It is a momentous week for the Improvement Analytics Unit. The first analysis conducted by the unit – our partnership with NHS England to provide rapid feedback on the impact of new care models and other national transformation programmes – was encouraging, both in terms of the results demonstrated by the care home initiative we analysed and also in terms of showing the feasibility and benefits of analysing data sets that span organisational settings.

Our analysis examined the impact of providing enhanced support for people living in care homes in Rushcliffe, Nottinghamshire. Residents of the participating care homes experienced 29% fewer A&E attendances and 23% fewer emergency admissions to hospital, when assessed against a matched comparison group.

The enhanced support package was developed by Principia, a multi-speciality community provider. It had several elements, one of which was to assign one general practice for each care home, and to encourage residents to consider moving to that practice. Not an easy change to make, since often people move to care homes at times of vulnerability. Residents were supported through the decision by Age UK Nottingham and Nottinghamshire.

The enhanced support included some other important elements. Named GPs visited residents who were registered at the aligned practice, reviewing their medicines on a regular basis and standardising the processes of care for people living with long-term health conditions. Community nursing teams provided practical advice and training to their nursing home colleagues, who had sometimes been isolated. Local care home managers were brought together in a network, and worked with the clinical commissioning group (CCG) to establish shared ownership of the changes being made.

To evaluate how successful the enhanced support was at reducing the use of secondary care, the Improvement Analytics Unit obtained linked care home and hospital data for Rushcliffe and similar areas of England. Then, the team selected matched comparison residents, who had similar age, gender, health conditions and prior hospital use to the Principia residents, and lived in similar care homes.

Compared with the matched group, the Principia residents experienced similar rates of emergency hospital admission up until the point at which they joined the care home. After moving to the care home, their admission rates were 23% lower than expected. Assuming that the two groups were comparable, the most likely explanation of the Improvement Analytics Unit's findings is that they reflect higher quality care for residents of the Principia care homes.

Some questions remain. In a study like this, we couldn’t rule out the possibility that the Principia and matched comparison residents differed in unobserved ways, potentially explaining the difference in the use of emergency care. And we didn’t study how the enhanced support operated in practice, or the mechanisms through which the various elements might have interacted with contextual factors to produce the lower rates of A&E use. Further evaluation is needed but, in the meantime, these findings are encouraging. 

Reductions in emergency hospital admissions are potentially good for the individuals concerned, and might help to manage demand for secondary care. So, these findings should be considered when designing approaches to improve care for people living in care homes in other parts of the country. 

While the evaluation is significant in its own right, the picture becomes more exciting when seen in the broader context of what the Improvement Analytics Unit will achieve over the coming years.

This analysis was produced in under nine months, which is not bad considering the great strides made beneath the surface to link care home and hospital data. But the Improvement Analytics Unit is planning to streamline further, potentially turning around an evaluation in just a few months. By 2019, the unit will be conducting up to 20 evaluations each year, feeding back analysis on a regular basis to the participating teams.

Access to timely and reliable data is important: a message that emerges clearly from the quality improvement work that the Health Foundation has supported over the last decade. Yet, when it comes to health care data, reality does not often match the hype. We might live in a ‘big data’ era according to popular imagination, but our recent review found worryingly low levels of analytical capability in many parts of the NHS. 

Initiatives like the Improvement Analytics Unit show that progress can be made in data linkage and care delivery, even in complex areas such as integrated care.

New care model vanguards can register their interest in working with the Improvement Analytics Unit by contacting

In the meantime, the Health Foundation is exploring other ways to support applied analytics within the NHS, with a wider focus than the new care models. More soon.

Further reading