurses in the ward. They did this not only when the patient had no family, but also for a unique interpretation by nurses only. We would deduce what the older patient's personality is from our own experiences. Further, we refer to the care workers in our team. For example, we often discuss about our patients such as “she keeps dressing neatly because she has tattoo,” and “her belongings look stylish,” and “she may like to apply makeup.” Therefore, we can grasp and listen to our patients' wishes through our conversations. (General, 30s)

Supporting families' decision making. This category concerned how nurses discussed the use of life-sustaining medical treatments (such as gastric fistula) with family members forced to make decisions about the future directions of medical treatment. These practices were directed toward the family members. The family tended to decide on important treatment-related matters on behalf of patients, especially patients without a clear capacity to make decisions. Nurses aided this decision making by “speaking about the patient's speculated intention,” such as telling them about the patients' pain or anxiety. Nurses imagined that it was difficult for patients to explain their intentions, particularly

given patients' severe conditions. However, they were discreet in providing their own opinions about medical treatment. After careful consideration of the family's own decision making, nurses judged the extent that they would inform the family of their own opinion as a nurse, often as delicately as they could. In the first half of my career, I made suggestions to the patient's family based on the viewpoint of the patient, instead of the family. But, after I had experienced this myself (My father was unwilling; but, I ended up deciding, as his family, that he should undergo a stomach wax), I learned that family members tend to choose lifesustaining treatment. This made my suggestions to (the) family shift to be more in line with their viewpoints. However, I have to be the person to tell the older patients, while still considering the family's feelings. My viewpoint is about 60% from patients, 40% from family. (Head, 50s) Nurses also “just provided information” to family members, and adopted an attitude of “supporting the family's decision making,” whereas avoiding letting their own personal ideas get in the way. Nurses' use of “just providing information” as a practice was influenced by their relationships with the patient's family, such as their degree of psychological closeness or the frequency that family members visited the hospital to meet the patient. If nurses could not build a good relationship with the family, they tended to “just provide information.” “Supporting the family's decision making” was accomplished either before the family had reached a decision or after. In the former situation, nurses attempted to prepare family members for their decision by presenting them with the reality of their situation; whereas in the latter, they tried to encourage family members to be satisfied with their decision. Facilitating decision making within patients' families. This category refers to how nurses begin talking to family members about end-oflife decision making and resolve conflicts among family members to help smoothen the decision making process. Attempting to begin talking could be done passively, such as in response to a question from the family, or actively through references to the death. I vaguely suggest to the family that the time for the patient's death is approaching by asking them, “Do you have the clothes for dressing the patient in the end?” (General, 30s) Nurses also “selected a proper negotiator in the family” when conflict arose among family members. The number of family members who participated in decision making tended to increase as the patients' death neared. When discussions became confusing, nurses would occasionally change the negotiator or appoint a new one(s). Rebuilding patients' daily life in the ward. The practices were especially important after nurses became aware of the patients' impending death and reflected nurses' struggles to ensure that patients could maintain daily life in the hospital and face death with dignity. Nurses first speculated about the patients' daily life from their inferences about patients' character; then, they “carried on with the patient's daily habits of the ward.” This involved replicating what the patient had done naturally and casually before their admission. (My patient's) family did not have any interest in doing something for the patient, but we played music CDs in spite of the family. Because hearing ability remains for a long time, all of the staff play music CD or talk to him whenever they visit his room. (Head, 40s)



