Article Text
PDF
Abstract
Objectives The aim of this study was to investigate concordance between patients and non-professional carers about factors associated with a ‘good death’ and other end-of-life decisions.
Methods Patients completed a questionnaire about end-of-life care issues, and were asked to rank the importance of factors linked to a ‘good death’. Carers also completed a questionnaire about end-of-life care issues relating to the patient, and whether or not they agreed with those choices (ie, medical treatments, PPD). Carers were also asked to rank the importance of factors linked to a ‘good death’ to the patient, and to them personally at that point in time.
Results Only 69% of patients stated they had discussed their preferences for end-of-life care with their respective carer. The rankings were similar for the patient and the carer's views of what was important for the patient, although the patients ranked ‘to be involved in decisions about my care’ as less important than the carers, while the carers ranked ‘to have sorted out my personal affairs’ as less important than the patients.
Conclusions When discussions around end-of-life choices do occur, carers generally appear to agree with the patients' preferences around end-of-life treatment, and preferred place of death.
- Cancer
- Terminal care
- End-of-life care
- Good death
- Patient
- Carer
Statistics from Altmetric.com
Introduction
The End of Life Care Strategy was developed with the aim of ‘promoting high quality care for all adults at the end of life’ in England.1 It highlights the importance of non-professional carers (‘carers are central to the team that cares for somebody at the end of life’, and they should be treated as ‘coworkers’ with the health and social care team), but notes that ‘the carer's needs and wishes may conflict with those of the dying person’.
Caring for patients with advanced cancer can be a very positive experience for carers (and patients),2 but it can also be associated with a variety of negative physical, psychological and social consequences.3 Moreover, the burden of caring may be exacerbated if there is non-concordance around end-of-life care decisions, and especially in relation to preferred place of care (PPC)/preferred place of death (PPD).
In a previous study, we investigated the relative importance of previously identified factors associated with a ‘good death’ in a cohort of patients with cancer from a single cancer centre.4 The aim of this study was to replicate the study in a more heterogeneous group of patients with cancer, but more specifically to investigate concordance between patients and non-professional carers about factors associated with a ‘good death’ and other important end-of-life decisions (eg, PPD).
Ethics
The study was sponsored by the Royal Surrey County Hospital, and received ethical approval from the NRES Committee South East Coast—Surrey.
Methods
The study was a prospective, observational study. The study was conducted at two hospitals (Royal Surrey County Hospital, Guildford; Ashford & St. Peter's Hospital, Chertsey), and five hospices (Princess Alice Hospice, Esher; St. Catherine's Hospice, Crawley; Woking Hospice, Woking; Sam Beare Hospice, Weybridge; St. Margaret's Hospice, Taunton) in the UK.
The study population consisted of patients and a cohabiting non-professional carer (family member). The inclusion criteria for patients for the study were: (1) diagnosis of locally advanced/metastatic cancer; (2) age >18 years; (3) known to relevant hospital/hospice specialist palliative care team; and (4) presence of a cohabiting non-professional carer. The exclusion criteria for patients for the study were: (1) inability to give informed consent; (2) inability to complete the study questionnaire; and (3) estimated prognosis <1 week. Similarly, carers had to be able to give informed consent, and be able to complete the study questionnaire.
Potential participants were given an information sheet about the study, time to consider the study (at least 24 h), and the opportunity to discuss the study (with researchers/other persons). The information sheet was slightly different for the patients and their carers (to reflect differences in the content of the study questionnaires). The consent sheet was also slightly different for the patients and their carers, and the patient and their carer needed to provide written consent prior to the start of the study. The study questionnaires were administered sequentially/independently to the patient and their carer.
Data were collected about the patient's demographics, home postcode (to determine indices of deprivation), cancer diagnosis and Eastern Cooperative Oncology Group (ECOG) performance status. Patients were asked to complete a questionnaire about a variety of end-of-life care issues, including PPD, ‘acceptable’ places of death, possession of a last will and testament, possession of an advance care plan, and whether or not they had had any discussions with their carer regarding end-of-life care choices. They were also asked to rank the importance of factors previously linked to a ‘good death;’5 these factors were ‘to feel my life is complete’, ‘to be at peace with my God’, ‘to be involved in decisions about my care’, ‘to be able to die at home’, ‘to have sorted out my personal affairs’, ‘to have sorted out my funeral arrangements’, ‘to not be a burden to my family’, ‘to not be a burden to society’, ‘to be mentally alert (until death)’ and ‘to have my pain/symptoms well controlled’.4 Patients were also asked to complete the Memorial Symptom Assessment Scale—short form (MSAS-SF).6
In addition, data were also collected about the carers’ demographics. Carers were asked to complete a questionnaire about a variety of end-of-life care issues relating to the patient, including whether or not they had had discussions with the patient regarding end-of-life care choices, and whether or not they agreed with those choices (ie, medical treatments, PPD). They were also asked to rank the importance of factors previously linked to a ‘good death’ to the patient, and to themselves (see above).4 ,5 Carers were also asked to complete the Zarit Burden Interview,7 which is a validated/recommended tool for assessing the impact of informal caring.8
The data were analysed using SPSS (V.19) software. Descriptive statistics were used to report much of the data. The summed ranking of factors important for a good death was determined by multiplying the number of patients/carers by a ‘score’ attached to the relevant factor ranking.4 Thus, the ‘most important’ factor was allocated a score of 7, the second-most important factor was allocated a score of 6, and so on (with the ‘least important’ factor allocated a score of 1). Univariate binary logistic analysis was used to determine the relationship between patient's PPD (home vs elsewhere) and other recorded variables (ie, age, gender, index of deprivation, cancer diagnosis, ECOG performance status, physical symptoms and psychological symptoms). Equally, univariate binary logistic analysis was used to determine the relationship between carer's preference for patient's death (home vs elsewhere) and other recorded variables (ie, age, gender, relationship and carer burden). A p value of ≤0.05 was deemed to be statistically significant.
