There are different ways to facilitate ethical decision-making in patient care in Europe and North America. In North America ethics consultations are offered in most hospitals mainly to help to protect patient rights and improve care as well as resolve conflicts. The ethicists are primarily clinicians without formal ethics education. The majority of consultations are one-to-one discussions with healthcare staff or the patient/family, resulting in recommended courses of action.1 Subjective evaluations of ethics consultation have shown high staff satisfaction.2 3

Although ethics consultations similar to those in the American model have been reported from Europe,4^–6 it seems that supporting healthcare teams in dealing with ethical problems through reflective practicei are more commonly advocated7^–11 and practised.8 11 12 In Europe there is a lack of studies evaluating reflective ethical practice. Two Swedish studies have evaluated long-term effects but neither detected any impact on job satisfaction, burnout or moral stress.13 14 In a Dutch study evaluating “moral case deliberation”, the learning was valued highly and the respondents felt more ethically competent.11

In the present study philosopher-ethicists led interprofessional ethics rounds regarding dialysis patients. Known patients were discussed with real social context,1 which provides for more effective learning through the use of an actual critical situation. The goals were to promote mutual understanding and stimulate ethical reflection by helping to identify and analyse ethical problems,15 but the philosophers were instructed not to offer solutions. The solutions were to be dealt with after the rounds. In a previous study, both goals have been evaluated through interviews with a sample of the ethics rounds participants, which revealed that the goals were partly fulfilled. Positive experiences included stimulation to broadened thinking and a sense of connecting between professional categories; negative experiences were associated with frustration with the lack of solutions.16

In summary, there seems to be a further need of European studies to illuminate reflective ethical practices. The aim of the present study was to evaluate whether the ethics rounds stimulated ethical reflection.

METHODS

This study had a descriptive and comparative design based primarily on qualitative and secondarily on quantitative data.

Model for ethics rounds

Philosopher-ethicist-led interprofessional rounds concerning patient care ethical problems were applied in nephrology departments at three Swedish hospitals. Dialysis patient cases were selected on the basis of their existing life-sustaining treatment and their expected capacity to make decisions. Four philosophers from a national network that assists professionals with clinical ethical problems were recruited. The rounds were held regularly in each department every other month for 8 months and each session lasted 90 minutes. A further description of the model is presented elsewhere.16

Questionnaire

Evaluation of the stimulation of ethical reflection was operationalised in terms of comparing the rating of expectations of gaining new insights versus reported insights and comparing descriptions of reasoning about the ethical problems before and after the ethics rounds. Based on the literature3 17 and two pilot tests, two study-specific questionnaires were developed with both open and closed-ended questions (table 1). For the closed answers a five-point adjective scale was used to measure the rating of expectation about gaining insights and new insights: none (1); low (2); moderate (3); rather high (4) and high (5). Insights were defined in this study as new thoughts, new angles of approach or increased knowledge about an ethical problem.

RESPONDENTS AND DATA COLLECTION

The sample comprised all staff at the three nephrology departments. They were informed of the opportunity to participate in the rounds during departmental meetings. Of 200 available personnel, 103 persons attended one to four rounds of a total of 12 ethics rounds, resulting in 194 attendances (table 2) and a median of 14 participants per round. Selection was non-random mainly because the staff working the day of the rounds had the opportunity to participate, on average 20 persons. Not being familiar with the patient and time constraints were reported as reasons for not attending. The first questionnaire was handed out immediately before each session and the second directly afterwards. The internal drop-out was nine not answering the first questionnaire and five the second. For separate questions it ranged from 1% to 24% (mean 4%).

