This site is intended for health professionals only

How can more people be helped to die at home?

At the Nuffield we find that although many interventions claim to make significant impacts on emergency hospital use, most do not deliver.   However we have recently published an evaluation of a service that looks like it might work. Our study of nearly 30,000 people who received the end-of-life home-based nursing service run by Marie Curie found evidence that, compared to controls, they were much more likely to die at home, had significantly lower use of hospital care, and incurred lower hospital care costs.

Before getting carried away, it’s worth bearing in mind a couple of caveats. First, this wasn’t a randomised trial. Although we carefully matched controls to those who received Marie Curie care on a wide range of demographic and clinical variables and prior hospital use, there may have been differences between the groups on factors such as availability of carer/family support and preferences for where they died that might have influenced their suitability for home-based end-of-life care.

Secondly, the reductions in hospital costs that we observed are likely to be offset – at least in part – by other costs, including the nursing service itself, as well as possibly increased use of other services due to people being cared for at home. Nevertheless we think the scales of these costs are probably relatively low compared to the cost of fewer hospital admissions

Our study did not try and say why the service worked, however two points however are really striking to me. First, this isn’t a highly complex scheme requiring massive service redesign. Although Marie Curie offers a range of service models, the very vast majority of those we studied received a nine hour day or overnight visit from nurses or healthcare assistants. . Further research is needed to try and understand the key factors (pain and symptom control, support for carers?) that make this type of service effective. However, it seemed to be sufficient to allow people to avoid emergency hospital care and for many of them to be able to die at home.

Second, the impact of Marie Curie Nursing Service care was greater for people who didn’t have a recent history of cancer. The bulk of the literature on palliative and end-of-life care has focused (unsurprisingly) on people with cancer. Our findings suggest that this sort of care might offer similar or even greater benefits to those with other conditions. This also highlights the need for more research into those conditions that might particularly benefit from this type of care.

Whatever the final costs amount to, there would a strong case for making such home based support at the end of life more widely available. Allowing those who so choose to die at home, and reducing the need for unplanned and often distressing hospital admissions at the end-of-life is clearly a good outcome from a patient and quality perspective.

The recent palliative care funding review suggested that there is significant unmet need, with around 90,000 people dying every year without access to palliative care that would have benefited them (out of around 470,000 deaths annually). Given QIPP targets around end-of-life care and constant pressure to shift resources away from expensive hospital care into the community, this type of service also has the potential to allow wider changes in the nature and scale of acute care provision.

Experience has taught us that interventions that can really change patterns of service delivery are hard to find. The findings of this study (together with the wider literature on home-based end-of-life care) suggest that this type of service can significantly improve the quality of care for people at the end of life.


Visit Pulse Reference for details on 140 symptoms, including easily searchable symptoms and categories, offering you a free platform to check symptoms and receive potential diagnoses during consultations.