Introduction
Nursing home care is complex in that many residents live with several chronic conditions (WHO, 2011), with increasing symptom prevalence (Estabrooks et al., Reference Estabrooks, Hoben and Poss2015), and that it includes both curative services and palliative care (Nakrem, Reference Nakrem2015). In Sweden, there is a wide range in time of residency in residential care: almost 50% die within the first year (Schön et al., Reference Schön, Lagergren and Kareholt2016), but nearly 30% of the women and 20% of the men are alive after 4 years (National Board of Health and Welfare, 2018). Regardless of the length of residency, each resident's length of stay is often equal to the length of their remaining life (Schön et al., Reference Schön, Lagergren and Kareholt2016); because of this, there is a high prevalence of palliative care needs in residential care (Morin et al., Reference Morin, Aubry and Frova2016). The residents are affected both physically and mentally by age and illness: 57% need help in activities of daily life, and 67% have cognitive impairment (SOU, 2017).
The high prevalence of cognitive impairment in nursing homes makes autonomy, when conceptualized as assuming full decisional capacity, difficult to apply. Instead, the interpretation of the autonomy concept as relational is more appropriate (Jacobs et al., Reference Jacobs, Snow and Allen2015). The relational view of autonomy suggests that individuals can never be fully independent (MacDonald, Reference MacDonald2002) and that there is no conflict in being autonomous and engaging others in decision-making (Walter & Ross, Reference Walter and Ross2014). Self-determination can be regarded as the process of exercising the individual's right to autonomy (Beauchamp, Reference Beauchamp2013). In palliative care, self-determination is defined as a decision-making process that includes the support and advice of others, for example, family members and healthcare professionals (Bakitas, Reference Bakitas2005). When physical frailty progresses, the possibilities to carry out personal choices are limited, and dependency on others can entail a loss of both executional and decisional control. However, loss of physical performance does not justify loss of freedom to make decisions, and it is crucial to protect the decisional, as well as the executional, autonomy in older persons with frailty to preserve their self-determination (Collopy, Reference Collopy1988).
Decision-making and quality of care (QoC) are connected as having the opportunity to participate in decisions about one's own health and welfare is an attribute of QoC (Donabedian, Reference Donabedian1990). In nursing homes, aspects related to the persons who provide the care and institutional aspects of the care process are perceived as pillars of QoC by residents and relatives. QoC in nursing homes is, according to family members, enhanced by care routines that are mutually developed with residents and, conversely, is obstructed when residents are expected to adapt to existing care routines (Vinsnes et al., Reference Vinsnes, Nakrem and Harkless2012). A nonpaternalistic attitude with respect for the autonomy of the residents, a close relationship between staff and residents, and possibilities for relatives to take part in decision-making are also described to enhance QoC (Rodriguez-Martin et al., Reference Rodriguez-Martin, Martinez-Andres and Cervera-Monteagudo2013). Staff in nursing homes recognize and want to achieve a relationship with the residents, although organizational factors limit their ability to shift their focus away from task-based activities (Beck et al., Reference Beck, Törnquist and Broström2012).
Older people can delegate end-of-life decision-making to others by indirectly expressing their wishes and describing their desired end-of-life outcomes. They believe that their appointed delegate will make appropriate decisions on their behalf and they can thereby achieve a sense of control without being directly responsible for specific decisions (Romo et al., Reference Romo, Allison and Smith2017). Nurses and physicians do not always involve nursing home residents in end-of-life discussions, although they are capable of decision-making (Dreyer et al., Reference Dreyer, Førde and Nortvedt2010). This can be remedied by having family members monitor the care (Davies & Nolan, Reference Davies and Nolan2006) and, by proxy decision-making, they can achieve autonomy and participation for the residents (Hedman et al., Reference Hedman, Häggström and Mamhidir2017). However, being a proxy decision-maker in end-of-life care can be a burden for family members because they perceive themselves as being responsible for the decisions that are made (Dreyer et al., Reference Dreyer, Førde and Nortvedt2009). Family members can also jeopardize the residents' autonomy at end of life when they want more active, life-prolonging treatments for the residents than the residents do themselves (Gjerberg et al., Reference Gjerberg, Førde and Pedersen2010).
