NHS North West London

Transcription

1 NHS North West London Shaping a Healthier Future Pre-Consultation Business Case Volume 3 Chapters 11 to 15 Edition: July 2012 Approved by JCPCT on 25 June 2012 Page 1

2 Table of Contents Page 1 Executive Summary Volume Introduction to the NHS in NW London Volume Introduction to the Shaping a healthier future programme Volume Stakeholder Engagement Volume Clinical Case for Change Volume Financial Base Case Volume Our vision for how to improve significantly the health of people in NW London Volume Delivering better care out of hospital and in hospital Volume Benefits of implementing the vision Volume Quality assurance of the vision Volume Determining the number of major hospitals needed Volume Determining options for five major hospitals Volume Developing criteria to evaluate options for major hospital sites Volume Appraisal of options for locating major hospital sites Volume Confirmation of options for consultation Volume Proposals common to all consultation options Volume Preferred option for consultation Volume Other options for consultation Volume Implementation plans Volume Plans for consultation Volume Approval process Volume Next steps Volume 5 24 App A A1) Programme Governance and A2) Programme Governance: Terms of Reference for Programme bodies Volume 6 App B Further information on stakeholder engagement Volume 7 App C Consolidated finance pack from F&BP modelling Volume 8 App D D1 to D8) 8 CCG Strategy documents Volumes 9-16 App E Out of Hospital Standards Volume 17 App F Methodology for impact on workforce Volume 17 App G Additional information on assurance Volume 17 App H Further information on travel analysis Volume 17 App I Evidence on impact of quality of estates Volume 17 App J Further information on estates analysis Volume 17 App K Western Eye Hospital proposed relocation Volume 17 App L Estimated impact of reconfiguration options on activity Volume 18 Page 2

3 11. Determining the number of major hospitals needed This chapter describes the process for determining how many major hospitals there should be in NW London. A set of hurdle criteria, developed by clinicians, were used to establish the optimal number and, based on this, clinicians believe NW London needs five major hospitals. Having fewer would be too expensive and take too long to deliver, although it would be clinically safe if implemented. If there were more than five, then clinical quality would fall as clinicians would not see enough patients and there would not be enough staff for sustainable clinical rotas. This chapter sets out: The high-level process for determining options The start of the process to determine options The analysis to determine the number of major hospitals. The recommendations in this chapter were based on the proposals set out in Chapters 7 and 8, and led to the next stage of the process to determine where the major hospitals should be located, set out in Chapter High-level options evaluation process A process was designed that enabled the programme to move through a funnel from an initial possibility of millions of options down to a small number of options to undergo further analysis, before agreeing the options that would go to consultation. Figure 11.1: Overview of process for developing and evaluating options 11.2 Starting the process to determine options If local clinicians considered every possible combination of reconfiguration options, the exhaustive list would be too long to be meaningful due to the potential millions of combinations of all the service delivery models on all the existing sites and, theoretically, on any number of new sites. Page 3

4 Local clinicians considered the implications of the acute clinical standards and the service delivery models and identified a number of key principles that should underpin the development of the options. Figure 11.2: Key principles underpinning development of long list of options By using these principles, clinicians agreed that the options development process would be driven by the location of the major hospitals in NW London to ensure the appropriate delivery of urgent and complex secondary care across NW London. This allowed the generation of a theoretical long list of consultation options that described how one or more major hospitals could be located on any of the existing sites in NW London, or even consider putting major hospitals on new sites. The next stage was to filter these options to a manageable list of options that was realistic and understandable Determining the number of major hospitals needed Clinicians used a set of hurdle criteria to establish the right number of major hospitals in the options. Each option needed to: Deliver the acute clinical standards Be deliverable within a realistic timeframe without creating a large risk to the consistent high delivery of services Demonstrate high level affordability. In practice, this meant defining and applying an agreed set of criteria and eliminating options where these were not met. Four criteria relating to meeting the emergency and A&E clinical standards, high-level affordability and deliverability were used to determine the number of major hospitals needed. These criteria are shown in Figure Page 4