5

Table 2 Categories and Concepts. Categories

Concepts

Seeking older patients' character with their family

Extracting the patient's story from the family Estimating patient's needs with the family Extracting the patient's character from the patient's appearance Discussing the patient's character with colleagues Speaking on patient's speculated intentions Just providing information Supporting the family's decision making Hinting about the patient's final hours to the family Selecting a proper negotiator in the family Carrying on with the patient's daily habits in the ward Healing patient's loneliness Permitting family members' exceptional demands Deviating from nursing treatment standards and placing responsibility on oneself Promoting withdrawal or withhold of life-sustaining medical treatment Carrying out the medical treatments to prolong life Securing the patient Relieving the patient's pain Bathing even in risky conditions Feeding the patient through the mouth despite the risk of mis-swallowing Sending patient home before death Monitoring patient's condition more strictly Taking advantage of their title of nurse manager Sharing the responsibility of risky practices Using objective persuasiveness Sharing awareness of dying through nursing duties Sharing awareness of dying through informal conversations Observing with high sensibility Regarding patient's emotional reactions as essential Exploring the validity of medical treatments Evaluating uncertainty

Seeking older patients' character without their family Supporting families' decision making

Facilitating decision making within patients' families Rebuilding patients' daily life in the ward

Sustaining patient's life span through medical care

Enhancing the patients' QOL by using risky care

Managing special risks

Sharing awareness of dying among nurses Gathering information to judge the feasibility of practices

Evaluating uncertainty Note. QOL ¼ quality of life.

Figure 1. Relationships between categories. Note. QOL ¼ quality of life.

Nurses also diverted patients' loneliness by visiting their beds frequently throughout the day. When family members asked nurses whether the patient could receive care or engage in an activity that deviated from the rules of the ward or treatment

regimen, nurses would occasionally “permit the family's exceptional demand,” reasoning that the patient was on his or her deathbed. For example, nurses might serve patients hospital meals outside of the designated period to ensure that patients could take


6


the meal with his or her family, or affixed the Japanese suffix “-chan” (generally used to refer to children) to the patient's name when referring to them. Nurses' acceptance of these deviations encouraged a more trusting relationship between the family and nurses. Nurses also “deviated from nursing treatment standards and placed responsibility on themselves,” such as by lending their own private property to patients. This occurred in the rural hospital that we studied, and the nurse involved said that the patient had been her neighbor and they had known each other before admission. “Promoting withdrawal or withholding life-sustaining medical treatment” was an aim to help patients achieve a “natural death.” The treatments they advocated for withdrawal or to withhold included drip and tube feeding after patients were unable to eat by mouth. For this practice, nurses often came into conflict with family members or doctors who were hoping to prolong the patient's life. To negotiate for the withdrawal or withhold, nurses would try to involve their managers or referred to patients' evident pain and the burden of medical treatment, such as consistently needing to insert needles into patients for intravenous drips or the need for sputum suction.

Sustaining patients' life span through medical care. “Carrying out medical treatments to prolong life” such as an intravenous drip, tube feeding, and sputum suction were performed. “Securing the patient” meant restraining their behavior by means of physical restraints or sedation. These were used to protect them from accidents in the ward. Overall, the practices emphasized restricting patient's daily life in the ward. The necessity of these treatments varied with nurses' perceptions of the burden and pain they brought to patientsdthat is, although some nurses said that these treatments were necessary despite the burden, other nurses believed that these treatments were not necessary because of the pain they caused in patients at the end of life. I do sputum suction; but, I know that it is painful for patients. If I don't do it, patients would be worse than they are now. (Head, 50s) I wonder how much it is necessary to give high-calorie transfusions for older patients at the end of life. They will die anyways because they are old enough. I do not think that they have to die in the hospital. In fact, it's better to be cared for by the family. My opinions might be behind the times… (General, 30s) However, nurses also aimed to ensure that they could replicate the patient's daily life without trouble by “relieving the patient's pain.” This was done through medication or nursing skills such as touching or rubbing the patient's body gently or changing their position.

Enhancing the patients' quality of life by using risky care. Even though they might endanger patients' lives, this practice was conducted by nurses and family to enrich patients' last days; they were typically based on inferences about patients' needs. This category was derived from “Rebuilding patients' daily life in the ward.” However, it was distinguished by the points as follows. Importantly, these practices were always carried out alongside “managing special risk.” The main feature of this practice was that it was accomplished even when nurses knew other, safer methods (e.g., bed bath, drip). Furthermore, nurses had to be careful to evaluate the effects of these practices on patients because they were often dangerous and patients could not often communicate how they felt. For this reason, nurses believed that such practices were important.

