If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Center for Collaborative Palliative Care, Faculty of Health and Life Sciences, Linnaeus University, Kalmar, SwedenSophiahemmet University, Stockholm, Sweden
Department of Health Care Sciences, Palliative Research Centre, Ersta Sköndal University College, Stockholm, SwedenCapio Palliative Care Unit, Dalen Hospital, Stockholm, Sweden
Unit of Medical Ethics, Department of Clinical Sciences, Faculty of Medicine, Lund University, Lund, SwedenDepartment of Nephrology, Skåne University Hospital, Malmö, Sweden
End-stage kidney disease (ESKD) is characterized by high physical and psychological burden, and therefore, more knowledge about the palliative care provided close to death is needed.
Objectives
To describe symptom prevalence, relief, and management during the last week of life, as well as end-of-life communication, in patients with ESKD.
Methods
This study was based on data from the Swedish Register of Palliative Care. Patients aged 18 or older who died from a chronic kidney disease, with or without dialysis treatment (International Classification of Diseases, Tenth Revision, Sweden; N18.5 or N18.9), during 2011 and 2012 were selected.
Results
About 472 patients were included. Of six predefined symptoms, pain was the most prevalent (69%), followed by respiratory secretion (46%), anxiety (41%), confusion (30%), shortness of breath (22%), and nausea (17%). Of patients with pain and/or anxiety, 32% and 44%, respectively, were only partly relieved or not relieved at all. Of patients with the other symptoms, a majority (55%–84%) were partly relieved or not relieved at all. End-of-life discussions were reported in 41% of patients and 71% of families. A minority died in specialized palliative care: 8% in hospice/inpatient palliative care and 5% in palliative home care. Of all patients, 19% died alone. Bereavement support was offered to 38% of families.
Conclusion
Even if death is expected, most patients dying with ESKD had unmet palliative care needs regarding symptom management, advance care planning, and bereavement support.
The World Health Organization emphasizes the need for palliative care, comprising symptom management, team work, communication, relationships, and family support. Palliative care aims to prevent and relieve suffering and enhance the quality of life of patients with a life-threatening illness, such as end-stage kidney disease (ESKD), and it can be combined with treatment aimed to prolong life,
for example, dialysis treatment. Patients with ESKD whether on conservative care (no dialysis) or on dialysis suffer from a variety of symptoms; for example, pruritus, pain, fatigue, sleep disturbance, nausea, muscle cramp, restless legs, anorexia, depression, and dyspnea.
Many patients are old and have a multiplicity of comorbidities that add to their illness situation. The associated mortality is high even when the patients are on dialysis, about 20% annually.
Survival outcomes of supportive care versus dialysis therapies for elderly patients with end-stage kidney disease: a systematic review and meta-analysis.
but mortality rates imply that many patients with ESKD are living their last year of life. Hence, these patients will need discussions on prognosis and end of life to be able to make decisions and plan for end of life according to their personal values.
Their trajectory may also involve decision making regarding withholding or withdrawing dialysis. A review of patients' and family members' perspectives of end-of-life care in chronic kidney disease highlights their situation with for example growing physical and psychosocial suffering, personal vulnerability, relational responsibilities, and existential issues.
With the goal to improve end of life for patients with ESKD, it is important to explore what end-of-life care they and their families' receive. There is a need for further comprehensive studies describing the care situation and key components of palliative care activities for this group of patients when close to death. Therefore, the aim of the present study was to describe end-of-life care of patients with ESKD in terms of symptom prevalence, relief, and management during the last week of life, as well as aspects of end-of-life communication for patients and families.
Methods
Design
This registry study is based on data from the Swedish Register of Palliative care (SRPC) (http://palliativ.se). The SRPC is a national quality register for end-of-life care for all deaths in Sweden, irrespective of diagnosis or health care setting. The goal of the SRPC is to improve quality of care for all people dying an expected death.
