Ethical issues experienced during palliative care provision in nursing homes

Design

A two-phase, exploratory, sequential, mixed-methods design based on the taxonomy of Creswell and Plano Clark ¹² was identified as the most appropriate design to fully address the research aim. Insufficient prior research was available to justify an explanatory design; therefore, it would be inappropriate to conduct the quantitative research first. In addition, the information gained from the qualitative exploration was used to shape the quantitative survey, and therefore the two methods could not run in parallel. The first phase focused on RN and HCA experiences of ethical issues and the second phase focused on understanding the frequency and level of distress reported by RNs and HCAs associated with these issues across Northern Ireland. Data collection for Phases 1 and 2 occurred between December 2014 and November 2015 (Figure 1).

OPEN IN VIEWER

Figure 1.

Research design.

Participants

In Phase 1, semi-structured interviews were undertaken with the purposive sample of RNs and HCAs from 13 nursing homes across Northern Ireland who accepted an invitation to participate. Staff were recruited through the provision of an information pack and oral presentation about the study by the researcher. If they were willing to participate, they completed the consent form and the researcher contacted them directly using the provided contact information. Participants were recruited until saturation was reached and no new themes were emerging from the dataset, which was anticipated to be approximately 12 interviews based on the research literature. ¹³

In Phase 2, a cross-sectional survey was conducted with all RNs and HCAs employed in 18 nursing homes across Northern Ireland who were working on the day of data collection. Based on feedback from Phase 1, the researcher recruited in person at the nursing homes at arranged times, and staff were invited to participate. In the United Kingdom, no data were available to indicate the total number of RNs and HCAs employed in nursing homes, and therefore a power calculation was not possible. To allow for an exploratory factor analysis (EFA) which was conducted as part of the larger study, Costello and Osbourne ¹⁴ recommend a sample size of 200. To ensure safe staffing levels, the Regulation and Quality Improvement Authority ¹⁵ recommend a ratio of 35% RNs to 65% HCAs; therefore, 70 RNs and 130 HCAs were considered the desired sample size. A convenience sample of 23 nursing homes was approached and 18 responded positively to an invitation to partake. The total number of RNs and HCAs employed per home ranged from 9 to 100 (mean = 33.2, standard deviation (SD) = 24.4).

In both phases, the inclusion criteria were as follows: currently employed in the nursing home as an RN or HCA; over 18 years old; and could provide informed consent.

Data collection

In Phase 1, qualitative, audio-recorded, semi-structured interviews using the critical incident technique ¹⁶ were used to ascertain the key ethical issues surrounding palliative care provision within the nursing home setting. An interview guide was developed based on a previous study ¹⁷ and pilot tested prior to use. Each participant undertook one interview. Individual interviews were held in a private room within the nursing home and lasted for 11–67 min (median = 30 min).

In Phase 2, a cross-sectional survey was used to assess the frequency and level of distress experienced by RNs and HCAs associated with ethical issues during palliative care provision in the nursing home. The survey was undertaken within the nursing home in which they were employed, located across Northern Ireland. Participants completed the surveys during their shift and returned them to the collection point on the same day.

The ‘Ethical Issues in Palliative Care for Nursing Homes’ (EPiCNH) ¹⁸ was used to collect data. The survey was developed through a literature review, interviews with key stakeholders to develop items, face validity testing with an expert panel and pilot testing with the population of interest. The scale consisted of 26 items, scored on two subscales: the Frequency Scale and the Distress Scale (further details on the psychometric properties of the instrument have been previously published ¹⁸ ). Participants were asked to rate each item from 0 to 4, with a score of 0 indicating that the item did not occur frequently or cause any level of distress, and a score of 4 indicating high frequency or distress.

Analysis of data

Phase 1 data were stored and managed through NVivo 10 software. And 10% of the audio recordings and transcripts were independently reviewed and analysed by the research team to ensure transcript accuracy and to increase credibility of the interpretation. Interviews were transcribed verbatim and analysed concurrently through thematic analysis. ¹⁹ This included a six-step process of familiarisation, generating codes, searching for themes, reviewing themes, defining and naming themes and producing a report.

Survey data (anonymised) were entered and managed using IBM SPSS Statistics v22. Ten percent were checked for transcription errors by a member of the research team (D.M.). Descriptive statistics were used in Phase 2 to summarise what was revealed through the survey and highlight which themes were most prevalent. T-tests were conducted to look for similarities and differences between RNs and HCAs in terms of both frequency and distress.

Integration

Following the guidance of Creswell and Plano Clark, ¹² integration was undertaken through data transformation and a spiralled comparison within the discussion. The data from Phase 1 were integrated with the literature review and informed the development of the survey instrument utilised in Phase 2. Phase 1 categories linked to items in the survey which allowed the items to be ranked in terms of frequency of occurrence and level of distress associated with the theme. Results were presented in a spiralled comparison, which means that data from Phases 1 and 2 are integrated and then described concurrently.

