How is end-of-life care provided across Manchester?

Two men chatting in hospital.

Up to 74% of people say they want to die in their usual place of residence. This means they want to die in their own home, or in a residential or nursing home. However, 58% of people actually die in acute hospitals. On average, people have over three admissions to hospital in their last year of life and spend almost 30 days in hospital beds. Unwanted hospital admissions at end-of-life cost the NHS around £180million every year.

Previous research suggests that whether people die in their usual place of residence can be affected by their level of poverty, their race, or their religion but little evidence exists to explain these issues. We, at the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care (NIHR CLAHRC) Greater Manchester, have designed a research project to look at the different reasons that explain where people die. The results of the project will give us the chance to make suggestions to improve overall standards of care and to reduce costs by avoiding unwanted hospital admissions before people die.

How have we been doing the research?

We got in touch with the Greater Manchester Public Health Network to access information from the Primary Care Mortality Database, which holds information including place and cause of death as well as demographic and geographic individual details. We used anonymous quarterly data, between January 2011 and March 2015, from all GP practices across the three clinical commissioning groups (CCGs) in Manchester: North, Central and South.

The information we got included; the total number of patient deaths; the number of deaths that happened in their own home, a care home, a hospital, a hospice or other place; and the cause of death. We also used information about poverty levels, religion and race from the Office for National Statistics census 2011.

What have we found out?

We used the information to find links between demographic, religious and race factors and the numbers of deaths in the usual place of residence using anonymised data from each GP practice in Manchester. These included the average age at death, the numbers of deaths related to each cause of death, and the average poverty level and ethnic and religious composition. This meant we were able to predict expected results for each practice based on the demographic information for their local population – and then measure how the observed rates compared to those predicted.

Between January 2011 and March 2015, on average, the gap between the observed and predicted performances of GP practices differed widely. Only a limited number of GP practices had outcomes close to what was predicted. Even after accounting for disease profiles and socio-economic characteristics, a big variation exists across and within the three Manchester CCG areas to which the GP practices belong. In particular, there is one CCG where GP practices’ observed outcomes are similar and one CCG where the variation is much higher.

An important part of this diversity is likely to lie in the actions taken in primary care, both at the GP practice and at the CCG level.

Who has been doing the research?

This research is being led by the NIHR CLAHRC Greater Manchester, an organisation made up of researchers based at the University of Manchester and NHS staff at Salford Royal NHS Foundation Trust. The Principal Investigator for the project is Damian Hodgson, Professor of Organisational Analysis at the Alliance Manchester Business School.

What happens next?

These initial results could be extended to account for other aspects of GP practices; their organisational characteristics, resource availability and the quality of services provided. The effect of implementing specific end-of-life care policies (such as the individual plan for end-of-life care) in GP practices can also be analysed. GP practices with higher discrepancy between observed and predicted outcomes could be used as case studies for further research which would explore what lies behind the variations in care outcomes that people experience at the end-of-life.

How can I find out more

To find out more about this research, please contact Michael Spence by email via or by telephone on 0161 206 8551. 

Open this page in PDF format.