The basis for the development of the register was the principles for end-of-life care put forward by the British Geriatrics Society, demanding, for example, competent symptom management and open end-of-life communication.
British Geriatrics Society. Palliative and end of life care for older people. BGS compendium document 4.8 2004. Revised version. Available from www.bgs.org.uk. Accessed January 8, 2017.
The register includes a web questionnaire that contains information about care interventions associated to key components of palliative care in the last week of life. However, the questions about end-of-life discussions and the last expressed wish for place of death are not necessarily related to the last week of life. The responsible physician and/or registered nurse at the health care setting of the patient's death complete the questionnaire after the patient's death. The registered information is based on medical records, together with health care professionals' recalled memories of the care period during the patient's last week of life.
Study Population and Data Collection
To be included, patients had to be 18 years or older with a diagnosed chronic kidney disease according to the diagnosis registry International Classification of Diseases, Tenth Revision, Sweden (ICD-10-SE) (N18.5 or N18.9) by the Swedish National Board of Health and Welfare, which is a national interpretation of the international diagnosis registry ICD-10, reported by physicians as the mortality reason and/or underlying cause of death. Furthermore, the death of the patient had to be registered as expected in the SRPC. The Swedish Causes of Death Certificate Register (the Swedish National Board of Health and Welfare), which covers all deaths in Sweden, was used to ascertain identification of deaths of patients with chronic kidney disease corresponding to the inclusion criteria during the data collection period. Data between January 1, 2011 and December 30, 2012 were used. In 2011, 53% (n = 47,670) and in 2012, 62% (n = 57,031) of all deaths in Sweden were registered in the SRPC.
Variables
Background variables for the study sample included sex, age, place of death, and number of days enrolled to the health care setting of death. Place and/or health care setting of death was categorized into general home care, specialized palliative home care, nursing home (i.e., short- and long-term care facilities), hospital ward, hospice/inpatient palliative care, and other. Number of days enrolled at the care setting of death was categorized into 0–3, 4–14, 15–30, 31–365, and more than 365 days.
The study variables included the presence and relief of six predefined symptoms (pain, nausea, anxiety, respiratory secretion, shortness of breath, and confusion), the use of validated scales for symptom assessment (e.g., numeric rating scales), individual injection prescriptions as needed (pro re nata) for symptom relief (pain, nausea, anxiety, and respiratory secretion), consultation from other health care specialists (e.g., palliative care or pain management) for symptom relief, end-of-life discussions with patients and/or family members, whether the patient's preference of place of death was met, bereavement support offered to family members, and having someone present at the moment of death. Prevalence and relief of each symptom were reported in the categories: not present, totally relieved, partly relieved, not relieved at all, or do not know. The other questions were reported in categories: yes, no, or unknown.
Data Analysis
Descriptive statistics were used to present background characteristics and study variables. Data were analyzed with STATA 14.1 (StataCorp LP, College Station, TX).
Results
Sample
A total of 472 patients with chronic kidney disease were included, of whom 188 (40%) were women and 284 (60%) were men. The mean age was 82.2 years (SD 8.7). Most patients (52%) had been enrolled for less than one month to the care setting of death and 13% for a very short period (zero to three days) (Table 1). Most (89%) of all the patients died in institutional care settings, of which 45% died in nursing homes (long-term 33% and short-term 12%). Thirty-five percent died in a hospital ward, and 8% died in a hospice/inpatient palliative care unit (Table 1).
According to registry reports by health care professionals, all six predefined symptoms in the questionnaire were prevalent during the last week of life. Pain was most frequently reported (69%), followed by respiratory secretion (46%), anxiety (41%), confusion (30%), shortness of breath (22%), and nausea (17%). Uncertainty about symptom prevalence was most frequent for anxiety (16%) (Table 2).