Ethical considerations

Ethical approval was granted by Queen’s University’s School of Nursing and Midwifery Research Ethics Committee (02.DPreshaw.08.15.M7.V4). All participants were supplied with a participant information sheet which discussed confidentiality, anonymity, disclosure of sensitive information and participant welfare. Written informed consent was obtained prior to participation in the interviews, whereas consent was implied based on the return of the survey. Governance was sought from the manager of each nursing home.

Participant profile

Table 1 portrays the demographic characteristics of the sample who completed the survey, including their work designation, age, gender, education and number of years spent within the nursing home.

OPEN IN VIEWER

Table 1.

Demographic characteristics of participants.

	Phase 1		Phase 2
Characteristic	RN	HCA	RN	HCA
Total	13	10	69	129
Gender
Male	1	1	10	13
Female	12	9	59	116
Missing	0	0	3	0
Age
18–25	3	4	9	36
26–35	2	1	20	41
36–45	3	1	9	20
46–55	1	2	21	18
56+	3	1	9	8
Missing	1	1	1	6
Employed in nursing home/years
<5	5	5	34	79
6–15	6	3	17	31
16–26	2	2	18	19
Missing	0	0	3	0
Highest qualification
None	2	3	3	23
NVQ Level 2	0	3	0	29
NVQ Level 3	0	4	0	40
Bachelor’s degree	6	0	31	10
RGN	4	0	20	0
Postgraduate qualification	1	0	5	0
Other	0	0	10	27

RN: registered nurse; HCA: healthcare assistants; NVQ: National Vocational Qualification; RGN: registered general nurse.

Main findings

The interviews revealed three themes related to ethical issues reported by healthcare providers during palliative care provision: ‘Ethical issues in practice’, ‘Relational ethical issues’ and ‘Organisational ethical issues’. Data from the two phases have been presented in a spiralled comparison (Table 2).

OPEN IN VIEWER

Table 2.

Tabular summary of integration.

Phase 1			Phase 2
Ethical issue		Examples of issue	Relevant item(s) from survey	Average scores, mean (SD)
				Frequency	Distress
Professional issues/issues in practice	Prioritising residents	Truth telling, managing symptoms, acting in the resident’s best interests	1. I find it difficult to protect a resident’s rights and dignity	.87 (.92)	1.65 (1.4)
			2. At times, I have not been honest with a resident because I thought it was in their best interest	1.78 (1.10)	1.82 (1.16)
	Effective practice	Clear communication, working cooperatively, keeping clear and accurate records	3. I have to follow a request by a senior clinician not to discuss the resident’s diagnosis with them when he or she asks for it	1.07 (1.31)	1.52 (1.46)
			9. I have to initiate extensive lifesaving actions when I think they only prolong death (e.g. PEG feeding, sub-cut fluids)	.94 (1.15)	1.31 (1.31)
			18. I witness staff avoiding residents at the end of life due to their fears about dying	.69 (.97)	1.28 (1.42)
	Fear of malpractice (HCAs only)	Not knowing what to do, fear of ‘abuse’ label, receiving conflicting care instructions	11. I don’t know what to do when unclear resident care instructions have been provided by a senior clinician	.92 (1.00)	1.52 (1.36)
	Role fidelity	Working within your professional boundaries, feeling undervalued, not understanding role in decision making	6. I have made a decision in the resident’s best interest to prevent them coming to harm or unnecessary risk	2.68 (1.24)a	1.54 (1.24)
			10. I don’t know what care to provide when no advance care plan had been agreed	.95 (1.08)	1.54 (1.41)
			12. I don’t feel confident to voice my opinion regarding palliative care decisions	1.04 (1.19)	1.44 (1.43)
			13. I feel powerless in decision making during resident care	.95 (1.11)	1.49 (1.46)
Relational issues	Issues with residents	Balancing the needs of multiple residents, resident demands on staff resources, lack of understanding from residents regarding medication	4. I struggle to provide care to a resident due to their verbal or physical resistance	1.48 (1.22)	2.05 (1.36)a
			5. I have to care for residents only accepting small amounts or refusing food and fluids near the end of life	2.70 (1.19)a	2.23 (1.33)a
	Issues with families	Witnessing family distress and being unable to help, coping with criticisms and demands from families	14. I witness distress from family or care partners	2.27 (1.17)a	2.28 (1.17)a
			15. I have to follow the family’s or care partner’s wishes for the resident’s care when I do not agree with them	2.03 (1.26)a	2.26 (1.25)a
	Conflicts/ dilemmas with colleagues	Concerns over competence	7. I witness how poor staff communication results in diminished quality of care to residents	1.87 (1.17)	2.48 (1.34)a
	Staff wellbeing	Distress and guilt experienced	16. I am asked to provide care to the resident according to a senior clinician, specialist palliative care nurse or charge nurse against my personal or professional opinion	1.20 (1.28)	1.58 (1.46)
Organisation issues	Lack of resources	Lack of staff and access to external services, knowledge and awareness, benefits and support, time	19. I am not able to provide the care I want due to lack of resources within the care home	1.26 (1.30)	1.84 (1.55)
			20. I don’t have enough time to provide the resident with the care he or she needs	1.77 (1.35)	2.36 (1.44)a
			24. I am involved in non-direct care activities which reduce time spent with the residents	1.69 (1.46)	1.90 (1.48)
			8. I have observed professional incompetence due to insufficient staff training for providing nursing care	1.49 (1.25)	2.17 (1.48)a
	Organisation of services	Lack of GP involvement, the relationship with the hospital, and the structure of care within the home	21. I am unable to provide the quality of care I want due to conflicting care directions by external health and social care services	.96 (1.08)	1.54 (1.45)
			22. I find GP support lacking for resident care	1.49 (1.38)	1.89 (1.47)
			23. I have been involved in what felt like an unnecessary hospital admission	.97(1.18)	1.53 (1.47)
	Quality of services	Management of pain, care at the end of life, recognising the limits of available care	17. I am required to provide palliative care for residents I don’t feel trained to care for	.70 (1.00)	1.28 (1.45)
			25. I feel the pain management is not satisfactory	.87 (1.09)	1.51 (1.55)
			26. I have witnessed end-of-life care which I felt was not satisfactory	.62 (1.09)	1.37 (1.67)