According to Swedish law, care should be of good quality and build on respect for the patients' self-determination (SFS, 2017). Despite this, daily routines, rather than individual preferences, direct life for those who live in nursing homes (Hellström & Sarvimäki, Reference Hellström and Sarvimäki2007) and concern about the care quality is reported by residents' family members as being one of their main anxieties (Davies & Nolan, Reference Davies and Nolan2006). This study aimed to provide knowledge about residents,’ in the palliative phase, and family members' perceptions of QoC and self-determination, conceptualized as having executional and decisional control, and to detect any differences between their experiences. The research questions were as follows.
• How do the residents perceive QoC and decision-making in everyday life and in life-changing situations?
• Do residents hand over decision-making to staff or family members; and do staff or family members make decisions against the residents' will?
• Are there differences between residents' and family members' perceptions of QoC and decision-making in everyday life and in life-changing situations?
Methods
Design
This cross-sectional study was based on data gathered from residents and family members in residential care facilities, representing both municipal and private operators, in all 10 districts of a large city in the southwest of Sweden. Data were collected during 2017 and 2018, using the instrument, Quality from the Patients' Perspective (QPP) (Larsson & Larsson, Reference Larsson and Larsson2002).
Procedure and participants
One registered nurse working at each residential care facility was designated as the main contact for the study and assisted in recruiting the residents. The main inclusion criteria were that the residents should reside in a general unit and be in a palliative phase, defined as having a maximum life expectancy of one year. The contact nurse used the “surprise question” (SQ) (Lynn, Reference Lynn2005), asking herself or himself, “Would I be surprised if this resident die within a year?” If the answer was no, the resident could be asked to participate in the study. The included residents then asked one family member to participate to obtain their perspectives. Speaking and understanding Swedish was also an inclusion criterion. Participating residents did not have to be able to read the questions or write their answers by themselves, as the first author (R.S.) assisted when necessary. Because of conditions such as impaired vision, paresis, or reduced strength, 83 residents were assisted when completing the questionnaire. Some residents did not have family members or did not want to ask them to participate. Two residents and one family member declined participation, and 18 family members did not return their questionnaires. Two of the residents' questionnaires had to be excluded because of the amount of missing data. The study was approved by the Regional Ethics Committee in Gothenburg (Dnr1036-15).
Data collection
Data on residents' and family members' perceptions of QoC, with a focus on self-determination, were collected using an abbreviated version of the QPP, specially designed for residential care facilities. The instrument measures the perception of the actual care received, called the perceived reality (PR), and the importance of each aspect of care, called the subjective importance (SI). Significantly lower PR than SI indicates a need for quality improvement (Wilde et al., Reference Wilde, Larsson and Larsson1994) and can also be interpreted as a low level of self-determination, as the instrument measures both executional and decisional activities. Both the QPP for residents and for family members had 24 original items and six additional study-specific items about decision-making in everyday life and in life-changing situations, all formulated as statements such as “I/My family member is treated with respect.” The statements were evaluated on a 4-point Likert scale, ranging from 1 (do not agree at all) to 4 (fully agree) for PR, and from 1 (of little or no importance) to 4 (of very great importance) for SI. Additional “not applicable” and “do not know” responses were used in both questionnaires when appropriate. To further investigate the decision-making process, four items about residents handing over decision-making to staff and family members and having decisions made against their will by staff and family members were added to both questionnaires, as were two items about perceived physical and mental health. The added items were formulated and designed with help from ImproveIt (www.improveit.se/kupp/), the company that provides QPP, to fit in with the other items in the questionnaires.
Data analysis
The IBM SPSS Statistic Data Editor Software, version 23, was used to analyze data. The responses from the total sample of residents (n = 112) were used in calculations of the residents' perceptions of care. When comparing residents' and family members' perceptions of care, the sample consisted of the residents whose family member had completed the questionnaire (n = 83). Demographic and clinical characteristics were examined with descriptive statistics. Wilcoxon's signed rank test was used for examining differences between PR and SI of QoC and in decision-making in everyday life and lifechanging situations in the separate groups of residents and family members. Wilcoxon's signed rank test was also used for investigating differences between residents' and family members' experiences in the paired analyses.
Results
Demographic and clinical characteristics
As presented in Table 1, the study population consisted of 112 residents and 83 family members, mostly from municipal residential care operators, with considerably more women in both groups. Heart disease was the most reported disease, and many residents had comorbidities. The residents' mental health was rated as generally better than their physical health by both residents and family members, but the family members rated the physical health as being worse than the residents did. The residents were significantly older than the family members because the majority of the residents had asked their children to participate. Most of the family members visited the residents once or several times a week.
