1 INTRODUCTION

The number of people aged 60 years and older will increase from 1 billion in 2019 to 1.4 billion by 2030 and 2.1 billion by 2050. This increase is occurring at a unique pace and will accelerate, particularly in developing countries, in coming decades (World Health Organization, 2021). Every fifth person in Sweden is ‘older’ (≥65 years of age), which is a proportion that is expected to increase to one in four within a few decades (Statistic Sweden, 2020). The increasing proportion of older people in society is a common finding in several western countries, which is reflected in the increased need for prehospital emergency care and the associated use of the ambulance services (AS) (Lowthian et al., 2011). This can be seen, for example, in Sweden, with an increase in ambulance assignments by 68% during 2009–2014 (Bremer, 2016). A significant contributor to this increase is the growing population of older people with greater healthcare needs. This is confirmed by a recent study indicating a considerable increase in dispatched ambulance resources for patients between 70 and 89 years of age, showing that 60% of the assignments involved older patients >70 years of age (Hjalmarsson et al., 2020). Because autonomy is an important ethical value for most patients, it is also important to understand how ambulance clinicians (ACs) perceive and manage patient autonomy. ACs in Sweden comprise mainly registered nurses (RN) with or without specialist training in prehospital emergency care. These RNs are sometimes paired with emergency medical technicians (EMTs), who are commonly trained as assistant nurses. Henceforth, in this paper, ACs refer to these nurses of differing roles.

In this study of ACs’ attitudes towards older patients’ self-determination, we define autonomy as ‘the power or right of self-government’, while self-determination is defined as ‘the act or power of deciding things for oneself’. Here, the concept ‘attitude’ is understood as ‘a way of thinking or feeling about a fact or state’ (Encyclopædia Britannica, 2020). Taken together, autonomy in older people in AS—as the first link in the care chain—is important for nurses in the following phases of older people nursing.

Self-determination in older people has been found to have a positive relation to quality of life (Bölenius et al., 2019; Kalfoss, 2010). A primary quality aspect of self-determination is to independently manage oneself without support (Johannesen et al., 2004). However, certain conditions must be met to exercise self-determination. Older persons need to feel safe in relationships, be able to influence decisions, and feel involved (Ekelund et al., 2014). Building trustful relationships over time is important in environments supporting older people's self-determination (Souesme & Ferrand, 2019).

There is a dearth of studies examining self-determination in older patients within AS settings. However, ethical conflicts have been found to be related to self-determination when patients refuse the care offered by the ACs (Sandman & Nordmark, 2006). In a recent study from the AS context, Bremer and Holmberg (2020) found that a number of ethical conflicts originated from problems related to the patient's self-determination. Ethical conflicts arose because of the ACs’ inadequate access to the patient's narrative, uncertainty regarding the patient's decision-making ability and conflicting assessments of the patient's best interest.

Research on older patients’ own experiences from an ethical perspective is lacking when it comes to urgent situations in need of ambulance care. In a study by Kluit et al. (2018), it was found that the decision to transport older people to hospital was influenced by whether the patient was presenting with acute conditions, despite a period of complaints varying between hours and years preceding the decision. The older patients saw hospital admission as inevitable due to an unsatisfactory care environment at home and positive expectations of hospital care. Shared decision-making was rarely seen, and an ethical dilemma occurred when the next of kin consented to hospitalisation against the wishes of the patient.

Additionally, older patients’ decision-making abilities in acute situations are often reduced and sometimes completely lacking. The patients’ inability to express themselves has direct consequences for patient autonomy, integrity and dignity. Due to the lack of patient self-determination, ACs encounter forms of surrogate decision-making, often with support and information from family members, but sometimes entirely on the basis of the care team's independent assessment of what might benefit the patient, based on the signs of illness and available medical history (Bremer et al., 2012; Bremer & Holmberg, 2020; Hagiwara et al., 2013; Holmberg et al., 2016).