Analysis

Qualitative analysis

First, the open-ended answers in the questionnaires were analyzed through inductive content analysis with co-assessment.18 The text was divided into meaning-units and preliminary subcategories were generated by constant comparison using the software NVivo.19 Those subcategories that shared similar meaning were collapsed, sorted and abstracted into main categories. Second, the analysis was then transformed into a deductive “directed content analysis”,20 because of the discovery of a pattern in the answers. This pattern seemed to be in line with the theoretical framework of the principles approach, persons approach and perspectives approach presented in a guide for practical ethical analysis by Boyd (see below).21 The third and fifth author examined these approaches to use as coding categories. Then the subcategories were categorised into new main categories and sorted into the three approaches, which involved a comprehensive process of moving between the empirical data and the framework in a continuous process of refining categories and sorting data. All co-authors scrutinised and discussed each step in the analysis process. Finally, the number of meaning-units for each subcategory were counted.

A principles approach is a theory-driven approach that focuses on whether a particular act is morally right21 and might resolve the ethical problem.22 What makes an act right depends on the moral principle being relied on, such as deontological and teleological theories. A person’s approach focuses on the moral agent, the person who performs the act. It is referred to as virtue ethics, which is concerned with the best kind of person to be and desired virtues, such as justice, kindness or bravery. A perspectives approach focuses on the case, which implies understanding of not just one person but, rather, a problematical situation.21 Boyd21 referred this last approach to hermeneutic ethics, which considers multiple contexts such as the psychological and social and acknowledges that multiple interpretive perspectives exist. This approach seeks to highlight complexities22 and implies interpretation through openness to different perspectives, which may lead to awareness of one’s prejudices and a new shared perspective among individuals.21 22

Quantitative analysis

The closed-ended answers (table 1) were analysed by descriptive and comparative non-parametric statistics. For the comparison between the expectation of gaining insights (question 1) and reported new insights (questions 4, 8 and 9), the sign test was calculated.23 To detect any effect of respondents attending more than one ethics round, a subgroup analysis was made with random selection of all participants from a collective list of all ethics rounds (n  =  103). Factors from the interview study16 suspected to affect perceived insights were profession, department and philosopher. These were tested using Kruskal–Wallis and then Mann–Whitney tests,23 calculated on the subgroup of 103. For the open-ended answers subcategories derived from the qualitative analysis were computed according to frequency of meaning-units.

Ethical considerations

Ethical considerations regarding the ethics rounds are published elsewhere.16 Ethical approval was obtained from the Regional Ethical Review Board in Uppsala, Sweden. The informed consent was based on written information regarding the study and voluntary participation. The questionnaires were returned to a box after the rounds, and confidentiality was guaranteed.

FINDINGS

To illustrate the ethics rounds, descriptions of the patient cases and a representative example of perceived insights are presented in table 3.

The median rating of expectation before the ethics rounds of gaining insight was rather high (4). Forty-one per cent of the respondents reported after the rounds a high to rather high rating (4–5) of new insights and 35% a moderate rating (3). A moderate rating was the median result found for all insights perceived (question 4), insights gained specifically from the ethicists or from other staff members (questions 8, 9). The ratings of reported insights calculated for all insights (question 4) as well as those gained specifically from the ethicist (question 8) were both statistically significantly lower than the expectations of gaining insights (p<0.001) (table 4). After random selection of all participants from a collective list of all ethics rounds (n  =  103), there was a higher percentage of dissatisfied respondents (59%, p<0.001). The 45 (24%) that reported a low rating of or no new insights gave the following reasons: they had already heard it all before from other staff members (21 meaning-units); had expected solutions to the ethical problems (13); the philosopher was too passive (13); they felt they already knew the answers (10) or were not familiar with the patient (7).

Regarding factors hypothesised to affect reported new insights, the median rating (rather high) of reported insights was significantly higher in rounds led by one of the philosophers (p = 0.02). The doctors had significantly lower expectations (moderate) of gaining insights compared with the other professions (p = 0.001). No significant differences were detected between departments.

Perceived ethical problems before and new insights after the rounds

The main findings are presented in tables 5 and 6. Complementary descriptions and comparisons before and after rounds are given below. Each subcategory had representation from an average of six rounds (range 3–11) and from all professions and at least two departments.

Two major problems were experienced with regard to the 12 patients: non-compliance and end-of-life issues (table 3). The respondents described the origin of the problems as similar but with different approaches, which could be sorted into principles, persons and perspectives approaches.