Table 1. Demographic-, health-, and social-related characteristics of residents and family members
SD, standard deviation.
* Residents from the total sample whose family member has completed the Quality from the Patients' Perspective.
Residents' perceptions of QoC and decision-making
As presented in Table 2, there were statistically significant differences between PR and SI values in 17 of 24 of the ordinary QPP-items, indicating that the QoC was lower than preferred in these areas. Lowest mean values in PR were found in support when feeling lonely; support when feeling worry, anxiety or fear; and in staff's time to talk to the residents. Highest mean values in PR were found in access to technical aid, support with personal hygiene, and support during meals.
Table 2. Comparisons between perceived reality and subjective importance in the group of residents and between residents and family members
p values refer to differences in Wilcoxon signed rank test. p values also shown for differences between perceived reality and subjective importance within the residents' group, n = 83, and within the family member group, n = 83. Statistical significance assumed at p ≤ 0.05.
PR, perceived reality; SD, standard deviation; SI, subjective importance.
* Residents from the total sample whose family member has completed the Quality from the Patients' Perspective.
In decision-making in everyday life and in life-changing situations, there were statistically significant differences between PR and SI values in four of six items, indicating that the actual experiences were not consistent with the residents' expectations. The lowest mean values in PR were found in staff knowing about the residents' will in life-changing decisions, whether staff would make the right decisions according to the residents' will, and family members knowing about the residents' will in life-changing decisions. The items concerning staff had both lower PR and SI values than the related items concerning family members (Table 2).
Concerning the residents' handing over of decision-making to staff or family members, and about residents having decisions made against their will, 65% (n = 73) never handed over decisions to the staff, and 60% (n = 63) never handed over decisions to their family members. Decisions made against the residents' will were not common: 81% (n = 90) reported that this never happened with staff and 96% (n = 103) that it never happened with family members (Table 1).
Comparison of residents' and family members' perceptions of QoC and decision-making
Mean values for the comparison between residents' and family-members' perceptions are presented in Table 2. Concerning QoC, in 14 of the 24 QPP items, both residents and family members had statistically significant differences between PR and SI values, indicating that both groups perceived QoC as being lower than preferred. Lowest mean values in PR were found in support when feeling lonely; support when feeling worry, anxiety or fear; and support to go outside. Both residents and family members rated these items with lowest mean values, but not in the same order. The highest PR values, for both residents and family members, were in access to technical aids and being treated with respect. Residents' third-highest PR values were in support with personal hygiene, and support to sit and lie comfortably, which were equally rated. Family members' third-highest PR was support during meals, which was also the only item in which both residents and family members were satisfied with the QoC. In the remaining 9 items, residents were content, but family members perceived QoC as being low. Generally, residents rated PR higher than family members did, although not all differences became statistically significant.
In decision-making in everyday life and in life-changing situations, there were statistically significant differences between PR and SI values in four of six items (the same as in the residents' group), for both residents and family members, indicating that their actual experiences were not consistent with their expectations. In addition, the family members also had statistically significant differences between PR and SI values in the item about them making the right decisions according to the residents' will.
Discussion
The main findings of the present study were that neither residents nor family members were satisfied with the QoC that the residents received. There were significant differences between the PR and SI values in both groups concerning QoC and decision-making in both everyday life and in life-changing situations. Generally, residents were more satisfied than family members were because they rated PR higher and had fewer differences between PR and SI. Even in the family members' highest ratings, there were significant differences between PR and SI, indicating a wish for better quality. The self-determination of the residents can be considered as being compromised, as both decisional and executional control were affected.
The results indicate a need for broad improvements; however, in access to technical aids, support during meals, support with personal hygiene, and support when visiting the toilet, the QoC was satisfactory. Prioritizing the residents' most basic needs of bodily care and practical tasks is common in residential care facilities (Beck et al., Reference Beck, Törnquist and Broström2012; Holmberg et al., Reference Holmberg, Hellström and Österlind2018). Routine and task-based care has been found to have a negative effect on residents' autonomy (Rodgers et al., Reference Rodgers, Welford and Murphy2012), because promoting autonomy requires opportunities to make genuine choices (MacDonald, Reference MacDonald2002). When the staff’s working day follows routines rather than acknowledging the residents' desires and wishes, the opportunity for making choices is less prevalent (Murphy & Welford, Reference Murphy and Welford2012).