5 Figure 11.3: Clinical Board rationale for determining the number of major hospitals Step 1 A major hospital is required to deliver high quality care Local clinicians recommend that the major hospital service model described in Chapter 8 is required to ensure the provision of high quality care based on the acute clinical standards. Based on the issues set out in the Case for Change, clinicians concluded that their desired clinical standards (see Chapter 7) could not be met if all nine current NW London acute sites (Central Middlesex, Charing Cross, Chelsea & Westminster, Ealing, Hammersmith, Hillingdon, Northwick Park, St Mary s and West Middlesex) were to become major hospital sites. The Case for Change highlighted the issues currently facing the current configuration of nine major hospitals, which include: Lack of available manpower with sufficient skills/experience Staff being unable to build and maintain their skills/experience if work is spread too thinly across the sites (even if the staff were available) The costs of providing staff (even if they were available) on a 12-24/7 basis compared to the volumes (and therefore revenue) of activity. Clinicians do not believe that there is currently sufficient `reason for changing the specialist hospitals (Royal Brompton, Harefield, Royal Marsden, St Mark s and the Royal National Orthopaedic). The focus for clinicians was then to determine how many major hospitals should be located in NW London to provide the highest quality healthcare Step 2 Consider the nine existing major hospital sites only and not new locations Clinicians have recommended that only the nine existing acute sites should be considered for future location of major hospitals. New brown or green locations are not suitable due to the timescale required to find and develop any site. Choosing to build a brand new site for major hospitals would also have extremely high capital requirements and would not support a financially viable health system in NW London in the future. Page 5

6 Planning permission would be required for any new site and the Clinical Board acknowledged the following comments about the general market for new sites in NW London: The general market for redevelopment property in West London remains strong, notwithstanding the national and global economic difficulties at present. This market is often led by residential interest, although several high profile commercial schemes have recently gained publicity, such as the development of the Westfield Shopping Centre (Shepherds Bush/ White City), the consultation stages of the Earl s Court redevelopment and proposed sale of the iconic BBC TV Centre. Land values which can reach and exceed 100 Million per acre for prime Mayfair sites, begin to fall as one moves west, reaching c. 5 to 10 Million in the area of White City and perhaps 3 Million in Park Royal / Greenford (outside the initial search area). It seems that the greatest likelihood of being able to identify a site suitable for the purpose under consideration (size, accessibility and so on) is for some land in an outdated existing use (Govt/ Local Authority/ Statutory Undertaker/ major Utility) to become surplus to present needs. Truly open market land sales are rare and often particularly costly, due to likely competition with residential developers. Work investigating new sites for other programmes in NW London has established that due to the general sensitivity of social, political and transport matters, as well as high values and historical/ complicated ownerships in this area, many sites take years, if not decades to see physical development take place Step 3 There should be three to five major hospitals in NW London to support the population The Clinical Board considered available evidence about the factors which contribute to high quality clinical care: Evidence of the links between senior staff presence and quality of care Patient volumes required to ensure staff build and maintain skills Technology required to support high quality care Interdependencies between different acute services and their required clinical support. Local clinicians identified that there should be between three to five major hospitals in NW London to support the projected population of 2 million. Clinicians identified that having more than 5 major hospitals would be likely to lead to suboptimal care: Clinical evidence highlights that involvement of senior medical personnel improves outcomes for example patients do less well when cared for in the evening or at weekends when there are fewer senior staff available. There is a constraint on available consultants with sufficient skills to cover emergency surgery rotas and A&E rotas - concentrating these specialist doctors on five or less sites will ensure consultant delivered care at least 12 hours a day. This will enable NW London to meet its quality standards and respond to any future workforce requirements Even if there were more specialist doctors and their teams available, spreading them across more than five sites would mean that they would each see fewer patients and therefore be unable to build and maintain the skills and expertise they need to ensure delivery of high quality care. A full 24/7 emergency care site requires a sufficiently large population catchment to ensure that consultants and their specialist teams Page 6

7 maintain their skills by conducting a sufficient number of operations. (This particularly applies to emergency surgery, critical care and interventional radiology). Figure 11.4: Minimum required number of emergency surgeons and implied catchment of population for differing numbers of major hospital sites The Case for Change noted that National shortages of some clinical staff groups, such as paediatricians, midwives, radiologists and pathologists, due to the numbers of individuals currently entering training, are expected to continue in the future. Even if there were more suitably trained staff in place, they would quickly begin to lose their skills as they would not be seeing sufficient volumes of patients. Currently, the nine acute sites require at least 45 surgeons on each of their rotas to provide the necessary cover, however this is only to provide for a relatively low catchment of around 238,000 people (see Figure 11.4). Local clinicians agreed that having more than five major hospital sites would not enable the specialist expertise to be concentrated into enough centres and seeing sufficient volumes of patients to deliver the highest quality care. The Clinical Board recommended that potential options with only one or two major hospitals would be too difficult to deliver and would compromise access Supporting evidence on number of hospitals providing Emergency Care For emergency care services, early involvement of senior medical personnel in the assessment and subsequent management of many acutely ill patients improves outcomes, i.e. patients suffer fewer complications and are less likely to die when they are cared for by senior, more experienced staff. Recent analysis across London has shown that patients attending and admitted to hospital during the evening, night or at the weekend are more likely to die than people admitted at times when more senior staff are available. Studies suggest that there is an association between reduced numbers of senior staff at weekends, and the observed increases in mortality seen at these times. In addition, evidence suggests that where a service has the same provision in place, seven days per week, there is no observed difference in mortality rates in the week and at the weekend. Page 7