Table 2Symptom Prevalence During the Last Week of Life (n = 472)
Reported symptom relief was highest for pain and anxiety. Despite this, 32% of patients with pain and 44% of patients with anxiety were only partly relieved or not relieved at all. Relief was lower for the other symptoms, with most patients being only partly relieved or not relieved at all: confusion 84%, nausea 62%, respiratory secretion 56%, and shortness of breath 55% (Fig. 1).
Fig. 1Relative frequencies (%) of symptom relief among patients with reported symptom prevalence.
The reported use of validated scales for symptom assessment was low. For assessment of pain, validated scales (e.g., numeric rating scales) were used on one or more occasion in 14% of patients. For the other symptoms, the use was 9%. Individual prescriptions for injections as needed (pro re nata) for symptom relief (reported for pain, nausea, anxiety, and/or respiratory secretion) were common and varied between 80% (respiratory secretion) and 90% (pain). During the patient's last week of life, external expertise in symptom relief was consulted for one-fifth (22%) of the patients (Table 3).
Table 3Symptom Management During the Last Week of Life (n = 472)
Study Variable
All Patients
Pain assessment with validated scales, n (%)
Yes
67 (14.2)
No
364 (77.1)
Unknown
41 (8.7)
Symptom assessment with validated scales other than pain, n (%)
End-of-Life Communication and Having Someone Present at Death
End-of-life discussions with a physician were reported in 41% of patients and 71% of family members. Whether end-of-life discussions were performed or not was reported as unknown in 22% of patients and 13% of families. The patient's preference for place of death was reported as met in 41% and unmet in 3%. However, 56% was reported as unknown (Table 4).
Table 4Aspects of Communication and Presence at Death (n = 472)
Study Variable
All Patients
End-of-life discussion with patients, n (%)
Yes
193 (40.9)
No
177 (37.5)
Unknown
102 (21.6)
End-of-life discussion with family members, n (%)
Yes
335 (71.0)
No
69 (14.6)
Unknown
59 (12.5)
No family members reported
9 (1.9)
Someone present at the moment of death, n (%)
No
88 (18.6)
Health care professionals
136 (28.8)
Family members
164 (34.8)
Health care professionals and family members
74 (15.7)
Unknown
10 (2.1)
Patient's preference for place of death was met, n (%)
About one-fifth (19%) of the patients died alone, whereas 51% had family members or family and health care professionals present. Health care professionals alone were present in 29% of patients' deaths (Table 4).
Bereavement support was reported to have been offered (one to two months after the death of the patient) to 38% of family members, whereas 31% had not been offered bereavement support and 30% was reported as unknown (Table 4).
Discussion
This registry study contributes to the growing literature on unmet palliative care needs of patients with ESKD and stresses that areas in which palliative care needs to be improved for these patients include symptom assessment and management, communication about end-of-life care including preferred site of death, and bereavement support.
The high symptom prevalence reported by clinicians is in line with earlier findings of a high symptom burden in patients with advanced kidney disease.
However, the SRPC comprises only six predefined symptoms, which means that other symptoms that are common for patients with advanced kidney disease may have been present. For instance, a study of patients with ESKD managed without dialysis showed that lack of energy, pruritus, drowsiness/somnolence, and concentration difficulties were the most prevalent and distressing symptoms besides the SRPC included shortness of breath and pain. Notably, the average number of self-reported symptoms during the last month of life was as many as 20,
found that in the last month of life, pain was reported by 73%. Also in the present study, the prevalence of pain was high. Moreover pain was reported as the most prevalent symptom, and about one-third of patients with pain were only partly relieved. Known reasons for pain are mineral bone disease secondary to chronic kidney disease and peripheral neuropathy but also comorbid conditions as, for example, peripheral artery disease. Moreover, some kidney diseases and also the dialysis treatment in itself may cause pain. The management of pain is complex (e.g., because of the risk of drug accumulation and the effect of dialysis treatment on pharmacokinetics) and requires more attention both clinically and in research.
have also described severe pain as an explanation for patients' decisions to withdraw dialysis. It should also be noted that besides pain, nausea and shortness of breath were also more prevalent in the study by Murtagh et al.
