Introduction

Even though medical improvements have reduced the mortality rates for patients afflicted by stroke, mortality during the first few days at hospital is significant. Today, there is an increasing recognition that the principles of palliative and supportive care are important components of meeting the needs of patients severely afflicted by stroke even in acute settings. However, there is limited knowledge about which factors have an impact on the end-of-life care (EoLC) for these patients or about how these last days of life are experienced from the family members’ perspective.

Aim

The overall aim of this thesis was to describe the EoLC of patients severely afflicted by stroke and to identify factors impacting upon EoLC for the patients and their family members in Sweden out of various contexts and methods.

Design and Methods

This thesis is based upon four papers employing qualitative, quantitative and mixed-method designs.

Paper I is a qualitative study based on focus-group interviews with 41 health-care professionals (HCPs) in different professions related to stroke care at three stroke units. The aim was to study ethical dilemmas, different approaches and what consequences they had among health HCPs; the data was analysed using content analysis. The result inspired the design and conduct of the following studies.

Paper II is a quantitative comparative study based on a retrospectively registered questionnaire from the Swedish Registry of Palliative Care (SRPC). Patients dying of stroke (n =1626) were compared with patients dying from cancer (n=1626), according to symptoms, symptom management and communication with the patient and family members during the last week of life. Data was statistically calculated using OR.

Paper III is a mixed-method study employing a sequential explanatory design. In the first, quantitative, part, 995 stroke patients who died in hospital were compared with 631 stroke patients who died at nursing homes, according to symptoms, symptom management and communication with the patient and family members during the last week of life. The quantitative data was statistically calculated using OR and the qualitative data was analysed using content analysis. Nine significant differences drawn from the quantitative results regarding care were chosen to be discussed by twelve nurses working in stroke units.

Paper IV is a qualitative interview study with the aim to study the family member descriptions of the trajectory from admission to the hospital until their loved one died. A semi-structured interview guide was used, and data was analysed using thematic analyses.

Results

Factors that had an impact on EoLC were consequences related to the difficulties around decision-making about withholding or withdrawing life-sustaining treatment. Non-decisions or not holding to the decision generated communication barriers causing obstacles in inter-professional collaboration and ethical dilemmas within the team (Study I). The absence of a mutual approach to care resulted in underprovided palliation, undignified medical treatment and ambiguity in care, generating feelings of distrust among the family members. The results described in Study I of underprovided palliation were further investigated in Study II. The differences in knowledge about whether symptoms were present or not in patients afflicted by stroke compared to patients with cancer were significant. For example, the HCPs in the stroke group did not know if pain was present nine times more frequently than in the cancer group. These differences in knowledge about whether symptoms were present of not were also identified in Study III. Here, HCPs at the nursing home more often had knowledge about whether a symptom was present in patients dying of stroke or not, compared to HCPs at hospitals. This study also identifies differences in the presence of the symptoms being compared and whether the patient’s suffering was fully relieved. The nurses working at stroke units explained that these differences were a consequence of the stroke unit’s aim (saving lives), no previous relationship to the patient and ambiguity in the evaluation of symptoms. The patient’s altered levels of consciousness increased the difficulties in evaluation. In addition, the hospital setting’s aim increased the risk of prolonged treatment, for example nutrition supplied during the last day of life, and underprovided palliation of for example, pain compared to nursing homes. In Study IV, family members were seeking trust through mutual collaboration and creating relationships with the HCPs. If the family did not feel that their search for trust was taken care of, it generated feelings of distrust allied with anxiety and memories of failing to do the best for their loved one at in the end of life. During the trajectory at the hospital, family members were “seeking trust in chaos”, “seeking clarity when deciding about living or dying” and finally they were “seeking trust in care as a final act of love”.

Conclusions:

The results of this thesis suggest that the absence of a coherent approach contributes to developing ethical dilemmas within the HCPs. The ambiguity in care had an impact on the quality on EoLC, with an increased risk of unnecessary suffering and questionable symptom management. The prevailing culture at acute-care hospitals affected the HCPs’ attitudes towards EoLC, with attention being predominantly on life-sustaining treatments. There is an increasing recognition of the need for improvement in the PC approach at acute-care hospitals in order to create equal quality of care during end of life, irrespective of the place of death for patients dying of stroke. Furthermore, family members need to feel trust, which is achieved through relationships and collaboration with HCPs. In the striving to accomplish a wholly compensatory care of quality during the trajectory, this thesis can be a source of knowledge and guidance for nurses and for teams at the stroke unit.