8 Royal Colleges have indicated that a population of 350, ,000 is required to have sufficient scale to run a high quality urgent surgery service where staff are seeing sufficient volumes to maintain skills 1. Alongside this evidence has shown that by establishing specialist centres and networks, patients will experience better clinical outcomes. An example of this in emergency surgery is around access to surgeons who are trained in laparoscopic surgery for emergency situations. Currently in NW London several sites have low levels of emergency surgeons and not all are able to conduct laparoscopic procedures, as shown in Figure By reducing the number of sites providing emergency surgery, there will be better cover at those specialist centres that do provide emergency surgery and those places will have teams of surgeons with a broader range of specialist skills to give patients the best expertise. Those teams of surgeons will also see greater number of patients helping them to maintain and develop their skills. Figure 11.5: Number of emergency surgeons and the percentage of which are laparoscopic trained for major hospital sites in NW London 2 Clinicians recommend that by changing to a configuration of 3-5 major hospitals in NW London, the right skills will be in place to meet the clinical standards and provide the highest quality of care Step 4 Only options that have five major hospital sites are viable in the medium term Clinicians have identified that only options that have five major hospitals are viable in the medium term. Whilst having less than five major hospitals would be likely to increase the efficiency with which the clinical standards could be delivered, moving to three or four sites would cause major disruption to existing services which could affect the consistent delivery of high quality services. Moving to fewer than five major hospital sites would require transferring a large number of existing services simultaneously across the region increasing the likelihood of: A long implementation timeframe (~7+ years) and period of change A large investment in capital to develop infrastructure on some sites during a period when access to capital investment is severely constrained. 1 Academy of Royal Colleges, Acute Healthcare Services Report, London Health Programmes, NHS London, Adult emergency services: Acute medicine and emergency general surgery Report, September 2011 Page 8

9 No sites currently have the capacity to deliver the volumes of activity in an option with less than five major hospitals, as shown in Figure Figure 11.6: Bed requirements per major hospital and current bed capacity 3 The current site with the greatest bed capacity is Northwick Park, with 576 beds. The smallest site is Central Middlesex with 227 beds. If options with three or four major hospitals were considered, thereby requiring between beds at each site, all sites that were included within that option (even if comprised of the largest hospitals in NW London) would need significant levels of expansion to meet demand. This scale of change would be likely to have a long delivery timeframe, due to the investment (and subsequent building / development work) required. By only considering options that have five major hospitals, there are several existing sites that are close to providing the capacity that is required. The Clinical Board recommended that five major hospital sites in NW London would provide the most sustainable model for delivering high quality care to the projected local population of 2 million. Building on this work and the clinical quality standards, local clinicians considered the implications for paediatrics and maternity services, cross-referencing this work with need to for change set out in sections and and other reviews, such as the pan-london pathway reviews. They recommended that: To meet clinical standards and best utilise workforce, maternity units in NW London should plan to cater for approximately 6000 births per annum on average There should be five paediatric inpatient units Given co-dependencies with paediatric services and neo-natal units, there should be five maternity units, to be part of the suggested five major hospital sites. The Clinical Board also recommended that there be an additional unit on another site if there were the specialist services to support it. These are the current clinical recommendations for maternity and paediatrics. In the future, commissioners may establish that there are not enough consultant staff for sustainable, safe 3 There may also be specialist and elective hospital sites so therefore the change in total bed base across the sector cannot be calculated using only these numbers; current bed capacity includes adult general and acute beds, adult day care and critical care i.e. excludes paediatric, maternity and other beds Page 9

10 clinical rotas and may wish to revisit the number of maternity and paediatric units needed in NW London to deliver high quality clinical care. Page 10