Older patients who need AS are often vulnerable and dependent, especially when their decision-making ability is impaired, which increases the risk of important ethical values being violated. When the patient is older and fragile, there is reason to believe that the risk is even more pronounced, making self-determination conditional (Ekelund et al., 2014). In urgent situations, older patients may have impaired decision-making ability or reduced ability to participate in decisions concerning their own care, making it more difficult for the ACs to understand and respect the patient's autonomy, as there is uncertainty about the patient's authentic wishes and own values. Hence, the risk of value conflicts is significant (Bremer & Holmberg, 2020). Older, acutely ill persons with impaired decision-making abilities also risk negative discrimination resulting from conscious or unconscious bias among ACs or the patient's needs are assessed based on the ACs’ own values (Bremer et al., 2015). This risk can be assumed to be greater when older patients do not have a social network that can act as surrogate decision-makers in cases where the older patients are unable to identify, articulate and fight for their needs. In a worst-case scenario, older patients can be given lower priority, even though their needs are just as, or possibly even more, significant compared to younger patients with similar needs. Consequently, the delivery of unequal care may result (Giordano, 2005; Rantala et al., 2016).

Taken together, the older patient's vulnerability in urgent situations exposes the asymmetrical and unequal power relationship in the care relationship between the AC and the patient, partly due to ACs’ own values and how they can influence the decisions made. Zaner (2000) suggests that it is the healthcare provider who has the power, through knowledge, skills, resources, social legitimacy and legal authority, and not the patient. Wiggins and Schwartz (2005) suggest that healthcare providers have the power to help, while at the same time, the patient becomes more dependent on them, exposing the patient to greater vulnerability and abuse of power. Consequently, the patient needs to trust not only the healthcare providers’ professional knowledge, but also their moral character. Through relational autonomy, the shortcomings in an individualistic or sovereign perspective on autonomy can balance the principle-based framework of beneficence, justice and equality (Donchin, 2001; Mackenzie, 2008; Stoljar, 2011). Mackenzie argues for a relational approach to the principle of respect for autonomy, which means an obligation on the part of healthcare providers to respect patients whose autonomy is impaired by promoting the autonomy competence of these patients. Based on these descriptions of power relations in care relationships and individualistic versus relational aspects of autonomy, there is reason to pay attention to the ACs’ management of their power and patient autonomy in care relationships with older patients.

Hence, the aim of this study is to empirically explore the attitudes among Swedish ACs about older patients’ self-determination in cases where patients have impaired decision-making ability, and who are in urgent need of care.

2 MATERIALS AND METHODS

An explorative design was adopted using a modified Delphi technique (Keeney et al., 2011). The Delphi method emanates from the assumption that a group's opinion, which in this case comprised a panel of Swedish ACs, is more valid than individual opinions.

2.1 Study setting and participants

The study was conducted in two regions (A and B) in southern Sweden. Region A covers approximately 5,600 square kilometres and 300,000 inhabitants. Region B covers approximately 11,200 square kilometres and 240,000 inhabitants. The AS in region A has eight ambulance stations, while Region B has fifteen, covering both rural and urban areas. In 2018, each region dispatched an AS resource to approximately 35,000 calls. The participants were recruited in both regions. The inclusion criteria were ACs with professional affiliations, that is, RN with or without specialist training, and EMTs. RNs with specialist training in ambulance care, that is prehospital emergency nurses (PEN), were the most dominant subspecialty in the RN group (Table 1). Stratified sampling was performed to achieve variation in age, gender, and professions.

TABLE 1. Ambulance clinician demographic characteristics Round 1 Round 2 Round 3 Round 4 Gender, n Men 15 14 13 13 Women 17 17 17 17 Age, years (mean) 25–65 (45) 25–65 (45) 25–65 (45) 25–65 (45) Experience from ASa, years (mean) 1,5–45 (16) 1.5–45 (16) 1.5–45 (16) 1.5–45 (16) ACb professional affiliation, n Emergency medical technician 7 6 5 5 Registered nurse 10 10 10 10 Ambulance nurse 14 14 14 14 Number of participants, n 31 30 29 29 a AS= ambulance services. b AC= ambulance clinicians. 2.2 Data collection and analysis

Data were collected and analysed over four rounds between November 2019 and April 2020. The analysis of the focus groups was carried out in Swedish by Swedish-speaking authors, as well as the analysis that led to the statements that were included in the instrument and the preliminary categories that constituted the result. After the items and the preliminary result categories had been translated into English, the fifth and sixth English-speaking authors also participated in the continued analysis process and the writing of the result.

