Geographical understanding of variation in place of death: the role of care services and end of life care improvement (GUIDE_Care Services)
Award Number
14/19/22Programme
Health and Social Care Delivery ResearchStatus / Stage
CompletedDates
2 July 2015 -1 April 2018
Duration (calculated)
02 years 08 monthsFunder(s)
NIHRFunding Amount
£300,854.91Funder/Grant study page
NIHRContracted Centre
King's College LondonPrincipal Investigator
Professor Wei GaoPI ORCID
0000-0001-8298-3415WHO Catergories
Models across the continuum of careTools and methodologies for interventions
Disease Type
Dementia (Unspecified)CPEC Review Info
| Reference ID | 170 |
|---|---|
| Researcher | Reside Team |
| Published | 12/06/2023 |
Data
| Award Number | 14/19/22 |
|---|---|
| Status / Stage | Completed |
| Start Date | 20150702 |
| End Date | 20180401 |
| Duration (calculated) | 02 years 08 months |
| Funder/Grant study page | NIHR |
| Contracted Centre | King's College London |
| Funding Amount | £300,854.91 |
Abstract
The UK end of life care strategy has set a target to enable more people to die in their place of choice. Most people (85-90%) prefer to die at home or in a hospice. However, hospital remains the most common place of death (PoD), with great local variation in home and hospice deaths. In our NIHR project GUIDE_Care we have identified trends and patient factors associated with this variation. This explained a quarter of the variation and even less for cancer, dementia and some other Long Term Conditions (LTC’s) (~10%), prompting a further search for stronger determinants of where people die. This follow on study builds on the GUIDE_Care findings to investigate the potential role of service factors, provide national and local data to improve end of life care and better meet patient and family preferences. To evaluate the role of service factors in place of death, to inform national and local end of life care improvement. 1) To evaluate the relative contribution of service factors in PoD; 2) To produce GIS maps to visualise patient and service factors adjusted for geographical variation in PoD; 3) To explore the inter-relationship of service and patient factors with PoD. Data and methods Design and setting: a population-based multi-level study in England. Data: we will use routine data sources at two levels: patient- and area-level. The former is the death registry data provided by ONS. The latter is primarily available in the public domain with no non-disclosure restrictions. The patient-level data includes records for all non-external causes of death, over a recent 5 year period (2008-2012), containing information on patient factors (eg. age, gender). We will investigate cancer (ICD-10: C00-C97) and other selected causes of death (eg. Dementia, LTC’s). Patient-level data is linked to area-level data through geographical identifiers. Analysis The outcome is PoD. We have identified three main groups of service factors: service provision, service utilisation and proximity to care facilities. In addition, we will calculate and adjust for population-based needs for palliative and end of life care. The relative contribution of service factors is evaluated at the area-level, in the context of patient factors and population based need. The PoD will be modelled as a proportion using beta regression, or as a count using Poisson regression. The R-square type measures are used to quantify the relative contribution of service factors. The residual variations are plotted using GIS maps to visualise the patient and service factor adjusted variation. The inter-relationship of service and patient factors with PoD is explored using Structural Equations Modelling (SEMs), or its alternatives. At patient-level, PoD is modelled as a multinomial variable and the clustering effect is accounted for under multilevel modelling or generalised estimating equation (GEE) framework. The results from area- and patient-level analysis are integrated and synthesised under the reported conceptual framework. Expected outputs These include scientific publications and conference presentations; reports for service users, commissioners, policy-makers and service providers; and, dissemination through traditional media briefings and social media (e.g. Twitter, YouTube, Podcasts).
Aims
To evaluate the role of service factors in place of death, to inform national and local end of life care improvement. 1) To evaluate the relative contribution of service factors in PoD; 2) To produce GIS maps to visualise patient and service factors adjusted for geographical variation in PoD; 3) To explore the inter-relationship of service and patient factors with PoD.