11 12. Determining options for five major hospitals This chapter describes the process for developing a smaller list of options for where the five major hospitals should be located. By applying further hurdle criteria, clinicians developed a set of eight options for further consideration. The eight options were based on eight of the current A&E sites as Central Middlesex is proposed to be designated as an elective hospital. In all cases, Hillingdon and Northwick Park are proposed to be designated as major hospitals with the remaining three major hospitals at Charing Cross or Chelsea and Westminster; Ealing or West Middlesex and Hammersmith or St Mary s. This chapter sets out: The analysis to determine the location of five major hospitals The proposed eight options to go for further evaluation. The analysis described in this chapter follows on from the analysis completed in Chapter Continuing the process to determine the location of five major hospitals Even by recommending five major hospitals for NW London, 126 potential options for reconfiguration remained. The next stage was to filter these options with five major hospitals to a manageable list of options that were realistic and understandable. To do this, clinicians used three more hurdle criteria relating to minimising changes to access and deliverability. Figure 12.1: Clinical Board rationale for determining the location of five major hospitals Calculating travel times A considerable amount of travel analysis was used during this part of the process to determine the location of five major hospitals. Travel time has been used in two ways to support the analysis of the potential reconfiguration options. The travel time analysis is important as it is used to: 1) Support predicting where activity may flow if one hospital no longer offers a service, for example, if hospital X is not offering A&E services, how much extra capacity will surrounding hospitals need to be able to cope with the additional volumes? 2) Reflect the potential change in actual travel time that may be experienced by the public as a result of the proposed reconfiguration, for example, if hospital X is not offering A&E services, how much extra time will I have to travel in emergencies? There are challenges in analysing travel time data. Journey times can be impacted by several factors such as the use of multiple forms of transport to get to hospital, traffic congestion varies during different times of the day and as a result of incidents or road works, Page 11

12 and ambulance travel times vary significantly depending on the type of incident (they do not always travel under blue light ) and time of day. As a result there is no comprehensive point to hospital database of actual visitor and staff journeys using different forms of transport. Throughout the travel analysis conducted, a number of elements have been considered: Transport types considered Private Car (peak and off-peak) and Public Transport peak travel times Areas covered Times between acute hospital sites Times from small areas (lower super output areas) within NWL to acute hospital sites From areas outside of NWL where >20% patients currently use NWL services Types of analyses Impact on Travel times o Individual hospital changes o For configuration options Population weighted averages, 95th percentile and maximum travel times Health deprivation as a function of changes in travel time Estimated activity flows from major hospital if it were to no longer provide a service. As patients rightly do not consider themselves constrained by commissioning boundaries, the travel analysis has included hospitals surrounding NW London where residents are currently being treated, e.g. Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, University College London Hospitals, Watford, Wexham Park and Wycombe. The analysis has also considered non-nw London residents, from the rest of London or the surrounding counties, who are treated within the region 4. Figure 12.2 shows how the impact of service changes on travel times are calculated. 4 Residents from postcode regions where more than 1 in 5 residents are currently treated at NW London hospitals are included. Page 12

13 Figure 12.2: Calculating the impact of service changes on travel times 12.2 Step 5 Northwick Park and Hillingdon Hospitals are proposed as major hospitals in all options to minimise impact on access There are 126 potential ways of configuring the nine current acute sites in NW London as five major hospital sites. The next step of the rationale to determine the location for the major hospitals was to consider the impact on access for patients and the public. Page 13

14 Figure 12.3: Impact on blue light travel times for each borough when A&E destination is changed (absolute change in blue light travel time vs current configuration (mins)) 5 5 TfL HSTAT travel time data. Blue light travel times estimated at 67% off-peak private car travel times. Notes for figure 12.3: 1. Current configuration of major hospitals = Northwick Park, Hillingdon, Ealing, West Middlesex, St Mary s, C&W and Charing Cross. 2. Hospitals outside of NWL included in analysis = Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, UCLH, Watford, Wrexham Park and Wycombe 3. Travel times from population areas outside of London not included Page 14

15 The areas shaded in the darkest blue would experience an increase in travel time of between 6 34 minutes, if residents there had to go to their next nearest hospital. The maps showing the current configuration minus either Northwick Park or Hillingdon have the greatest amount of area shaded, demonstrating that a larger area (and therefore significant population levels) would have to travel further if either of these hospitals were no longer major hospitals. This is in comparison with the maps for St Mary s, Chelsea & Westminster and Charing Cross, where very little of the maps are shaded in the darkest colour and the overall area of impact is much smaller. Figure 12.4: Impact on private car travel times for each borough when A&E destination is changed (peak) 6 6 TfL HSTAT travel times data. Notes from Figure 12.4: 1. Peak time = Morning peak = 7am 10am 2. Current configuration of major hospitals = Northwick Park, Hillingdon, Ealing, West Middlesex, St. Mary s, C&W, Charing Cross, except for Brent PCT where current configuration is assumed to also have CMH in order to provide a comparison 3. Hospitals outside of NWL included in analysis = Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, UCLH, Watford, Wexham Park and Wycombe 4. Travel times from areas outside of London not yet included (will be updated with all LSOAs where >20% patients use NWL hospitals) 5. Population weighted average used for travel times Page 15