Round 1

The data collection in the first round was carried out with 6 focus groups involving 4–6 participants in each group. The first and the last author moderated the focus group conversations (Morgan, 1997), one having the role as main-moderator and the other as co-moderator. Initially, the participants were informed about the aim of the study and the topic of the focus group conversation. The conversations were digitally recorded and lasted between 77 and 95 min (mean = 86 min) and transcribed verbatim. The focus groups were supported with an opening question: ‘What does patients’ self-determination mean to you?’ Bearing in mind the inductive focus of the study, low-moderator involvement was selected (Morgan, 1997), to allow interaction within the group and facilitate great depth in participant-generated data.

The data analysis started by reading the transcribed conversations several times to obtain a sense of the transcripts as a whole. The first and last authors undertook a manifest and descriptive qualitative content analysis. Statements expressing attitudes regarding older people's self-determination were extracted from the data and entered into an Excel document. The statements were then reformulated for clarification, with the intent of remaining close to the concepts and words used by the participants. Subsequently, the statements were compared with each other to reduce redundancy. Statements expressing similar attitudes were grouped together into categories expressing the overall structure of the data. The statements and the grouping were then discussed and adjusted together with the second, third and fourth authors. Round 1 resulted in four categories concerning the ACs’ attitudes to older people's self-determination, comprising a total of 108 statements, henceforth referred to as items. Finally, a face-validity check of the items was undertaken, involving the other Swedish-speaking authors.

Round 2

For data collection in the second round, the 108 items emerging in Round 1 were used to develop a questionnaire employed in Round 2 (Table 2). To capture additional attitudes, the participants were given the opportunity to describe, in free text, any additional attitudes that they considered important. To enhance validity, the original questionnaire was piloted with a group of people who did not participate in the main study (Hasson et al., 2000). The pilot group consisted of three experienced specialist trained ambulance nurses and one RN, all active researchers.

TABLE 2. Delphi flowchart of the four rounds Round 1 → Round 2 → Round 3 → Round 4 Participants, n 32 31 30 30 Response rate, % 100 97 97 100 Participant drop-out, n 0 1 1 0 Round activity Focus group conversations → Questionnaire 108 items → Questionnaire 59+1 items → Questionnaire 41 items ↓ ↓ ↓ With consensus, n 49 19 4 Without consensus, n 59 41 37 Total without/with consensus, n 37/72

The questionnaire for the following rounds was introduced with the question; ‘To what extent do you agree with the following items?’, encouraging the participants to rate the extent to which they agreed with each item using a five-point Likert scale, ranging from (1) ‘not agree’ to (5) ‘strongly agree’. For analytical purposes, the scale was trichotomised to a three-point scale before determining whether consensus had been reached (Jirwe et al., 2009; Rådestad et al., 2013). Thus, 1–2 on the Likert scale represented ‘not agree’, 3 represented ‘neutral’ and 4–5 represented ‘agree’. The level of consensus was set at 70% prior to data collection (Keeney et al., 2011). Thus, in the present study, an item was considered to have reached consensus when 70% or more participants agreed on any of the trichotomised scale responses.

For data analysis in this and the following rounds, descriptive statistics were used, focusing on mean values and standard deviation.

The questionnaire was distributed via e-mail to the 32 participants from Round 1. Three reminders were sent via e-mail. The participants who did not respond to the second reminder e-mail were contacted by telephone (n = 5). The questionnaire was available for 19 days. In total, 31 participants answered the questionnaire, resulting in a 97% response rate in Round 2.