One explanation for this could be that symptoms were self-reported by patients in their study. There were also differences in the data collection periods (last week vs. last month) and treatment of kidney disease (in the present study, ESKD treatment was not specified).
Most importantly, our results show that validated symptom assessments seldom were used. Similarly, previous studies show that symptoms are underestimated and undertreated in renal care.
This addresses a need for better symptom management. The use of validated assessment scales in clinical practice may contribute to an increased communication between patients and health care professionals.
Thus, regular assessment of symptom prevalence, intensity, and relief with validated scales combined with person-centered dialogues should be an important part of both early and late renal palliative care, as this could contribute to enhanced well-being at the end of life. For example, Edmonton Symptom Assessment System Revised: Renal (ESAS-r: renal)
Although death was reported as expected, end-of-life discussions with physicians were conducted with only about two-fifths of patients, whereas end-of-life discussions with family members were more common. More frequently reported discussions with family members may indicate that they were conducted in a late phase when the patient was too ill to participate in the discussion and decision making. In accordance with the present findings, lack of communication about the future and the end of life has also been described in interviews with patients or relatives.
Reasons for discussions with patients not taking place could be lack of shared decision-making processes together with prognostic uncertainty. Therefore, the approach of end of life may not be acknowledged, and patients are then less likely to receive the end-of-life care they wish for. However, there are tools such as the surprise question that aim to identify patients who may be nearing end of life.
found that the reasons for the absence of these discussions may partly also be that renal physicians estimate that patients do not want or need this information. It has however been found that most patients want discussions about their prognosis
A systematic review of prognostic/end-of-life communication with adults in the advanced stages of a life-limiting illness: patient/caregiver preferences for the content, style, and timing of information.
Yet, in a recent study of end-of-life care of patients with kidney disease dying in hospital wards, it was found that none of 100 patients had an advance care plan.
In the present study, the patients' preference of place of death was reported as met in about two-fifths of patients. These results may partly be because of a lack of end-of-life discussions and patient involvement earlier on during the disease trajectory, when the patient is still able to participate in a conversation. Lack of end-of-life discussions, that is, less possibility for both the patient and the family members to prepare, could also be one of the reasons why one-fifth of the patients died alone and that almost 50% died without family members present. Family members' presence may also be related to that even if death is reported as expected it may have been difficult to foresee the last days. Lack of end-of-life discussions may also be one reason why few patients died in specialist palliative care. Delayed or insufficient discussions may also partly explain why only about one-tenth of patients died at home, although dying at home is a common preference.
A recent comprehensive study also showed that people with chronic kidney disease are less likely to die at home, compared with people without chronic kidney disease.
The results of the present study show a need for integration of palliative care with proactive actions in renal care earlier on in patients' illness trajectories (cf., e.g., Sawatzky et al.
). This corresponds with health care professionals' perceptions of inadequate care of patients in dialysis care, in areas such as bereavement support, spiritual support, and end-of-life discussions.
Moreover in a European survey, most nephrologists reported that palliative care had not been part of their core curriculum or recent medical education.
Hence, both primary palliative care training of nephrologists and renal nurses and collaboration with palliative care specialists should be emphasized in future renal care. One example is a pilot study where Feely et al.
found that collaborations such as palliative care consultations in an outpatient hemodialysis clinic were well received by both patients and care professionals and that documentation of care goals improved greatly. Furthermore, the value of palliative care has been shown in different contexts, as a review of randomized studies found that palliative care interventions improve patients' quality of life and satisfaction with care at the end of life.
Supporting the family is an important component of palliative care. Still, in the present study, bereavement support (at the health care setting of the patient's death) was offered to less than half of the patients' families. Importantly, after a long period of dialysis treatment, close family members may also need confirmation and closure with dialysis staff who are well known to them, irrespective of conversations with staff at the setting of the patient's death.