In the present study, both residents and family members claimed that staff did not have enough time to talk to the residents and, consequently, they perceived support when feeling lonely, and support when feeling worry, anxiety, or fear as low. In Sweden, 66% of residents in residential care facilities are troubled by loneliness. Residents who often perceive loneliness are less satisfied with the care they receive and the possibility for self-determination, participation, and individual adjustments (National Board of Health and Welfare, 2016). Loneliness is described by older persons as a complex and private matter requiring good relationships to discuss (Kharicha et al., Reference Kharicha, Iliffe and Manthorpe2017). This might constitute a problem for the participants in the present study as there were significant differences between PR and SI in receiving help from staff known to the resident, which indicates that there was a lack of close relationships between residents and staff. Both residents and family members also perceived a lack of engagement from staff. Residents depend on staff, not only for clinical care, but also for their psychosocial wellbeing (Nakrem et al., Reference Nakrem, Vinsnes and Seim2011), and commitment to the care relationship should be promoted to maintain and develop QoC for older people (Häggström et al., Reference Häggström, Mamhidir and Kihlgren2010). In-depth knowledge and interaction are necessary to promote autonomy (Moser et al., Reference Moser, Houtepen and Widdershoven2007), and nursing home staff need to recognize the individuality and know the habits, preferences, and tastes of the residents (Naess et al., Reference Naess, Fjaer and Vabo2016).
In the present study, both residents and family members were discontent with the staff knowledge about the residents' preferences in daily decision-making as well as in life-changing decisions, such as hospital admission, intravenous fluids, and cardiopulmonary resuscitation (CPR). They also perceived insufficient QoC in having the opportunity to participate in decisions about healthcare, care and rehabilitation, and in having the opportunity to influence how the support will be provided. These results indicate a need for improvement in the cooperation between residents, staff, and family members. Vinsnes et al. (Reference Vinsnes, Nakrem and Harkless2012) found that family members perceive themselves as an essential communication link between residents and staff and in having a unique ability to interpret the residents' wishes. In the present study, the residents seemed to adhere to the perception about the family members' ability to interpret their needs, as they trust them to make the right decisions according to their will if they were to take over their decision making. This is interesting because both residents and family members rated PR lower than SI in decision-making in life-changing situations, indicating a need for more discussions between them in this regard. This result is consistent with those of Bollig et al. (Reference Bollig, Gjengedal and Rosland2016), who found that most residents believed that their relatives would know their wishes in end-of-life-care without having discussed them. In the present study, as well as in Bollig et al. (Reference Bollig, Gjengedal and Rosland2016), the family members are unsure about their ability to make the right decisions and, according to Bollig et al. (Reference Bollig, Gjengedal and Rosland2016), the relatives perceived decision making as a burden when they were not aware of the residents' wishes.
Nursing home residents consider death to be a natural end to a long life, but they rarely speak about it to staff or relatives (Österlind et al., Reference Österlind, Ternestedt and Hansebo2017). Speaking about dying and death can be seen as an unpleasant and sad topic by the residents, but also as unnecessary, as they perceive themselves as being in fairly good health (Ingravallo et al., Reference Ingravallo, Mignani and Mariani2018). In the present study, many residents rated their health as being fairly good or very good, despite having several diagnoses and being dependent on staff for activities of daily living. This might partly explain the lack of conversations about life-changing situations, but it is surprising that the topic was not raised by staff more often, as the residents in the study were included because staff estimated their remaining lifetime to be a maximum one year.
Advance care planning (ACP), where people can make legally binding decisions about care in advance or appoint a specific person as a surrogate decision-maker for the future, is beneficial in nursing homes. It gives residents a voice in healthcare decisions, even after they have lost the capacity to take part in such discussions (Martin et al., Reference Martin, Hayes and Gregorevic2016). Few nursing homes have routine procedures for discussing these questions with residents and relatives at an early stage (Gjerberg et al., Reference Gjerberg, Førde and Bjørndal2011), even though death is regarded, by staff, as being natural and expected (Holmberg et al., Reference Holmberg, Hellström and Österlind2018). Caregivers find it difficult to talk about palliative care unless the residents or their relatives initiate the dialog themselves (Häggström et al., Reference Häggström, Mamhidir and Kihlgren2010). This constitutes a problem because most people in nursing homes seem to need a person from outside the family to start conversations about ACP (Bollig et al., Reference Bollig, Gjengedal and Rosland2016). Discussing ACP also requires trust and closeness (Ingravallo et al., Reference Ingravallo, Mignani and Mariani2018), and the results of the present study have already established a lack of time to talk, a lack of engagement, and insufficient relationships between residents and staff. This might further explain the staff lacking knowledge about decision-making in daily life and in life-changing situations as well as the low values relating to trust in the staff’s ability to make the right decisions according to the residents' will.