SD: standard deviation; PEG: percutaneous endoscopic gastrostomy; HCAs: healthcare assistants; GP: general practitioner.

^a Highest scoring items.

Ethical issues in practice

Ethical issues in practice were found when participants experienced conflict between what they felt was right and what was their duty of care. This included respecting human rights, conflicting bioethical principles (autonomy, beneficence, non-maleficence and justice) and nursing ethical principles (fairness, respect of dignity and truthfulness, issues with autonomy due to a lack of advance care planning, and the use of futile or aggressive treatment). The quotes below demonstrate the challenges in being truthful, and conflicting ethical principles of resident autonomy and beneficence:

I’m a nurse so I shouldn’t lie to my resident…but sometimes it’s very difficult. (Participant O, RN, p. 4)

you do feel you’re going against their wish but at the same time, you need to get washed…or you need to eat, you’re only trying to encourage something for your own good…but you do feel uncomfortable doing it. (Participant L, HCA, p. 5)

Relational ethics

Relational issues focused on the conflicts within the relationships of staff, families and residents including attentiveness to resident care needs, responsiveness to criticism and demands from families and care partners, and concerns about the competence of colleagues. The following quotes demonstrate specific examples of conflict with families over the best interests of the resident and concerns over the compassion of a colleague:

I had a family and the mother’s 91, she has dementia and they insisted she was taken off her night sedation with the result of the woman didn’t sleep and was up all night shouting. (Participant B, RN, p. 8)

I think her [the nurse’s] manner was sharp and maybe not as sympathetic as she should have been. (Participant A, RN, p. 12)

Organisational ethics

Organisational issues were a consequence of the structure and quality of services available within the nursing home and included distributive justice (resource allocation and quality of services), the organisation of services, particularly the coordination with, and responsiveness of, locum general practitioners (GPs), and staff wellbeing. The quotes below demonstrate the personal struggle with lack of time to get to know the residents, and the quality of services to ensure residents feel like a person and not a number:

Sitting with someone and having a chat goes a long way…you just don’t have that time since coming here. (Participant M, RN, p. 5)

I don’t believe somebody should feel like a number on a door and to me…that’s the way people are. (Participant L, HCA, p. 7)

Integration of datasets

The data from Phases 1 and 2 were integrated and presented in a spiralled comparison to draw out three key findings. First, relational ethical issues are most distressing to healthcare providers when providing palliative care. Elements include attentiveness, responsibility, competence and responsiveness, and how these can conflict during daily palliative care provision. Examples of this finding include the following: interactions between RNs and HCAs and the families of patients receiving palliative care resulted in frequent ethical issues and a high level of distress; resistance from residents during care provision was very distressing for most healthcare providers; and poor communication between staff was both frequent and distressing. Second, staff training in palliative care is still lacking, but desired by healthcare providers to improve experiences of palliative care provision and reduce ethical issues. Finally, communication to facilitate shared decision making between families and the multidisciplinary team is essential to improve palliative care and reduce ethical issues.