16 The analysis shown in Figure 12.4 further supports the likely impact on residents of each of the boroughs according to the different configurations. Should Northwick Park or Hillingdon no longer provide major hospital services, residents in Harrow and Hillingdon would be most affected, with peak car journeys likely to increase by a greater amount compared with other potential configurations. To minimise impact on access, local clinicians recommended that Northwick Park and Hillingdon should be major hospitals in all options because they are more geographically remote compared with the other sites Step 6 Central Middlesex should not be considered as a major hospital site After recommending that Hillingdon and Northwick Park be proposed as major hospitals, local clinicians considered if any sites were unsuitable to be major hospitals. Local clinicians recommended that Central Middlesex should not be considered as a major hospital due to insufficient levels of clinical quality to meet the standards for major hospitals and due to the size of the site. Over recent years, several services have been removed from the Central Middlesex hospital because they were clinically unsustainable. This includes emergency surgery, inpatient paediatrics and obstetrics. The site itself is particularly small, with 35,000 m 2 of clinical space and 245 beds. It also serves a small local catchment population. Should Central Middlesex be retained as a major hospital, because it is the smallest of the existing acute sites, it would be likely to require the largest proportional expansion of any of the current sites to accommodate the activity predicted for a typical major hospital Step 7 Geographic distribution of the remaining sites is proposed to minimise the impact of changes on local residents After recommending that Northwick Park and Hillingdon should become locations as a major hospital, and that Central Middlesex should not, there remained 20 possible configuration options utilising the other six sites. Local clinicians agreed that consideration should be given to the geographic distribution of the remaining options for the location of the other three major hospital sites. Local clinicians reviewed likely flows of activity, should a particular site no longer be available. For example, if patients normally travelled to Charing Cross, should Charing Cross not be a major hospital, the analysis considered the next hospital that they would go to. Figure 12.4 sets out the patient flows that would occur if either of West Middlesex or Ealing were not proposed as a major hospital. If West Middlesex became a major hospital and Ealing was not a major hospital, then around 52% of patients who currently use Ealing would go to West Middlesex instead. Transversely, 41% of patients would head to Ealing if it was a major hospital, and West Middlesex was not, where they currently use West Middlesex. By contrast, much lower percentages of patients (and ambulances) would switch to other hospitals. By considering an option of locating a major hospital at either Ealing or West Middlesex, the impact of any change on local residents will be minimised. Page 16

17 Figure 12.5: Potential activity flows for options based on Blue Light proxy travel time to next nearest hospital 7 The numbers highlighted in blue indicate those sites that would receive more than 20% of the activity of a hospital that was no longer offering major hospital services. For example, if Hillingdon was not designated a major hospital, then 71% of its activity would move to Ealing. The numbers show the close relationship between Ealing and West Middlesex with between 41% - 52% of activity transferring between the two sites, should one of them not be a major hospital. Other key relationships are between Hammersmith and St Mary s (where between 59%-62% of activity would move between the two sites should one of them not provide services) and between Charing Cross and Westminster (where between 43%-57% of activity would move between the two sites should one of them not provide services). This analysis was also completed for private car travel times, as shown in Figure TfL HSTAT travel times data. Notes from Figure 12.5: 1. Blue Light travel times estimated as 67% Off-peak Private Car travel times. 2. Current configuration includes Central Middlesex and Hammersmith to understand flow to / from those sites 3. Hospitals outside of NWL included in analysis = Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, UCLH, Watford, Wexham Park and Wycombe 4. Travel times from population areas outside of London where >20% patients use NWL hospitals are included Page 17

18 Figure 12.6: Potential activity flows for options based on Private Car travel time to next nearest hospital 8 The pattern for Figure 12.6 is very similar to Figure For example, the close relationship between Hammersmith and St Mary s is shown with between 74%-77% of activity transferring between the hospitals should one of them close. For Chelsea & Westminster and Charing Cross between 55%-59% of activity would move. If a major hospital is located at one of each of these key geographic pairings, the travel impact of this change on local populations is small. This analysis based on public transport travel times may be found in Appendix H. 8 TfL HSTAT travel times data. Notes from Figure 12.6: 5. Blue Light travel times estimated as 67% Off-peak Private Car travel times. 6. Current configuration includes Central Middlesex and Hammersmith to understand flow to / from those sites 7. Hospitals outside of NWL included in analysis = Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, UCLH, Watford, Wexham Park and Wycombe 8. Travel times from population areas outside of London where >20% patients use NWL hospitals are included Page 18