Results
In total, 120 patient/carer dyads participated in the study; 45 patients were recruited from hospitals, and 75 were recruited from hospices. The median age of the patients was 67 years (range 37–93 years), and there were 57 men and 63 women. Table 1 contains further demographic information about the study patients. The median age of the carers was 65 years (range 21–84 years), and there were 47 men and 73 women, of which 99 carers were spouses (41 husbands, 58 wives), 5 were ‘partners’ (3 men, 2 women), 12 were children (3 sons, 9 daughters), 2 were parents (2 mothers), 1 was a sibling (sister), and in 1 case the relationship was not stated.
Demographic information about study patients
One hundred and three (86%) patients had a last will and testament, but only 33 (27%) had an advance care plan (ie, 15.5% patients recruited from the hospitals, 34.5% patients recruited from the hospices). Of note, some carers (n=13) were unsure whether the patient had an advance care plan (ACP), some carers (n=11) were unaware that the patient had an ACP, and some carers (n=10) mistakenly thought that the patient had an ACP.
Nevertheless, 83 (69%) patients stated they had discussed their preferences for end-of-life care with their respective carer (‘have you discussed your wishes about medical treatment if you were to become seriously ill (and unable to make decisions for yourself) with your family?’), with 80 carers stating that such a discussion had taken place. Of the 40 carers who stated there had been no discussion, only three were not the spouse or partner (ie, 2 daughters, 1 son). Similarly, 87 (73%) patients stated that they had discussed their PPD with their respective carers, with 89 carers stating that such a discussion had taken place. Of the 31 carers who stated there had been no discussion only four were not the spouse or partner (ie, 1 daughter, 1 son, 1 mother, 1 sister). One hundred and one (84%) carers had a last will and testament, and 30 (25%) had an ACP. Interestingly, one-third (10) of carers with an ACP were paired with patients without an ACP.
Table 2 contains information about patients’ PPD, and also ‘acceptable’ places of death (APD). Home was the most popular APD, and the most popular PPD (proportion: 50.4%; 95% CI 41.4% to 59.4%). Hospital was an APD for 21% patients in total, but 35% patients recruited from the cancer centre (ie, Royal Surrey County Hospital). In general, the carer supported the patient's PPD (when aware of their decision): 90% agreed, 4.5% disagreed, and 5.5% did not know whether they agreed or not. Similarly, in general, the carer supported the patient's other preferences for end-of-life treatment (when aware of their decision): 96% agreed, 3% disagreed, and 1% did not know whether they agreed or not.
Preferred place of death and acceptable places of death for study patients
Patient's PPD (home vs other sites) was not associated with patient demographics (age, gender), index of deprivation, cancer diagnosis, ECOG performance status, presence of physical symptoms and/or presence of psychological symptoms (all p >0.050). It should be noted that patients had a median of seven (range 1–12) physical symptoms, and a median of three (range 0–6) psychological symptoms. Carer's PPD for the patient was not associated with carer demographics (age, gender, relationship), but was associated with Zarit Burden category. Thus, carers in the ‘moderate to severe’ burden category were less likely to choose home as the PPD (Fisher's exact test: p=0.037).
Table 3 and figure 1 show the summed rankings of the importance of factors linked to a ‘good death’ for patients and carers. The rankings were similar for the patient and the carer's views of what was important for the patient, although the patients ranked ‘to be involved in decisions about my care’ as less important than the carers (carers—3rd most important factor; patients—5th most important factor), while the carers ranked ‘to have sorted out my personal affairs’ as less important than the patients (patients—3rd most important factor; carers—5th most important factor). Equally, the rankings were similar for the patient and the carer's views of what was important for themselves, although the carers ranked ‘to be able to die at home’ as less important than the patients (patients—6th most important factor; carers—8th most important factor).
Ranking of factors important for a ‘good death’ by patients and their carers
Summed rankings of the importance of factors linked to a ‘good death’ for patients and their carers.