The (terminally) patient said, “I hurt,” but looks happy after bathing. Other staff and I also felt that it was a burden for the patient. (So) when monitoring (their) vital signs, I had them take a bath until the last minute, and while bathing, I determined that the patient's condition was tense and throbbing. (General, 30s) Managing special risk. “Enhancing the patients' quality of life (QOL) by using risky care” referred to taking steps to improve the QOL of at-risk patients, whereas this category referred to the special methods that nurses may need to implement to manage risky situations. For example, when engaging in “sending patients home before death,” nurses “monitored patient's condition more strictly” than usual and prepared for their return home several days in advance of it. In these cases, what was important for both patients and nurses was not the effect of the practice on patients' condition, but the mere realization of its necessity. Nurse managers often persuaded their colleagues, doctors, and patients' families to pursue the abovementioned risky care by “taking advantage of their title of nurse manager.” It may be because I was his primary nurse. If another nurse had suggested doing it, other staff would deny or wonder if it not appropriate. But, as a primary nurse, I can propose sending the patient back home. I don't know if this is good or not (ethically). (Chief, 30s) When nurses engaged in these risky practices, they told the doctors and their superiors about the challenges they might face (“sharing the responsibility of risky practices”), which helped them overcome those challenges. This was sometimes done before practice, whereas at other times it was done as an after-the-fact report. Of course, I told the doctor. I said, “I fed him a scoop (food) with the spoon.” He responded, “Alright, but I will pretend that I did not hear it.” (Chief, 30s) When conflict about whether to pursue risky practice arose among the staff, they relied on “objective persuasiveness,” such as measuring patients' vital signs. Sharing awareness of dying among nurses. This category dealt with how nurses ascertained other nurses' understanding of the awareness of dying and shared their own awareness to promote decision making and end-of-life care. “Sharing awareness of dying through nursing duties” refers to awareness obtained by transferring patients to another room to observe them more easily or discussing during routine conferences in the ward and “sharing awareness of dying through informal conversations” occurred during daily conversations between nurses in the ward. We have a room called the observation room in front of our staff station. When we perceived her condition as being very serious and unpredictable, we moved the patient to the observation room in order to observe them anytime from the station by opening the door or the windows. It also means that the patient can always perceive the presence of a nurse near her. (Chief, 40s)

Gathering information to judge the feasibility of practices. This category involved obtaining information from a variety of sources to judge the validity of their practices. To understand patients' responses to their practices, nurses observed older patients' often slight motor responses or reactions “with high sensibility.” Particularly at the end-of-life stage, nurses also “regarded patients' emotional reactions as essential,” including their moods and



expressions, rather than relying solely on objective numerical data about patients' physical conditions. The “exploring the validity of medical treatments” concept referred to how nurses considered to what extent the medical staff provided medical treatments in an objective way and with a regard for the balance between natural aging and the patient's will, reactions, and disease. Nurses engaged in such information gathering to clarify patients' characters. Although she was usually void of expressions, her muscular tone caused by contraction had relaxed, and her face became softer (when the patient had a bath). (Head, 40s)

Evaluating uncertainty: how do nurses perceive their practices?. It was common for nurses to claim that the criteria for evaluating nursing practices at the end of life were uncertain. Many of the nurses considered a “calm death” as the main criterion for a good death. However, nurses did not always realize that patients' death was the natural course, and some nurses felt that their own practice had been poor. In those cases, nurses reported that the family's response to the patients' death was an important criterion regardless of their impressions of their own practices. Indeed, it was difficult for nurses to judge whether their practices had been appropriate if they did not receive a response from the family. I have no idea what the best practices are. They held a funeral, and after they finished, all they could do (was) begin to process whether the patient's dying process was good or not. We can find out from the family's evaluation at the very end whether our practices were right or not. (General, 30s) Even when nurses believed that they had engaged in inappropriate practices, their impressions changed for the better when the patient's family talked to them about the patient's death (often just to thank the nurse). In this study, none of the nurses reported that the family had in any way disparaged their nursing. Some (bereaved) family members will come hug me, crying, and saying, “Thank you so much,” in the end. The first time I met her, we worried about our care, because she had an ideal vision of hospital care. The staff and I talked about how we were right, and the family agreed and showed gratitude for our care at the end. (Head, 50s)