Round 3

For data collection in this round, a second questionnaire was developed, comprising the items that did not reach consensus in Round 2 (n = 60) together with one new item that emerged from the open-ended question (Table 2). Feedback containing the group mean values was provided for each item to stimulate the participants to reflect upon these values in relation to their reconsidered answers in the data collection round. The questionnaire was distributed via e-mail to the 31 remaining participants. Three reminders were sent via e-mail. The participants who did not respond to the second reminder were contacted by phone (n = 5). The questionnaire was available for 17 days. One participant declined further participation. In total, 30 participants answered the questionnaire, resulting in a 97% response rate in Round 3.

The data analysis followed the procedure presented under the heading ‘2.2.2 Round 2’.

Round 4

For data collection in this round, a third questionnaire was developed, comprising the items that did not reach consensus in Round 3, resulting in 41 items. The same procedure as in Round 3 was followed, with the questionnaire distributed to the 30 remaining participants (Table 2). Three e-mail reminders were distributed. Those who did not respond to the second reminder were contacted by phone (n = 4). The questionnaire was available for 13 days, resulting in a 100% response rate within this round. In total, 30 participants finished all four rounds, giving a response rate of 94%. The responses were analysed as in Round 3, and an additional 4 items reached consensus.

The data analysis followed the procedure presented under the heading ‘2.2.2 Round 2’.

2.3 Ethical considerations

The study was carried out in line with the Declaration of Helsinki (The World Medical Association, 2013). Permission was granted by the Swedish Ethical Review Authority prior to the study (No. 2019-02127). All participants received verbal and written information about participation being voluntary and that they could withdraw consent to participate at any time without stating the reason. Participants were given the opportunity to ask questions and have them answered by the research team. Consent to participate was obtained from the participants. During the focus group interviews, maintaining confidentiality within the group was emphasised.

3 RESULTS

The results indicate an ambition to respect older patients’ self-determination, preferably in collaboration with the patient. This ambition was impeded by the influence of the AC colleague, significant others and/or other healthcare professionals. Alternatively, collaboration with others was also found to be an important aspect of respecting the patient's self-determination. The patients were perceived as sometimes not being willing or able to participate, thus delaying the ACs’ decisions. However, ACs still experienced having to make clinical decisions for the patient, using their power over the patient because of their professional status. Additionally, the context, with dyadic teams and caring for one patient at a time, together with the varying patient population, may both support and undermine the ability to respect the self-determination of the individual patient.

3.1 Round 1

The analysis of the focus group conversations generated 108 items grouped into four overarching categories, namely Category (1) attitudes regarding the patient (n = 35); Category (2) attitudes regarding the patient relationship (n=8); Category (3) attitudes regarding oneself and one's colleagues (n = 45); and Category (4) attitudes regarding other involved factors (n = 20).

3.2 Round 2

Forty-nine of the 108 items reached the consensus level of ≥70% in Round 2 (Table 3 and Table 4): Category 1 (n=11); Category 2 (n=5); Category 3 (n=25); and Category 4 (n=8).