19 Figure 12.7: Changes in travel time for each borough where a major hospital is located at one of each of the key geographic choices (private car peak time) (% change) 9 9 TfL HSTAT travel times data. Notes from Figure 12.7: 1. Peak time = Morning peak = 7am 10am 2. Current configuration of Major hospital hospitals = Northwick Park, Hillingdon, Ealing, West Middlesex, St. Mary s, C&W, Charing Cross 3. Hospitals outside of NWL included in analysis = Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, UCLH, Watford, Wexham Park and Wycombe 4. Negative numbers show the improvement in travel times for some residents due to Hammersmith being in a new Major hospital in some of the options Page 19

20 The largest increase in travel times would be for residents in Ealing under configuration options where Ealing was not a major hospital. However they are likely to only see a maximum increase of between 10%-13% to their average travel time depending on the other variations of configuration. In contrast, if neither Ealing nor West Middlesex were designated as major hospital sites, residents in Ealing would be significantly affected. Under this scenario Ealing residents would be likely to experience between 23%-30% increase in travel time to reach the next nearest hospital as shown in Figure Figure 12.8: Changes in travel time for each borough if there is no major hospital located at either Ealing or West Middlesex (private car peak time) (% change) TfL HSTAT travel times data. Notes from Figure 12.8: 1. Peak time = Morning peak = 7am 10am 2. Current configuration of Major hospital hospitals = Northwick Park, Hillingdon, Ealing, West Middlesex, St. Mary s, C&W, Charing Cross 3. Hospitals outside of NWL included in analysis = Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, UCLH, Watford, Wexham Park and Wycombe 4. Negative numbers show the improvement in travel times for some residents due to Hammersmith being in a new Major hospital in some of the options Page 20

21 Figure 12.9: Changes in travel time for each borough if there is no major hospital located at either Chelsea & Westminster or Charing Cross (private car peak time) (% change) 11 Figure 12.9 further highlights the impact of not locating a major hospital at one of the sites in the geographic pairings. If neither Charing Cross nor Chelsea & Westminster were designated as major hospital sites, residents in Hammersmith and Fulham would be significantly affected. In this scenario, residents would experience between 48%-57% increase in average travel times to the next nearest major hospital. The analysis is completed by looking at the impact on residents if neither Hammersmith nor St Mary s are designated as a major hospital site. 11 TfL HSTAT travel times data. Notes from Figure 12.9: 1. Peak time = Morning peak = 7am 10am 2. Current configuration of Major hospital hospitals = Northwick Park, Hillingdon, Ealing, West Middlesex, St. Mary s, C&W, Charing Cross 3. Hospitals outside of NWL included in analysis = Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, UCLH, Watford, Wexham Park and Wycombe 4. Negative numbers show the improvement in travel times for some residents due to Hammersmith being in a new Major hospital in some of the options Page 21

22 Figure 12.10: Changes in travel time for each borough if there is no major hospital located at either Hammersmith or St Mary s (private car peak time) (% change) 12 For residents in Westminster, the analysis confirms that removing major hospital services from both Hammersmith and St Mary s would see average travel times for residents increase by around 27%. In addition, residents from the neighbouring boroughs of Hammersmith & Fulham and Kensington & Chelsea would also be negatively impacted. In reviewing this analysis, clinicians proposed that a geographic distribution of the remaining three locations for major hospitals to minimise the impact of changes on local residents, should be found. They agreed that the location of the remaining three major hospital sites would be between: Either Charing Cross or Chelsea & Westminster Either Ealing or West Middlesex Either Hammersmith or St Mary s. The Clinical Board recommended to the Programme Board that this was based on the analysis that demonstrated that: The impact of removing one hospital in each pair on peak car travel times concluded that removing one of each pair had little impact on travel times 12 TfL HSTAT travel times data. Notes from Figure 12.10: 1. Peak time = Morning peak = 7am 10am 2. Current configuration of Major hospital hospitals = Northwick Park, Hillingdon, Ealing, West Middlesex, St. Mary s, C&W, Charing Cross 3. Hospitals outside of NWL included in analysis = Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, UCLH, Watford, Wexham Park and Wycombe. 4. Negative numbers show the improvement in travel times for some residents due to Hammersmith being in a new Major hospital in some of the options Page 22