Discussion This study described the characteristics of nursing practice in the long-termecare wards for older adults at their end of life. A prominent characteristic of such practices was uncertainty, including about the care needs of older patients, the ethical considerations of “Enhancing the patients' QOL by using risky care,” and the criteria used to judge their nursing practice after patients' deaths. A previous study [17] stated that uncertainty was defined as the inability to determine the meaning of illness-related events and uncertainty occurs in a situation where the decision maker is unable to assign definite value to objects or events and/or is unable to predict outcomes accurately. In most cases in our study, however, patients were not the decision makers and it was sometimes necessary to determine a decision maker. Although an above study presented that uncertainty has four aspectsdambiguity, complexity, lack of information, and unpredictabilitydin this study, there was another uncertaintydthe patient. Nurses were affected by having to judge the balance between their practices and evaluation after patients died. In addition, uncertainty was evaluated by inferences built on personality

7

dispositions. Nurses in long-termecare wards could try to obtain these perspectives through “seeking” with family or colleagues instead of dying patients. After evaluating, uncertainty was judged as either dangerous or an opportunity. When it was the former, nurses could focus on sustaining patients' lives; when it was the latter, nurses could attempt to improve patients' QOL. Although the end was adaptation or integration, as said by the uncertainty theory, it is unclear what the end in this study was [18,19]. There would be a shift in decision makers because of the uncertainty of patients' will and/or needs in our study. Therefore, it might be necessary to change strategies for trying to end uncertainty to include other perspectives, including those of the family or nurses. The nurses mentioned they were often unable to decide whether their care had been appropriate or not after the patient's death. It was unique among the nurses that, on receiving positive feedback, they tended to accept their practices, even when they were not sure that they had acted appropriately themselves. However, nurses in long-termecare wards tended to talk less about death when compared with nurses in other wards, despite their interest in the topic [10]. It therefore might be necessary to increase their opportunities to talk about death considering the uncertainty from various perspectives. A common issue for all the nurses involved in end-of-life care is finding a balance between prolonging patients' lives and improving their QOL. Thompson et al [20,21] described the process of end-oflife nursing in an acute care unit, claiming that the most important concept in this process is “facilitating and maintaining a lane change”. In this case, “lane change” meant the transition in thinking about acute treatment aimed at curing to actively pursuing comfort care. This dichotomy was similar to our study's concepts of “sustaining” and “rebuilding.” However, Thompson et al also explained that lane changes are irreversible, making them important to consider whether a lane change was completed or not. By contrast, the balance referred to in our study was implied to be changeable, given that the core category reflected how nurses were seeking it out. The likely reason for the malleability of this balance is not only the uncertainty of the patients' wishes and/or needs, but also the difficulty in predicting the patients' condition as a result of their long-term disease [22]. Another peculiarity of our study was that the nurses often communicated with family members rather than patients. Casey et al [23] pointed out that “knowing the person” was an important factor of dying well in acute care settings. Although this concept seemed similar to the “seeking” referred to in our study, Casey et al said that “knowing” required collection of information only from the patient. The process of “seeking,” by contrast, relied on the family and nurses' colleagues, which suggest that they play an important role as information sources as well. However, in “seeking,” nurses would probably face more uncertainty and would need to act in spite of it. In addition, a previous study reported that it was difficult for older adult patients and their caregivers to change awareness [24]. Nurses in long-termecare wards try to resolve uncertainty through “facilitating” or “supporting” because nurses want to perceive patients' deaths as professionals. However, this is difficult, as shown by the fact that nurses often talked about the communication gap between medical staff and family. A previous study also presented that it was difficult for family members to discuss end-of-life issues [25]. A system should be developed where nurses can discuss their patients and nursing practice more frequently and freely with patients' family members. Attempting to find this balance inevitably brought ethical conflicts. Furthermore, at the same time, they attempted “enhancing the patients' QOL by using risky care” and tried to