Principles approach

This approach focusing on the act and moral principles dominated before the rounds (74% of the meaning-units) compared with after (10%) (tables 5 and 6). The majority of the ethical problems before rounds related to the principle of respect for autonomy, non-maleficence and justice. Non-maleficence concerned suffering, such as pain, anguish or confusion. It was difficult to know when to withdraw dialysis, considering prolonged suffering and improved health. Respect for autonomy concerned patients whose participation was considered to be problematical because the patient was either too ill to have the capacity or lacked awareness of his/her illness and thus acted in a self-destructive way. There was ambiguity about who had the responsibility to decide about treatment—the doctor, the family or the patient—and in whose interest it was to continue treatment. Lack of adequate information, such as withholding the truth about a poor prognosis or giving false hope, was perceived before rounds as hindering patients’ making important decisions. After the rounds the importance of adequate information was the only principles insight mentioned.

Persons approach

Respondents who reflected on their role as moral agents considered their personal responsibility both before and after rounds (tables 5 and 6). Beforehand they could experience difficulties in helping suffering or aggressive patients, which could arouse feelings of powerlessness, frustration and of tormenting the patients. Nurses wondered about which mental posture to have with non-compliant or suicidal patients, should they be persuasive or forceful. After the rounds the respondents perceived insights of their reactions and how they affected their actions. They also perceived insights about boundaries for responsibility, which could imply either widening or limiting. Limiting could imply insight into the association between what one ought to do and can do. Widening could mean acknowledgement that ethical problems are everyone’s responsibility instead of blaming others. Doctors perceived insights about not being responsible enough.

Perspectives approach

This approach, focusing on the case and the understanding of the situation, dominated after the rounds (72% of the meaning-units) (table 6) in contrast to before (14%) (table 5). Before the rounds there were descriptions of difficulties in understanding the patient, such as why they were being aggressive or expressing a desire to die but still showing up for dialysis. After the rounds respondents extended their perspective on the patient, such as stating they could see aggression as a possible sign of crisis. When the patient conveys death wishes to the nurse but not to the doctor, this might mean a feeling of hopelessness but not wanting to die. The rounds were perceived to produce more questions than answers, giving the insight of more dilemmas and that the right answer was impossible to reach. Respondents realised how little they had understood and questioned their pre-understanding of the situation.

Besides new insights directly connected to patients’ situations, there was an increased awareness of relations to other professions. This implied understanding how other professionals think, especially nurses gaining an increased understanding of doctors. They understood the doctors’ loneliness in trying to make the right decisions. The doctors, on the other hand, learned to understand other staff members’ feelings of powerlessness and emotional connection to long-term patients. There was also an insight that the professionals shared the view of the situation as being difficult, hearing that others have the same thoughts. Others had their pre-understanding confirmed, reinforcing how far doctors and nurses stand from each other.

Perceptions before rounds of how the ethical problems should be solved and new insights afterwards

Principles approach

The principles approach was used about twice as often before rounds as after (tables 5 and 6). At both times, suggestions for problem solving centred primarily on promoting patient participation. Compared with before rounds, the new insights were more specific, such as improving routines for information and documentation and assessing the patient’s decision-making capacity when helping him or her with decisions.

Persons approach

To be brave seemed to be a desired virtue before the rounds. The focus was to become more assertive in relation to the doctors, such as questioning why dialysis treatment continues, but also daring to discuss sensitive issues with staff members. Although some respondents described questioning the doctors after the rounds, the new insights were usually focused on being assertive in relation to the patients. This may be interpreted as being firm rather than brave, as patients are in an inferior position. Respondents received strength from the rounds to confront non-compliant patients about making contracts and drawing clear limits on what is allowed medically.

Perspectives approach

The perspectives approach regarding how to solve ethical problems dominated the other approaches both before and after rounds. The same solutions were described after rounds as before but most came from different respondents. They were either focused on enhancing patient/family contact or staff collaboration (tables 5 and 6).