The majority of residents in the present study claimed that they never handed over decision making to staff or family members. This finding is consistent with previous research showing that having decision-making power is important in residents' everyday life, as well as being a factor of QoC (Donabedian, Reference Donabedian1990). Although the residents in the present study claimed that they made most of their decisions by themselves, and were not afraid to express their views to the staff, both residents and family members perceived that they would like the residents to participate more in decisions about healthcare, and care and rehabilitation, and to be able to influence how the support would be carried out to a greater extent. In addition, 14% of the residents claimed that staff sometimes made decisions against their will. A reasonable explanation might be that the residents try to hold on to their self-determination, but, in reality, have little opportunity to make genuine choices as they depend on the circumstances in the residential care facility. As one resident wrote in the comment field in the questionnaire: “Since I moved to the residential care facility I have become used to doing as others say, although I don't always want to. Woman, aged 99 years.”
Methodological considerations and study limitations
Because people are considered to be approaching the end of life when they are likely to die within the next 12 months (General Medical Council, 2010), the palliative phase in this study was defined as having a maximum life expectancy of one year. The accuracy of the SQ in predicting survival or death in nursing homes varies, from 57% to 67%, depending on the specific timeframes used (3 months, 6 months, or seasonal) (Rice et al., Reference Rice, Hunter and Hsu2018). However, no timeframe of one year has, to our knowledge, been tested with residential care facilities as the only context. This means that the accuracy of the SQ was not tested for the timeframe and context of the present study, which might be a limitation in the sample procedure. Assisting in reading and/or filling in the questionnaire allowed residents who usually cannot participate in this kind of study to make their voices heard. This strategy aligns with the World Medical Association's (WMA) declaration of Helsinki, which states that groups that are underrepresented in research should be provided appropriate access to participation (WMA, 2013). However, it is a limitation that people who are unable to speak Swedish were excluded because the QPP has not been translated into other languages.
Conclusion and implications
The shortcomings in QoC and self-determination in the present study demonstrate a need for broad improvements, especially in the area of psychosocial care. As confiding in staff about end-of-life preferences or when feeling lonely or worried requires closeness and trust, the managers in residential care facilities must ensure that appropriate conditions for building trustful relationships are established. Continuity and time are important factors when building relationships, which implies a need for at least one specially assigned staff to each resident (i.e. contact person). Time should be set aside for regular activities or conversations between the resident and the contact person and focus on areas important to the resident. These regular meetings can serve as a basis for improving the staff’s knowledge about the residents' preferences and ultimately will help staff interpret wishes and accomplish activities when the residents can no longer express their will. Residents and their family members agree on the areas in need of improvement and the high consistency between them suggests that family members are appropriate proxy decision-makers, although there are some differences in their perceptions. There is a need to support family members when acting as proxy decision-makers in end-of-life care and facilitate discussions about end-of-life preferences between them in time. Because there is a high prevalence of palliative care needs in residential care, a palliative approach early in the residents' illness trajectory would be beneficial. Palliative care both affirms life and regarding dying as a normal process (WHO, 2002), which makes a palliative approach suitable in decision-making in the residents' everyday life as well as in planning for end-of-life care. An early implementation of palliative care with a focus on what brings quality to each resident's life could facilitate the shift from task-based care to what really matters to the resident and thus enhance the possibilities for self-determination.
Author ORCIDs
R.N. Ramona Schenell, 0000-0002-6991-8798
Acknowledgments
We are very grateful to the residents and their family members who participated in this study. This work was financially supported by The Solstickan Foundation, Sweden; Emil and Maria Palm's Foundation, Sweden; Agneta Prytz-Folke's and Gösta Folke's Foundation, Sweden; Herbert and Karin Jacobsson's Foundation, Sweden; Helge Ax:son Johnson's Foundation Sweden; Adlerbertska's Research Foundation, Sweden; Nurses for Palliative Care, Sweden; and The Swedish Society of Nursing, Sweden.
Open access
