Determinants of emergency department attendance among people with dementia approaching the end of life.

Student thesis: Doctoral ThesisDoctor of Philosophy

Abstract

Background: Emergency department (ED) attendance among people with dementia is common and increases towards the end of life. Attendance can be distressing for people with dementia and is associated with poorer health outcomes. Understanding the determinants of ED attendance among people with dementia approaching the end of life is essential to inform policies that aim to improve end-of-life dementia care and reduce secondary care pressures.

Aim: To conceptualise the determinants of ED attendance among people with dementia approaching the end of life. Objectives: 1) To explore the drivers of ED attendance among people with dementia across the illness trajectory; 2) To examine individual- and service-level factors associated with ED attendance among people with dementia in their last year of life; and 3) To develop a conceptual model of ED attendance among people with dementia towards the end of life.

Methods: Informed by pragmatism, a convergent-parallel mixed methods design was used, mapped against three phases of the PRECEDE component of the PRECEDE-PROCEED health programme planning model. The first phase (social assessment) comprised a remote qualitative interview study exploring drivers of ED attendance among people with dementia across the illness trajectory, analysed using reflexive thematic analysis. The second phase (epidemiological assessment), conducted in parallel, comprised a population-based retrospective cohort study of individual- and service-level factors associated with ED attendance among people with dementia in the last year of life, analysed using multilevel modelling. The third phase (educational and ecological assessment) comprised a mixed methods integration to develop a conceptual model of ED attendance among people with dementia approaching the end of life. This was based on successive integration of claims from phases one and two, according to predisposing, reinforcing, and enabling factors. Ongoing engagement with public representatives facilitated the development of each phase. Policy implications were threaded throughout.

Results: Phase 1 (social assessment): From two dyad and 33 individual interviews (people with dementia=10; current caregivers=11; bereaved caregivers=16), three themes were developed: 1) Navigating a ‘push system’, 2) ED as the ‘last resort’, and 3) Taking dementia ‘seriously’. Themes describe a discrepancy between individual priorities and the configuration of health and social care, wherein ED attendance is often the path of least resistance for people with dementia who experience barriers to accessing timely community care and support. The discrepancy may reflect wider societal stigma and a lack of systemic prioritisation of dementia.

Phase 2 (epidemiological assessment): Of 74,486 decedents, 83% had at least one ED attendance in the last year of life. Factors associated with more ED attendances included living in urban areas (IRR 1.06, 95% CI 1.04-1.08), being of South Asian ethnicity (IRR 1.07, 95% CI 1.02-1.13), and having chronic respiratory disease as the underlying cause of death (IRR 1.17, 95% CI 1.14-1.20). Factors associated with fewer ED attendances included having higher socioeconomic position (IRR 0.92, 95% CI 0.90-0.94) and being a resident in a local authority with more nursing home beds (IRR 0.85, 95% CI 0.78-0.93), but not residential home beds.

Phase 3 (educational and ecological assessment): Based on the meta-inference, ED attendance among people with dementia approaching the end of life was conceptualised as a product of multiple intersecting predisposing characteristics (e.g., ethnicity, comorbidity, socioeconomic position) that interact with local reinforcing factors (e.g., residence and geographical location).
Systemic priorities and societal preconceptions influence how these factors interact and determine if a person with dementia is empowered to access community clinical care, or reliant on attending the ED towards the end of life.

Conclusions: Based on the findings of this thesis, it is posited that ED attendance is influenced by barriers to accessing timely and responsive community-based clinical continuity, which may be exacerbated among minoritised individuals and driven, in part, by a systemic failure to recognise dementia as a life-limiting illness and wider societal stigma. Findings are situated in previous literature and the current policy context, advocating an intersectional approach to policy development and practice, greater systemic prioritisation of dementia, improved post-diagnostic care that includes the end of life, and increased public education. The thesis also provides baseline observations, prioritised targets for change, and key components of an interventional programme to inform future research into safely reducing ED attendance among people with dementia.
Date of Award1 Sept 2023
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorKatherine Sleeman (Supervisor) & Catherine Evans (Supervisor)

Cite this

'