TABLE 3. Items for which consensus was reached in Categories 1 and 2 Mean value Standard deviation Consensus reached in round Category 1 Agree An elderly patient's wishes may vary over time 4.6 0.8 2 It is difficult to assess a patient's capacity for self-determination if the patient is not fully capable of making a decision 4.5 0.8 2 Self-determination can entail refusing life-saving treatment. 4.4 1.2 2 There is a difference between being able and willing to exercise self-determination 4.2 0.8 2 If a patient is unable to speak for himself/herself, ambulance clinicians have less knowledge of the patient's wishes compared with others present around the patient (e.g. relatives or healthcare professionals) 4.2 0.8 2 If a patient has previously participated in a documented ELC (End of Life Care) discussion, it is easy to respect the patient's self-determination and not administer life-saving treatment in the event of cardiac arrest 4.2 1.2 2 There are elderly patients who do not want self-determination 4.1 0.7 3 Elderly patients who represent a danger to themselves or their surroundings have no right to self-determination 4.0 1.0 2 Elderly patients have a personal responsibility for their decisions and their consequences 3.9 0.7 3 Younger patients are more likely to demand self-determination than older ones 3.9 0.9 3 Even if a patient has critical medical symptoms, his/her self-determination must be respected 3.8 0.8 3 An elderly patient who wants to go to the accident & emergency department must not be denied this 3.8 1.0 3 Elderly patients have faith in the competence of ambulance clinicians and do not therefore request self-determination 3.7 0.4 3 It is common for patients to relinquish their self-determination to ambulance clinicians 3.7 0.6 2 Neutral An elderly patient is rarely involved in the decision to call an ambulance 3.0 0.5 4 It is easier to disregard the self-determination of elderly patients who are acting out 3.0 0.6 4 Not agree A patient's capacity for self-determination is limited if it requires a lot of time to respect it 2.3 0.7 4 Elderly patients may not make their own decisions about where they are to be transported 2.2 0.8 3 An elderly patient has an obligation to say whether he/she wants pain relief 2.1 0.8 3 The capacity of elderly patients for self-determination depends on their physical condition 2.1 1.2 2 If an elderly patient's self-determination is in conflict with the ambulance service's guidelines, the written consent of the patient needs to be obtained. 2 0.9 2 Elderly patients suffering from psychological ill health have no right to self-determination 1.7 0.8 2 Category 2 Agree Respecting an older patient's self-determination is about making a decision together with the patient 4.5 0.8 2 Protecting a patient's self-determination is the joint responsibility of me and the patient 4.3 1.0 2 During a conversation with a patient, I assess his/her capacity for self-determination 4.2 0.8 2 Decisions based on a patient's self-determination are always made in consultation with the patient 4.0 1.0 2 It is easier to protect a patient's self-determination if only the patient is present 3.9 1.0 2 Neutral If I am unable to judge during a conversation whether a patient has the capacity for self-determination, the assessment is based instead on a generalisation with reference to previous care meetings 2.9 0.6 4 TABLE 4. Items for which consensus was reached in Categories 3 and 4 Mean value Standard deviation Consensus reached in round Category 3 Agree Protecting an elderly patient's self-determination means protecting his/her participation. 4.7 0.6 2 There is a difference between motivating, recommending and persuading a patient. 4,7 0.6 2 It is important that I make the patient realise the seriousness of his/her situation. 4.7 0.6 2 To protect a patient's self-determination is to protect his/her dignity. 4.7 0.7 2 In CPR situations, I often lack information about the patient's wishes. 4,6 0.7 2 It is my task to create the conditions for a patient's self-determination by helping him/her to understand that there are different care options. 4.6 0.8 2 A responsive approach is needed to be able to respect the patient's self-determination. 4.5 0.6 3 I disregard the patient's self-determination if he/she threatens to commit suicide. 4.4 0.9 2 It takes creativity to accommodate the patient's self-determination. 4.3 0.8 2 It takes courage to respect a patient's self-determination. 4.3 0.8 2 There are situations in which I feel that the best thing for a patient does not corresponds with the patient's own opinion. 4.3 0.9 2 It is my job to make a patient understand that the patient has the right to self-determination. 4.3 0.9 2 It takes experience and confidence in the profession to protect a patient's self-determination. 4.3 1.2 2 The experience, personality and interest of paramedics influence the extent to which the patient's self-determination can be respected. 4,2 0.8 2 Forcing a patient is to abuse his/her self-determination. 4.1 0.6 3 I have an obligation to respect the patient's decision. 4.1 0.9 2 There are situations in which the patient's self-determination comes into conflict with what I considered dignified. 4.1 1.0 2 I use my colleague as support when deciding whether a patient has the capacity for self-determination. 4.1 1.0 2 To be able go along with a patient's wishes and show respect for his/her self-determination, it is sometimes necessary that my colleague and I switch roles. 4,0 1.1 2