23 The impact on peak car travel times on populations if both hospitals in a pair were removed which showed a large impact on some residents Activity flows between Trusts based on blue-light and private car travel times which showed the highest patient flows were between the pairs The proposed eight options to undergo further evaluation As a result of the process set out above and in Chapter 11, local clinicians recommended that eight options were put forward as a list of options for more detailed analysis and evaluation. Each of these eight options located a major hospital at Northwick Park and Hillingdon Hospital and Central Middlesex was not proposed as a major hospital. Figure 12.11: Clinical Board recommendations for list of eight options for the location of major hospitals in NW London Page 23

24 Figure sets out the application of these options apply on a geographical basis. Figure 12.12: Map illustrating list of eight options to undergo further analysis Chapter 14 describes the evaluation performed on this list of eight options to identify a smaller group of options to undergo further detailed analysis. Once the options for consultation were identified, further work was completed to decide which locations were most appropriate for enhanced primary care services, elective centres and specialist centres. Page 24

25 13. Developing evaluation criteria to determine options for major hospital sites This chapter describes the evaluation criteria that were used for determining where the major hospitals should be located. The criteria were developed by clinicians and tested with wider groups and the public. The criteria are focused on quality of care, access to services, value for money, deliverability and the impact on research & education. This chapter sets out: Principles for developing options The evaluation criteria The questions beneath each of the evaluation criteria Principles for developing options Clear principles were applied to the work that has fed into the development and appraisal of reconfiguration options. The work has been clinically led, with recommendations coming from the Clinical Board (which includes clinical representatives for each provider and for each CCG) and an Expert Clinical Panel, which has provided external challenge to help test and refine our proposals. Further testing and refinement has taken place based on discussions with patients, patient representative groups, a wider clinical group, members of the public, local authorities and local HOSCs. Using these principles and by following a robust process, we have developed proposals for which service delivery model(s) will be located at each of the 9 existing acute hospital sites. The start of this process and the focal point for the overall set of recommendations for reconfiguring services in NW London began with the selection of locations for major hospitals. Following these decisions, further work was undertaken to confirm the location of other services. This chapter concentrates on the initial stage of the process around major hospitals. Chapters 16, 17 and 18 provide further detail on the recommendations for other services Evaluation criteria Further analysis of the potential options for consultation would be done using an agreed set of evaluation criteria, developed by clinicians with involvement from providers, patients and their representatives and the public. An initial set of draft evaluation criteria were developed and the tested at the engagement event on 15 th February Participants were asked to rank the criteria that were most important in determining how options should be evaluated. Page 25

26 Figure 13.1: Ranking of evaluation criteria by the public and clinicians at stakeholder event As expected, Quality of Care was the most important criteria for both the public and clinicians. There were not significant differences between the public and clinicians on other criteria, although the public ranked Patient Experience and Patient Choice slightly higher than clinicians. Some additional criteria were suggested by both groups during the event, as shown in Figure Figure 13.2: Additional evaluation criteria suggested by the public and clinicians The clinical board reviewed these suggestions and developed them further where appropriate. From this, additional aspects to the evaluation criteria were added and these are shown in Figure Page 26

27 Figure 13.3: Additional criteria that have been included in the evaluation criteria However, some of the additional suggested criteria were not included as they would not differentiate between different options. These are shown in Figure These criteria are vitally important but they are not being used as part of the evaluation criteria as they do not differentiate between sites where services may be located. Figure 13.4: Additional criteria that have not been included in the evaluation criteria A final set of evaluation criteria was refined and agreed by the Programme Board to be used as the basis for selecting which options should be included from the medium list into the list of options for further analysis. Page 27

28 Figure 13.5: Final set of evaluation criteria 13.3 The questions beneath each of the evaluation criteria It is important to understand that these evaluation criteria are useful only if they enable the JCPCT to differentiate between different options for consultation. Some criteria will be considered more important than others; but some of the less important criteria will have more influence because they enable the JCPCT to see the relative merits of options. This section describes how each sub-criterion was used, the question that was being answered and the measures that are being used as part of the evaluation to differentiate between the eight options, effectively by comparing the three pairs of sites: Charing Cross Hospital or Chelsea and Westminster Hospital Ealing Hospital or West Middlesex Hospital Hammersmith Hospital or St Mary s Hospital Clinical quality The importance of clinical quality cannot be overestimated. This criterion was the basis of the proposal to have five major hospital sites to ensure that there is sufficient, senior clinical staff to provide safe sustainable rotas. Evaluation question: Which options would provide better clinical quality in future using clinical surveys and indicators? The assessment used current mortality rates at Trust rather than site level. The Clinical Board was mindful that current clinical quality at Trust level was not a useable proxy for future clinical quality at site level after reconfiguration was complete. Therefore, even though clinical quality is the most important criterion, it does not help to differentiate between pairs of options. Page 28