Enhancing patient/family contact implied giving psychological support, accomplished through close and honest contact. After rounds the additional insights involved helping patients mourn being severely ill or encouraging the healthy person. This also implied trying to understand the patient as a whole human being and understand the non-compliant behaviour.

Regarding staff focus, there were frequent requests for interprofessional dialogue to understand doctors’ reasoning but also to obtain a whole picture of the patient with input from different professions and an opportunity to share different perspectives. After the rounds there was increased insight about the need to enhance the dialogue between doctors and nurses. This was suggested by half of the doctors responding. There were also insights about the need to implement regular team conferences, described similarly before and after rounds. Respondents wished to reach a consensus for care in order to achieve a common attitude towards non-compliant patients and avoid burdensome treatment for the severely ill.

Evaluation of the goal to stimulate ethical reflection

The goal, operationalised as new insights, was not completely reached. Although 76% of respondents reported a moderate to high rating of new insights, 64% reported insights as low or lower than expected (table 4). Furthermore, reasoning about the ethical problems produced more meaning-units before the rounds than after.

DISCUSSION

The ethics rounds did not seem to stimulate the ethical reflection that the respondents had expected. Dominating new insights did not seem to fit into traditional normative ethics, but was instead interpreted as hermeneutic ethics, such as extended perspectives on the patient and increased awareness of relations to other professions. The request for further interprofessional dialogue to solve ethical problems dominated and this correlates with findings from the ethics rounds interview study.16 It might be beneficial to speculate about reasons for unmet expectations, as the reasons given were not exhaustive and this may add to suggestions of how to make reflective ethical practice successful.

First, the respondents might have expected to gain principle-based insights about patient participation and suffering. The perspectives approach (hermeneutic ethics), involving gaining new perspectives, does not provide quick solutions or resolutions to problems, which clinical practice seems to request.22 One of the reported reasons for not perceiving insights was expected solutions to the ethical problems, which correlate with nurses’ experiences in the interview study.16 Schneiderman et al3 showed, among other things, that ethics consultation to help identify and analyse ethical problems yielded high staff satisfaction. This may be due to the simultaneous help with solving, which was also ranked high. The findings in the present study may, however, not be interpreted as an overall negative result; they might show how ethics works in clinical practice, as a time-consuming collaboration. Rather than resulting in a resolution or change in a healthcare provider’s basic stance, it may lead towards a richer interpretation.22 This may be a sign of enhanced ethical competence, which might help in solving ethical problems for future patients.

Second, the type of leadership in the ethics rounds seemed to affect reported insights. It might seem controversial to use philosophers as clinical ethicists. It can be defended because of philosophers’ methodological ability to build careful reasoned analysis24 and because an outsider may bring new perspectives to healthcare professionals, who may be blind to their own prejudices.22 Comments in this study and experiences from the interview study16 revealed both positive and negative attitudes. Positive experiences in the interview study were associated with the philosopher provoking participants to break from habitual ways of thinking, whereas negative experiences were associated with failure to make the knowledge applicable to the real world.16 Positive comments as secondary findings in the present study were associated with the philosophers stimulating critical reflection and clarifying ethical problems; the negative comments concerned passivity. According to Fox et al,1 fewer than 5% of the ethics consultants in North America are philosophers and they saw the lack of ethics education as a cause of concern. The style of leadership, however, seemed to be more crucial than the type of profession. There was a significantly higher rating of insights from one of the philosophers in this study. This philosopher was admired in the previous interview study for being structured and for maintaining discipline as well as balance in power between the nurses and doctors.16 This perhaps implies that with discipline more voices and thereby perspectives are able to come forth.

Third, some of the patients’ problems were of long duration and seemed difficult to solve. This might explain negative perceptions because the respondents felt they had already heard it all before from other staff members and results from the interview study showed a sense of resignation to a lack of change.16 This may indicate the need for proactive ethics rounds before ethical problems become severe. Furthermore, the insight of promoting patient participation seems obvious and the absence of it perhaps demonstrates how difficult it is to achieve. This is confirmed by studies showing low patient participation in end-of-life decisions.25 Reflections about how to make patients comply with treatment occurred frequently in the answers. A new insight was to give psychological support to patients and another, which is not recommended in previous studies, was to make demands on them. Instead, empowering and providing a reward system is recommended.26 It seems reasonable to assume then that not all insights from the ethics rounds are morally acceptable. There were similar categories before and after rounds, especially on how to solve problems, which would confirm the perceptions of not gaining new insights. The categories were derived mainly from different individuals, however, which might imply that the respondents received insights from each other.

