Mind the gap!

The Third SSNAP Annual Report

Case Study: Choosing the best model of stroke services for patients

Submitted by Dr Angus Ramsey, Senior Research Associate, Department of Applied Health Research, University College London (UCL) angus.ramsay@ucl.ac.uk

Based on: Impact of centralising acute stroke services in English metropolitan areas on mortality and length of hospital stay: difference-in-differences analysis
Choosing the best model of stroke services for patients
In 2010, Greater Manchester and London centralised their acute stroke services into a small number of specialist “Hyperacute Stroke Units” (HASUs). In Greater Manchester, only patients arriving at hospital within four hours of stroke were eligible for treatment in a HASU; in London, all stroke patients were eligible. In London 93% of patients were treated in a HASU, compared to only 39% of Greater Manchester patients.
In addition, the centralisations were implemented differently: in London, there was:
• greater prioritisation of service standards,
• clinical networks took a more hands-on approach to facilitating change,
• the whole system was launched on the same day.
Patients’ length of stay reduced more in both London and Greater Manchester than in the rest of England. However, only London was associated with significantly greater reductions in stroke patient mortality. Therefore, the London centralisation was associated with better patient outcomes. Using national stroke audit data in a study funded by the National Institute for Health Research (NIHR), the researchers analysed the impact of these centralisations on the provision of evidence-based clinical interventions, including rapid access to brain imaging, admission to a stroke unit, and specialist assessments. It was found that, following centralisation, London patients were overall significantly more likely to receive interventions than in Greater Manchester and elsewhere in England. This suggests that centralised systems that admit all stroke patients to HASUs, as in London, are significantly more likely to provide evidence-based care. This may explain why such systems are associated with better patient outcomes.

Royal College of Physicians, 11 St Andrews Place, Regent's Park, London NW1 4LE
Registered charity no. 210508
Terms and conditions    Cookie policy