In 2008 Maxine Power was the Chief Investigator on Stroke 90:10, a randomised trial which sought to determine whether participating in a quality improvement collaborative could improve adherence to a bundle of care process known to reduce stroke mortality when performed reliably.
The programme ran from July 2008 to December 2010 and involved 26 hospitals in the North West of England. It was the first time that an improvement programme dedicated to stroke had been implemented on this scale.
In the UK, stroke affects up to 110,000 patients per year resulting in long term disability and unacceptably high mortality with up to 30% of patients dying within one month (correct at the time of study in 2008). Data from the 1990s indicated that UK patients had ‘poorer outcomes than many European countries’ .The National Sentinel Audit of Stroke carried out by all acute trusts collected data on nine key process indicators and ‘monitored the rate of progress in stroke care and services in England, Wales and Northern Ireland on a two year cycle from 1998 to 2010’. It had repeatedly shown a wide gap between the evidence base and practice, i.e., only 70% of patients received aspirin with 48 hours. Multiple challenges were identified across the region, such as a lack of regular data collection and an absence of a community for sharing improvement ideas and work.
The nine key process indicators were grouped into two bundles: one focussed on early hours care and one on rehabilitation following stroke. The aim of the programme was to ‘improve compliance to an average of 90% on each of the nine processes by the next National Audit (2010); this would be measured against the baseline average of 72% in 2008. In addition, Stroke 90:10 aimed to improve reliability, accelerate the rates of improvement across organisations through shared learning and build capability for quality improvement.
Through a Breakthrough Series (BTS) collaborative model (a short term ‘learning system that brings together a large number teams from hospitals or clinics to seek improvement in a focused topic area’) hospitals were randomly allocated to the intervention (n = 13) or control (n =13). The intervention group (phase 1) participated in the programme from January 2009 to October 2010 (22 months), with the control group (phase 2) participating from January 2010 to October 2010.
The ‘phase 1’ teams worked to increase compliance to two bundles (collection of care processes). This percentage of compliance would be compared against baseline data.
The ‘phase 2’ teams introduced refined interventions and implemented learning developed during ‘phase 1’.
Each hospital taking part dedicated a multi-disciplinary team led by a stroke physician to the collaborative and these teams were supported in their improvement work through learning sessions and an expert faculty.
From the study it was found that:
For Bundle 1, the ‘phase 1’ (intervention) sites improved from 20% to 65% compliance and the ‘phase 2’ (control) sites from 27% to 62% compliance during the same period. For Bundle 2, the ‘phase 1’ (intervention) sites improved from 29% to 71% compliance and the ‘phase 2’ (control) sites from 18% to 74% compliance.
The study also determined that the BTS collaborative model can facilitate improvements in stroke care, that those who are new to improvement can learn quickly when introduced to an established collaborative and that these findings have ‘significant implications for models of diffusion and spread’.
Power, M., Tyrell, P.J., Rudd, A.G., Tully, M.P., Dalton, D., Marshall, M., Chappell, I., Corgié, D., Goldmann, D., Webb, D., Dixon-Woods, M., Parry, G. (2014). Did a quality improvement collaborative make stroke care better? A cluster randomised trial. Implementation Science, 9(40)
Carter, P., Ozieranski, P., McNicol, S., Power, M., Dixon-Woods, M. (2014). How collaborative are quality improvement collaboratives: a qualitative study in stroke care. Implementation Science, 9(32).