Service redesign should be tested as rigorously as new treatments, NHS chief says
BMJ 2014; 348 doi: https://doi.org/10.1136/bmj.g3744 (Published 05 June 2014) Cite this as: BMJ 2014;348:g3744
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Simon Stevens' address to the NHS Confederation Changes stated the need for rigorous testing of new approaches to service delivery. In the same speech (1), he also stressed the need for different solutions for diverse communities, holding up London's reform of stroke services as proof of the importance of the centralisation of services. The evidence on which this is based comes from a before-and-after study showing a '12% reduction in death rates.'
The BMJ has led the way in requiring authors to express benefit as absolute, and not relative, risk reduction. The study (2) compared outcomes in 307 patients treated for stroke before the introduction of 8 hyper-acute stroke units with those of 3156 patients treated after their introduction, finding that adjusted 90 day survival rates increased from 87.2% to 88.7%, an absolute increase of 1.5%. Extrapolated to the total number of 6438 stroke patients admitted across London, this corresponds to 96 people per year. Moreover the total costs were predicted to fall, largely as a result of reduced length of stay.
The introduction of evidence-based service reconfiguration is very welcome. But there is a risk of extrapolating from that evidence to justify the sort of reconfiguration of services being implemented across London, something being done by the Royal Colleges and the Chair of the Academy of Medical Royal Colleges (3,4). The claim is that lives will be saved by concentrating A&E care in a smaller number of specialist centres, while downgrading departments in other hospitals to become Urgent Care Centres. But a proper risk-benefit analysis would need to compare the estimated benefits with the possible adverse consequences of reconfiguration. On one side, these comprise 96 out of 6438 stroke patients, and approximately 120 of 7000 patients in London with acute myocardial infarction - assuming an estimated absolute risk reduction of 1.7% in rates of 6 month death or re-infarction from immediate PCI (5). But these two conditions make up only 0.5% of the 2.6 million A&E attendances across London (6), and there does not appear to have been any attempt to quantify the impact on the 99.5%. Indeed, even people who have had a stroke may wish to weigh the relative benefits of a gain of ~8 months of quality life expectancy with being at a greater distance from their family during their admission.
If the Royal Colleges and their Academy want to enter the debate, they should do so using rigorous science. This implies the need to explore more fully the benefits and disutilities of service reconfiguration, rather than simply claiming that change is justified on the basis of studies of single conditions with significant P values.
1. Triggle N. Dropping the 'N' in NHS http://www.bbc.com/news/health-27719958?utm_content=bufferd4a34&utm_medi...
2. Hunter RM, Davie C, Rudd A et al. Impact on clinical and cost cutcomes of a centralized approach to acute stroke care in London: a comparative effectiveness before and after model. PLoS ONE 8(8): e70420. doi:10.1371/journal.pone.0070420
3. Douglas N, Gilmore I, Field S et al. NHS change must be driven by clinical evidence. Guardian, 28 April 2010. http://www.theguardian.com/society/2010/apr/29/nhs-change-clinical-evidence
4. Campbell D. The doctors' leader who says it's time to make tough decisions on the NHS. Guardian, 24 July 2012. http://www.theguardian.com/society/2012/jul/24/terence-stephenson-doctor...
5. Cantor WJ, Fitchett D, Borgundvaag B, et al. Routine early angioplasty after fibrinolysis for acute myocardial infarction. NEJM (2009) 360: 2705-18.
6. HSCIC. Attendance rates at major A&E departments highest in London. 28 Jan 2014. http://www.hscic.gov.uk/article/3875/Attendance-rates-at-major-AE-depart...
Competing interests: No competing interests
Re: Service redesign should be tested as rigorously as new treatments, NHS chief says
John Yudkin is clearly right [1] that even if the “hub-and-spoke” reorganisation of acute stroke care in London has been a stunning success, it tells us nothing about whether services for other conditions should be similarly centralised. Unfortunately the published evidence about the London stroke service is anything but robust.
This example of “rigorous science” consisted of a comparison of about 1 year’s worth of acute stroke admissions to London hospitals, before and after the changes [2]. The study combined data from the community-based South London Stroke Register (SLSR), an internal audit done in two hospitals and data from London hospitals taking part in the national SINAP audit. Clearly patient groups admitted to the same hospital before and after a change in admissions pathway are not comparable, so the only valid comparison is between the 205 first-ever stroke cases in the SLSR in the “before”, and 105 in the “after” period. With such numbers the 95% confidence intervals on the estimated absolute difference in fatality rates would be around +8.5% (incorporating differences in survival time to increase statistical efficiency makes no sense clinically).
We have shown [3] that diagnostic uncertainties and clerical errors can cause disagreements in the classification of “acute stroke” cases by different registration systems, running simultaneously within the same hospital. This can produce differences in mortality estimates well above the 1.5% on which the claims of success in London are based [3]. Indeed it is strange to promote a service reorganisation focused on the need to increase thrombolysis rates and thus “save lives”, when intravenous thrombolysis has never been shown to reduce deaths from acute stroke [4].
The new system of acute stroke care in London may well be a success, but it bodes ill if this is a model of the rigorous evidence needed for other major service reconfigurations. Commissioners trying to negotiate the minefield of “competition” rules will have little to inform their choices beyond political rhetoric and marketing hype.
1. Yudkin JS. Service redesign should be tested as rigorously as new treatments, NHS chief says. BMJ 2014;348:g3744
2. Hunter RM, Davie C, Rudd A et al. Impact on clinical and cost outcomes of a centralized approach to acute stroke care in London: a comparative effectiveness before and after model. PLoS ONE 8(8): e70420. doi:10.1371/journal.pone.0070420
3. Barer D, Cassidy T. Effects of diagnostic uncertainty and misclassification on hospital performance indicators for acute stroke care. Clinical Medicine: accepted for publication 2014.
4. Wardlaw JM, Murray V, Berge E, et al. Recombinant tissue plasminogen activator for acute ischaemic stroke: an updated systematic review and meta-analysis. Lancet. 2012; 379(9834): 2364–2372
Competing interests: No competing interests