8


mitigate the ethical uncertainty through “managing special risk.” Glaser and Strauss referred to such risky care as “special treatments” in their work [14]. However, there were differences between their study and ours concerning the location's responsibility toward care. For example, in long-termecare wards, risky care was performed even when older patients' own wishes were unclear, and the responsibility was borne by the patients' family and medical staff. The “special treatments” often reduced the medical staff's burden, whereas risky care involved nurses voluntarily sacrificing their time and energy. “Enhancing the patients' QOL by using risky care” was only possible because of the culture of the long-termecare ward. Previous studies have reported that in acute wards, it can be difficult for nurses to justify having long talks with dying patients because their duties of addressing patients' physical needs take precedence [26]. Costello [27] noted that the hospital nursing culture is evidently influenced by the culture of the acute wards, whereas nurse managers tend to influence the cultures of the hospital wards. Although nurse managers might promote their subordinates to feel they are providing quality care, these feelings are likely to be influenced by managers' view of what is considered good practice or not. It is thus necessary for nursing managers to arrange for educational support for nurses in their ward to help deliver higher quality care for older people at their end of life. A previous study on long-termecare facilities outside of Japan showed that both patients' family and medical staff tended to think that the deaths of the older people for whom they care are positive experiences [28]. An important aspect of these positive impressions was maintaining effective communication between patients' family and medical staff. In Japanese long-termecare wards, the number of opportunities to communicate differs depending on patients' families because of long-term hospitalization. Although long-term hospitalizations enable family members and medical staff to build relationships of trust, family members may not realize the finiteness of patients' lives, which may result in infrequent family visits. Ensuring that families understand this is important, particularly because it allows nurses to set up regular meetings with patients' families and significant others. In addition, it is necessary for medical staff to maintain high-quality nursing records, as it allows them to grasp older patients' conditions clearly and sensitively, which often change slowly over time. Strengths and limitations The cases that nurses spoke in this study were ones that were particularly impressive to nurses, which helped ensure that the data were of a sufficient strength to clarify the features and problems relating to nursing practices in these wards. The data were saturated and analyzed by an interdisciplinary team as well. Despite these points, we need further research that includes bereaved family members and other medical staff (e.g., doctors). Conclusion The nursing practices directed at older adults in long-termecare wards at the end of life were characterized as “balancing enhancement of patients' daily life quality and life-sustaining care in the face of uncertainty about the patient's character.” All nurses shared the common goal of ensuring a natural death for all patients. Nurses' acceptance and evaluations of their own care were influenced by the family's responses after patients died. Further research is necessary to develop further evaluation criteria and educational programs concerning the nursing care of older adults at the end of their lives.