Discussion
The results of the study suggest good concordance between patients and carers in terms of end-of-life care treatment (96%), and PPD (90%). However, only 69% patients had discussed end-of-life care treatment with their carer, and only 73% patients had discussed PPD with their carer, in spite of a diagnosis of advanced cancer (and a referral to a specialist palliative care team). By contrast, Stajduhar et al9 reported limited concordance between Canadian patients and carers in terms of PPD (49.3%). Thus, the carers were more likely to choose hospital, while the patients were more likely to state ‘does not matter’. Other studies, in other countries/cultures, have also reported lower levels of concordance for end-of-life treatments10 ,11 and PPD.12 ,13
An explanation for the apparent disparity of our results relates to the methodology of the study: in previous studies the carers have been asked to state their own preferences,9 while in our study the carers were asked whether or not they agreed with the patient's preferences (and we did not check that the carer's understanding of the patient's preferences was accurate/up-to-date). Thus, end-of-life decisions often change over time, especially in terms of PPC/PPD. For example, Agar et al14 reported that 32% of patients changed their PPD over the course of their study, with many patients changing their minds more than once (and less patients wanting to die at home at the end of the study compared with the beginning of the study).
Most patients in our study had a last will and testament, but relatively few patients had an ACP. Advance care planning is an integral part of the End of Life Care Strategy in England,1 and is actively encouraged by all the specialist palliative care teams engaged in the study. Advance care planning has been associated with positive outcomes for patients (eg, decrease in hospitalisations at the end of life, decrease in life-sustaining interventions at the end of life),15 but may also have negative outcomes for patients and carers (eg, increase in distress, increase in interpersonal stress).16
Advance care planning is unusual in the general population,17 and there has been somewhat limited uptake within healthcare populations.18 Abel et al19 reported that 56% of patients in their hospice completed an ACP, despite the document being routinely distributed to patients (and carers). It should be noted, that in this study the presence of an ACP was validated by the recording of PPD in the hospice electronic patient record (rather than the presence of a formal document). Reasons for non-completion of an ACP are varied, and include patient factors, healthcare professional factors, healthcare service factors and legal issues.18 Of note, an intensive intervention to promote ACPs in Australian patients with lung cancer resulted in only 30% patients completing a formal document.16
The concept of a ‘good death’ varies across cultures,20 ,21 and also within the same culture.5 ,22 In this study, the most important factors associated with a good death for the patients were ‘to have my pain/symptoms well controlled’, ‘to not be a burden to my family’, ‘to have sorted out my personal affairs’, ‘to be mentally alert (until death)’ and ‘to be involved in decisions about my care’, respectively. Nine patients ranked ‘to be able to die at home’ as their most important factor, and overall, patients ranked this domain as the sixth most important factor. These patient rankings of factors that are important for a good death are similar to those reported in an earlier study in the UK.4 Moreover, other studies have reported the relative lack of importance of place of death to patients,5 ,23 and, indeed, to members of the general public.24
In general, there was reasonable concordance between patients and their carers regarding the relative importance of factors associated with a good death. However, carers thought that patients would be more concerned about being involved in decisions, and less concerned about sorting out their personal affairs. Furthermore, carers were even less concerned about themselves dying at home (eighth most important factor). Other studies have also reported differences in the priorities at the end of life between patients and carers,23 and between patients and healthcare professionals.5 ,25
The main limitation of the study is the fact that it was conducted among a mainly white population that lived in a generally privileged (as opposed to deprived) areas of the UK. As a result, it is difficult to extrapolate the data to patients with cancer in the rest of the UK. Equally, it is difficult to extrapolate the data to patients with non-malignant disease (especially as previous studies suggest important differences between patients with cancer and patients with non-malignant diseases).9
Conclusion
Many patients with advanced cancer appear reluctant to discuss end-of-life choices with their carers, and even among those who do discuss end-of-life choices with carers, there is a reticence to complete an ACP (but no reticence to complete a last will and testament). When such discussions do occur, carers generally appear to agree with the patients’ preferences around end-of-life treatment, and PPD.
One important finding of this study is the relative importance to patients of sorting out their ‘personal affairs’. Hence, it is important that end-of-life discussions do not just focus on healthcare issues such as the patient's PPD, but also address practical issues such as the patient's personal affairs. Moreover, such discussions need to be done in a timely manner, since it may take a long time to sort out certain personal affairs.
References
Footnotes
Acknowledgements The authors would like to acknowledge the important contributions of Alison Buchanan, Dr Lucy Harris, Dr Liz Reed, Dr Kate Shorthose and Kabir Mohammed (statistician).
Contributors AD came up with the idea for the study, wrote the protocol, was chief investigator on the study, and wrote the paper. MW helped to write the protocol, was lead research nurse for the study, and contributed to the paper. All the other authors reviewed the protocol, were principal investigators on the study, and contributed to the paper.
Funding The study was funded by the Palliative Care Research Fund at the Royal Surrey County Hospital.
Competing interests None declared.
Ethics approval NRES Committee South East Coast—Surrey.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement We are willing to share our data with other healthcare researchers (subject to a written agreement). We do not have any unpublished data from the study.