It could be argued that instead of answering about new insights into ethical problems, many respondents seemed to describe how the dialogue of the ethics rounds worked, a prerequisite for the ethical process. This seems to be in line with the hermeneutic and dialogical ethics method used by Molewijk et al,11 which involves seeing another persons’ point of view and learning is the result of one’s extended perspective. This was illustrated in the categories “Extended perspective on the patient” and “Increased awareness of relations to other professions”. These reflections could be interpreted as not ethical reflections. We decided not to judge, however, but instead to investigate what the respondents perceived as ethical insights. This led to Boyd’s approach21 involving adding hermeneutic ethics to the traditional normative approaches of principle and virtue ethics. Hermeneutic ethics is, according to Leder,22 distinct from other theories such as sociological theories as it sheds light on what should be. As Leder has pointed out,22 hermeneutic ethics is not in opposition to other ethical discourses but is the very space of dialogue wherein they may be articulated.

Methodological considerations

Many of the findings from this study confirm the findings from the previous ethics rounds interview study.16 The combination of different data collection modes may be seen as a triangulation approach, which strengthens the trustworthiness.

Even though the ethics rounds were based on only 12 patient cases, the perceived ethical problems involved seem to be in accordance with the dominating problems experienced by nurses in industrial countries.27 They did not seem to be specific for each patient case, as all subcategories were identified in several rounds. All rounds except one (round 4) seemed to concern life-sustaining treatment, as non-compliance to dialysis treatment is associated with high mortality.28 29 The reasoning may therefore be transferable to reflective ethical practice in wards caring for patients with life-sustaining treatment.

The reasoning about the ethical problems produced a smaller number of meaning-units after the rounds than before. This could signal either a lack of insight or difficulties in reflecting immediately after the rounds (which some respondents mentioned). Letting the respondents answer later was considered but was rejected because of the risk of a high dropout rate. The participants were informed that ethical problems would not be solved during the rounds; instead they were to deal with them afterwards. There was, however, a question posed before the rounds enquiring into how ethical problems might be solved, which may have raised the expectation that solutions to problems would be forthcoming.

Neither previous Swedish studies using instruments for evaluating reflective ethical practice showed positive results. There were methodological reflections over the shortcomings of the conduct of the research, applying ethics rounds over too short a time and the chosen scales’ inability to measure complex social processes.13 14 These reflections also seem to be of concern in the present study. When ethical reflection is not aimed at solving problems, contrary to American ethics consultations,3 evaluation measures seem difficult to find.

Despite the strength of the rigorous co-assessment from the co-authors, there were some problems associated with the sorting of Boyd’s approaches.21 One was distinguishing the principles approach from the perspectives approach, such as in the category “Reach a consensus for care”. With this kind of data it seems impossible to find distinct categories and implies that the frequencies of subcategories cannot be exact. Boyd’s approaches21 are not really a theory, rather a guide for teaching purposes and in practice not mutually exclusive. Each approach may be useful at different times and in combination.

CONCLUSION AND IMPLICATIONS

Even though the expected ethical reflection was not realised, this study shows the need for interprofessional reflective ethical practice. It may offer an understanding of how ethics might work in clinical reality with known patients that is context sensitive. To make reflective ethical practice successful, we suggest, as in the previous interview study (see model),16 a balance between ethical reflection and problem solving.

Participants’ expectations of outcomes of ethics rounds might be explored before initiating any further studies. Because both the present and the previous findings16 suggest that process skills30 are important, further studies are needed to explore what kind of ethicist leadership is most beneficial for the outcomes. The responsibility for clinical leadership to arrange and motivate ethical reflection is also of interest.