Conflicts of interest The authors declare no conflicts of interest. Acknowledgments This study was supported by JSPS KAKENHI Grant Number JP26671034. The authors would like to thank Dr. Yoshizo Minami and Dr. Chika Honda, Miss Tomoe Shiromoto, Mr. Yuki Maura for valuable advices and all the nurses who so willingly participated in this study. References 1. Cabinet Office, Government of Japan. Annual report on the aging society: 2014 [Internet]. Tokyo: Cabinet Office Japan; 2014 [cited 2016 Feb 9]. Available from: http://www8.cao.go.jp/kourei/english/annualreport/2014/2014pdf_e.html 2. Fukui S, Yoshiuchi K, Fujita J, Sawai M, Watanabe M. Japanese people's preference for place of end-of-life care and death: a population-based nationwide survey. J Pain Symptom Manag. 2011;42(6):882e92. http://dx.doi.org/10.1016/j.jpainsymman.2011.02.024 3. Bell CL, Somogyi-Zalud E, Masaki KH. Methodological review: measured and reported congruence between preferred and actual place of death. Palliat Med. 2009;23(6):482e90. http://dx.doi.org/10.1177/0269216309106318 4. Cabinet Office, Government of Japan. The consciousness survey on health to older adults in 2012 [Internet]. Tokyo: Cabinet Office Japan; 2013 [cited 2016 Feb 9]. Available from: http://www8.cao.go.jp/kourei/ishiki/h24/sougou/ zentai/index.html. Japanese. 5. Statistics Japan. Vital statistics [Internet]. Tokyo: Ministry of Health Labour and Welfare; 2015 [cited 2016 Feb 9]. Available from: http://www.e-stat.go.jp/SG1/ estat/GL38020101.do?_toGL38020101_&tstatCode¼000001028897 6. Ministry of Health, Labour and Welfare. Health and medical services. In: Annual Health, Labour and Welfare Report 2011e2012 [Internet]. Tokyo: Annual Health, Labour and Welfare; 2012. p. p28e37 [cited 2016 Feb 9]. Available from: http://www.mhlw.go.jp/english/wp/wp-hw6/dl/02e.pdf 7. Noguchi W, Matushima E, Matsushita T, Kobayashi M. The present state of “death with dignity” in Japan. J Nat Inst Public Health. 2006;55(3):208e12. Japanese. 8. Kamiyama C. Terminal care nurses' images of a “good death”. J Jpn Acad Nurs Sci. 2007;27(3):75e83. Japanese. 9. Mizuho Information and Research Institute. Report of a cross-sectional study on the deathwatch for older adults in long-term-care hospitals and facilities [Internet]. Tokyo: Mizuho Information & Research Institute; 2014 [cited 2016 Feb 9]. Available from: https://www.mizuho-ir.co.jp/case/research/pdf/mhlw_ kaigo2014_04.pdf. Japanese. 10. Takahara K, Takeda K. Attitudes of nurses to terminal care for patients at skilled nursing facilities: the aspect of the presence or absence of anxiety in terminal care. Kawasaki J Med Welf. 2014;23(2):285e90. Japanese. 11. Takehisa Y. The importance of the sanatorium-type medical treatment wards which performs the chronic period medical treatment. Nihon Ronen Igakkai Zasshi. 2011;48(3):239e42. Japanese. 12. Kinoshita Y. Modified grounded theory approach. Tokyo: Koubundou; 2007. p. 306. Japanese. 13. Pope C, Mays N. Qualitative research in health care. 3rd ed. London: BMJ Books; 2007. p. 1e11. 14. Glaser BG, Strauss A. Discovery of grounded theory: strategies for qualitative research. Mill Valley, CA: Sociology Press; 1967. 15. Ministry of Health, Labour and Welfare. The survey on the hospitalization of chronical care [Internet]. Tokyo: Ministry of Health, Labour and Welfare; 2005 [cited 2016 Feb 9]. Available from: http://www.wam.go.jp/wamappl/bb11gs20. nsf/0/86ff47e5ca7c3b65492570c4002539ee/$FILE/2-3,4.pdf. Japanese. 16. Lincoln YS, Guba EG. Naturalistic inquiry. Newbury Park, CA: SAGE Publications; 1985. 17. Mishel MH. Perceived uncertainty and stress in illness. Res Nurs Health. 1984;7(3):167e71. 18. Mishel MH. Uncertainty in illness. Image J Nurs Sch. 1988;20(4):225e32. 19. Mishel MH. Reconceptualization of the uncertainty in illness theory. Image J Nurs Sch. 1990;22(4):256e62. 20. Thompson G, McClement S, Daeninck P. Nurses' perceptions of quality end-oflife care on an acute medical ward. J Adv Nurs. 2006;53(2):169e77. http://dx.doi.org/10.1111/j.1365-2648.2006.03712.x 21. Thompson G, McClement S, Daeninck P. “Changing lanes”: facilitating the transition from curative to palliative care. J Palliat Care. 2006;22(2):91e8. 22. Lynn J, Adamson DM. Living well at the end of life: adapting health care to serious chronic illness in old age. Washington: Rand Health; 2003. 23. Casey D, Murphy K, Ni Leime A, Larkin P, Payne S, Froggatt KA, et al. Dying well: factors that influence the provision of good end-of-life care for older people in acute and long-stay care settings in Ireland. J Clin Nurs. 2011;20(13e14): 1824e33. http://dx.doi.org/10.1111/j.1365-2702.2010.03628.x