Interpretation of main findings

Decision making to ensure fair and ethical treatment, balancing autonomy, justice, non-maleficence and effective communication to support beneficent care were the most frequently raised issues in practice, highlighting the need to address these areas directly through improved training. Managing the challenging behaviour and negative symptoms, particularly in terms of resisting care, was a clear example of the struggle between balancing justice and non-maleficence. It is also supported by previous studies, for example, Gjerberg et al. ²¹ found RNs and auxiliary nurses to report ethical dilemmas due to demanding residents who ended up with more resources and breaches in patient autonomy. The conflicting ethical principles of beneficence and autonomy were commonly cited throughout the literature, ^{22 –24} for example, disagreements between nurses and patients were reported when the best patient outcomes were limited by the patient’s wishes. This research has added to the knowledge base by demonstrating how communication challenges can lead to ethical issues irrespective of the capacity of the resident.

Relational issues were found to be the most frequent and most distressing, setting apart palliative care ethical issues from ethical issues experienced during everyday nursing care. This research added to the evidence base by demonstrating how a lack of family understanding can lead to unrealistic demands within palliative care; however, while openness with the family is important, the resident must remain central to the decision-making process. The influence of family distress on care may suggest a need to include families actively in decision-making processes from the beginning of the care journey. Interacting with family members was found to be a concern for RNs and HCAs in both this study and many previous studies including those by Enes and de Vries, ²⁵ Schaffer ²⁶ and Gjerberg et al. ²⁷ These studies found meeting relative’s needs, interacting with families, communicating about death and disagreements over treatment to be the major ethical issues. Lopez et al. ^28,29 identified nurses as feeling ‘in the middle’ of the decision-making processes and ambiguous in terms of their specific role as they wished to create a plan that was acceptable to the GP, the family and the patient. Trying to satisfy everyone within the decision-making process was identified by nurses as a concern in this study; however, HCAs previously reported getting on with the job instead of involving themselves in decision making. ³⁰

Lack of staff and problems with GP involvement were highlighted as the main organisational ethical issues, indicating a need to increase resources and improve coordination of GP care within nursing homes. Many issues involving the GP were reported within the literature. Dreyer et al. ³¹ found that homes with only one GP visit per week also had limited coordination of care, and Enes and de Vries ²⁵ found lack of GP knowledge of symptom control to be an ethical issue. Involving GPs in nursing home care proved challenging for participants both in this study and in Schaffer’s ²⁶ study, where difficulty contacting GPs was one of the most frequently occurring ethical problems. While restraint was commonly found as an ethical issue throughout the literature, ¹¹ it was not identified by participants in this research. This could be caused by a lack of understanding of restraint and how it may conflict with a person’s right to freedom, or it could be a result of better management and use of restraint.

Strengths and limitations

The main strengths of this research include the rigorous pilot testing of the interview schedule and the survey instrument prior to use; team collaboration throughout the analysis process and the practical element of a mixed-methods approach which resulted in clear and direct messages for service improvement. Limitations have also been recognised. This study was only conducted in one part of the United Kingdom, limiting its wider application. The challenges of researching ethical issues which have previously been poorly defined, and the recognition that nursing homes who are willing to participate in research may be systematically different to those who are unwilling to participate in research may limit the applicability of findings across all nursing homes.

What this study adds

The key implications for policy include the need for inclusion of ethical decision making within palliative care education and support for the United Kingdom Clinical Ethics Network’s (UKCEN) argument ³² that it is time for formal recognition of clinical ethical committees, with the services of such committees made available to nursing homes. This recommendation has global implications for palliative care provision, linking to the World Health Organization’s (WHO) ³³ acknowledgement of the need for training and debate on the ethical aspects of a range of issues for those receiving palliative care.

The overwhelming similarities between RNs and HCAs highlight the importance of the shared environment in the experience of ethical issues. The key implications for practice include the need for multidisciplinary training and support for RNs and HCAs who are struggling to build strong networks with residents, families or colleagues, and a means of incentivising training in palliative care to help overcome common ethical issues. Training staff to take account of patient and family values and incorporate this into decision making has been recognised by the WHO ³³ as a means of addressing important ethical issues. Distribution of knowledge in ethical decision making in palliative care may be achieved by support of local knowledge transfer projects which are less expensive to run and accessible remotely, for example, Project Extension for Community Health Outcomes (ECHO). ECHO is an evidence-based telementoring intervention that connects primary care providers with expert teams of specialist providers via regularly scheduled videoconference. ³⁴ In addition, the implementation of strategies to strengthen and organise services, access and referral systems that are fully integrated across healthcare delivery would address many sources of ethical issues raised in this research.

Future research may consider exploring how to develop a multidisciplinary approach to palliative care education and ethical decision making, why training in palliative care is unattended and evaluating patterns of usage to search for methods of improving attendance. In addition, researchers should seek to understand the family’s perspective when engaging with nursing homes, including the development of family-friendly strategies to increase the likelihood of families engaging in care decisions.