In our original proposal, we identified a gap in the research knowledge relating to older adults’ experiences of transitions at the end of life. Despite anecdotal evidence of distressing, unnecessary and costly hospital admissions near to death, there has been little published academic work in this area. We set out to gain a deep understanding of the patients’ perceptions through interviews, and planned to follow that with a more structured approach to data collection with bereaved carers. After a small number of interviews with carers, it became clear that this would not do justice to the richness of the data the carers wished to share, and we sought permission to conduct in-depth interviews with the relatively large group of bereaved carers.
Our original objectives were to:
explore the effect of transitions towards the end of life on health status, quality of life, symptom control and satisfaction with care
understand the factors that influence decisions about transitions in the nature and location of care
elicit patient and provider views on the appropriateness of different transition patterns and the factors that constrain or shape decisions
map (describe) transitions in and out of hospital at the end of life for older people in England
identify individual- and service-level factors associated with frequency of transitions.
The structured data collected from carers have provided some common information on satisfaction with different aspects of care in a range of settings, but primarily home and hospital. The mix of patient diagnoses has meant that there was no shared concept of symptom control. Questions on pain control were asked of all carers but were relevant to only a proportion. Carers’ feelings about how well the condition was managed did emerge from the interviews, but not in a format that leant itself to ready comparison across the whole sample. All three sets of interviews have contributed to the data to inform our understanding of perceived quality of life, satisfaction with care and the influences on decisions and consequences of transitions for patients and families. The interview data have also identified some individual- and service-level factors that are important to note. However, early in the data collection with professionals, it became apparent the notion of an appropriate move was too simplistic and did not reflect the multiple, interacting influences on decisions that were apparent to professionals and families. In answer to this question, we have presented an overview of the factors that emerged as most influential, and most often raised by participants.
Reasons for diversions from original protocol
Our work differs from the original protocol in the following ways.
1. Addition of patient interviews
This was suggested by reviewers of the original application.
2. Addition of qualitative interviews with bereaved carers to questionnaire administration
We sought permission from the funder to do this, after reviewing data from the first few interviews. The interviewees were sharing rich and useful qualitative data with the researchers that were not being adequately captured in the quantitative responses. This added to the workload of the project, as the time taken to analyse 118 lengthy in-depth interviews was considerable. No additional resources were required.
3. Design of linked data analysis
Data analysis was limited to patients with heart failure and lung cancer, in line with the patient interviews. Constraints of data availability meant that we were able to analyse data relating to people who were admitted to hospital and died, rather than data on people who died from specific causes.