R. Odachi et al. / Asian Nursing Research xxx (2017) 1e9 24. Freid TR, Bradley EH, O'Leary J. Changes in prognostic awareness among seriously ill older persons and their caregivers. J Palliat Med. 2006;9(1):62e9. http://dx.doi.org/10.1089/jpm.2006.9.61 25. Rafferty KA, Cramer EM, Priddis D. Managing end-of-life uncertainty: applying problematic integration theory to spousal communication about death and dying. Am J Hosp Palliat Care. 2016;33(1):69e76. http://dx.doi.org/10.1177/1049909114550675 26. Clarke A, Ross H. Influences on nurses' communications with older people at the end of life: perceptions and experiences of nurses working in palliative care

9

and general medicine. Int J Older People Nurs. 2006;1(1):34e43. http://dx.doi.org/10.1111/j.1748-3743.2006.00006.x 27. Costello J. Nursing older dying patients: findings from an ethnographic study of death and dying in elderly care wards. J Adv Nurs. 2001;35(1):59e68. http://dx.doi.org/10.1046/j.1365-2648.2001.01822.x 28. Sloane PD, Zimmerman S, Hanson L, Mitchell CM, Riedel-Leo C, Custis-Buie V. End-of-life care in assisted living and related residential care settings: comparison with nursing homes. J Am Geriatr Soc. 2003;51(11):1587e94. http://dx.doi.org/10.1046/j.1532-5415.2003.51511.x


Improving end-of-life care in hospitals: a qualitative analysis of bereaved families' experiences and suggestions.

Acute hospitals recognised for quality of end of life care.

Cultural experiences of immigrant nurses at two hospitals in Chile.

A Trial of electronic surveillance feedback for quality improvement at Nurses Improving Care for Healthsystem Elders (NICHE) hospitals.

The medicare flex program: a partner in improving quality of care in critical access hospitals.

Improving Midwifery Care in Ugandan Public Hospitals: The Midwives' Perspective.

Intensive care nurses' experiences of end-of-life care.

Manager traits and quality-of-care performance in hospitals.

Patient satisfaction and quality of surgical care in US hospitals.

Quality and equity of care in U.S. hospitals.

Male ICU nurses' experiences of taking care of dying patients and their families: a gender analysis.

Quality of nursing care in psychiatric wards of university hospitals in northwest of iran from the perceptions of nurses.

Patients' perceptions of care are associated with quality of hospital care: a survey of 4605 hospitals.

Improving paediatric and neonatal care in rural district hospitals in the highlands of Papua New Guinea: a quality improvement approach.

Advance care planning and end-of-life care in a network of rural Western Australian hospitals.

A Balancing Act: Experiences of Nurses and Physicians When Making End-of-Life Decisions in Intensive Care Units.

Hand Hygiene Adherence Among Health Care Workers at Japanese Hospitals: A Multicenter Observational Study in Japan.

Primary care in urban hospitals.

Massachusetts Department of Public health. Assessing the quality of care in the Department's hospitals.

Care at the Very End-of-Life: Dying Cancer Patients and Their Chosen Family's Needs.

Hospitals with the highest intensive care utilization provide lower quality pneumonia care to the elderly.

End-of-life health care utilization in hospitals with compared to those without palliative care programs.

Supporting adherence and improving quality of life in haemophilia care.

End-of-life care pathways for improving outcomes in caring for the dying.