Therefore, the nurse or physician at the dialysis clinic should contact the closest relative, independently of other bereavement support.
Methodological Considerations
The reason for the decision to include the ICD-10-SE codes N18.5 or N18.9 was to ascertain identification of the target patient group. The diagnosis code N18.5 according to the ICD-10-SE is ESKD regardless of whether the patient is on dialysis or not. The diagnosis code N18.9 when stated as the mortality reason or underlying cause of death should stand for ESKD because a kidney disease of less degree would hardly be considered as a cause of death according to the formulation in the Swedish cause-of-death registry. Together with the inclusion criteria of expected deaths, we believe that this is a strength as this population is likely to have been identified as a relevant target group for palliative care. Still, there is a possibility that there are patients with ESKD who were not included. However, this is a general problem with registry studies as the data quality depends on the validity of the registration. Another strength of the present study is that the end-of-life questionnaire had been validated repeatedly.
There are several limitations of this study. One is that not all care units report to the SRPC, which means that not all patients with ESKD are represented in the register. Reporting units actively decide to join the SRPC, which also means that they may be more attentive to end-of-life care, thus possibly resulting in selection bias. It has been shown that registration in SRPC is associated with improved quality of care.
Hence, the general palliative care situation for these patients may be poorer than we report in this article. Even if 53% and 62% of all deaths in Sweden were reported in the SRPC during 2011 and 2012, respectively, the share of deaths of patients with chronic kidney disease that are registered is unknown. Another limitation is that the SRPC only comprises six symptoms, which means that other important symptoms may be missed; hence, there may be a need for an extended symptom measure in SRPC for diagnoses as ESKD. Validated scales were rarely used for symptom assessments, which may influence the accuracy of data. Recall bias, because the registry is completed retrospectively, is another limitation. The amount of reported unknown responses may be related not only to recall problems but also to a lack of documentation in patient records. A further major concern is that data are clinician reported and not reported by the patient or family members. Hence, there is a need for further research into patient- and family reported data at the end of life. Still, retrospective data contribute to knowledge of end-of-life care,
and this study has made it possible to gain information on the last week of life in people with ESKD, thereby identifying areas for improvement. A disadvantage of registry studies is that available background characteristics are often limited. No information is available for important background or medical characteristics, such as ethnicity, socioeconomic status, causes of CKD, and others.
We were not able to differentiate data between those treated with dialysis or conservative care or if dialysis was withdrawn. The study sample had an advanced age and a high percentage of institutional deaths that influences generalizability. There is a need for further studies of palliative care not only in relation to age and setting but also in relation to other factors, such as gender and ethnicity.
Conclusion
Although dying an expected death, patients with ESKD had unmet palliative care needs. Reports from health care professionals about patients' last week of life showed insufficient symptom assessment and relief, and findings suggest a lack of end-of-life discussions and bereavement support. Altogether, the results contribute to the knowledge that there are remaining challenges in the provision of care for dying patients with ESKD and a need for integrated palliative care.
Disclosures and Acknowledgments
The authors thank the National Register Committee and the SRPC for providing the data, particularly Greger Fransson. A special thanks to Sofia McGarvey for language editing.
This study did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Ethical Approval: Ethical approval for the study was granted by the Regional Ethical Review Board in Stockholm, Sweden (no. 2013/1576-31/3).
Survival outcomes of supportive care versus dialysis therapies for elderly patients with end-stage kidney disease: a systematic review and meta-analysis.
British Geriatrics Society. Palliative and end of life care for older people. BGS compendium document 4.8 2004. Revised version. Available from www.bgs.org.uk. Accessed January 8, 2017.
A systematic review of prognostic/end-of-life communication with adults in the advanced stages of a life-limiting illness: patient/caregiver preferences for the content, style, and timing of information.
30494-3&id=gr1.jpg){kind=link}