29 Patient experience Patient experience will improve across all the options due to the higher clinical quality of care. There may be some differentiation that can be determined by assessing the level of patient experience at each of the Trusts and the quality of estates on each of the sites. Evaluation question: Which options would provide a better experience for patients using patient experience surveys and looking at the quality of the buildings and facilities? The measures used for the assessment were: Patient experience data using CQC standardised scores for: o How would you rate the care you received? o Did you feel you were treated with respect? o Were you involved as much as you wanted to be? Quality of estates, looking at: o Area of not functionally suitable NHS space o Estate dating post-1964 o Estate dating post Distance and time to access services Distance and time to access services is always among the chief concerns of the public and patients. This criterion was a fundamental part of the basis for creating the medium list of eight options and proposing that Northwick Park and Hillingdon as major hospital sites. The medium list therefore covers three major hospitals in Inner NW London, Hounslow and Ealing and therefore changes in travel across the options is significantly less than in the first stage of options development. Evaluation question: Do any options keep to a minimum the increase in the average or total time it takes people to get to hospital by ambulance, car (at off-peak and peak times) and public transport? The measures used for the assessment were: Blue light travel times (average and maximum) Private car off peak travel times (average and maximum) Private car peak travel times (average and maximum) Public transport peak travel times (average and maximum) Patient choice Patient choice is a one of the 4 Tests set out by the Secretary of State as well as a principle within the NHS. The service reconfiguration options centralise specialist services, which is necessary to provide high clinical quality. Alongside this, NW London still aims to maximise the choice of high quality services available to its population. Evaluation question: Which options would give people in NW London the greatest choice of hospitals for emergency care, maternity care and planned care across the greatest number of trusts? The measures used for the assessment were: The reduction in the number of sites delivering: o Emergency care o Obstetrics Page 29

30 o Elective Care o Outpatients and diagnostics The number of Trusts with major hospital sites Capital cost to the system The future configuration of services needs to be affordable for the system to be sustainable. A major part of the financial assessment is the capital expenditure that would be needed to create any option and raise the quality of the estates to the standards that have been set. Evaluation question: Which options would have the lowest capital costs (cost of buildings and equipment)? The measures used for the assessment were: Estimated capital costs for new additional capacity, based on the number of new beds for each site Estimated capital costs for relocating specialist services, including maternity services Estimated costs of completing backlog maintenance at each existing site Estimated net receipts from land sale when sites become local hospitals Estimated capital cost per site for building new local hospitals Transition costs The implementation plans show that there is a need to create the future services out of hospital and at the major hospital sites before the services are decommissioned elsewhere. The cost of double-running will contribute to the relative cost of each configuration option. Evaluation question: Which options would have the lowest cost of transferring services between hospitals? The measures used for the assessment were: The number, and hence the cost, of beds transferred between sites for: o Adult, specialist and critical care o Maternity o Paediatrics Viable Trusts and sites Another aspect of the affordability and sustainability of any future reconfiguration is the viability of each site and Trust, given the expected future activity patterns. Evaluation question: Which options would have the lowest yearly subsidy and the fewest hospitals and Trusts with a financial surplus of less than 1% (the lowest acceptable level of financial surplus allowed for Trusts in the NHS)? The measures used for the assessment were: Estimated subsidies required to achieve a 1% net surplus for each site Estimated number of Trusts achieving a 1% net surplus Estimated number of sites achieving a 1% net surplus. Page 30

31 Surplus for acute sector The assessment of viable Trusts and sites indicates the degree to which each component of the acute sector is viable. This measure to look at the overall surplus for the acute sector enables commissioners to assess the relative sustainability of the overall acute sector. Evaluation question: Which options will give the largest financial surplus across all hospitals, to make sure that the proposed changes are affordable? The measure used for the assessment was: Total annual surplus / deficit across NW London acute sites Net Present Value A fundamental part of any economic assessment is the Net Present Value (NPV) of the proposed changes. Commissioners agreed the financial assumptions within the NPV calculation and the length of time it would take for each option to be completed. This allows us to calculate the NPV for each option relative to doing nothing. Evaluation question: Which options will give the largest Net Present Value (overall financial benefit) over the next 20 years? The measures used for the assessment were: The Total NPV, relative to doing nothing, calculated by: o Adding the benefits, which were: Consolidation savings Net change to fixed costs Capital receipts. o And subtracting the investments needed and the costs, which were: Up front capital investment Ongoing replacement Capex Operating costs for new assets One-off transition costs Workforce The availability of sufficient workforce with the right skills was a significant factor in determining that we need five major hospitals in NW London. Once this number was established, the programme wanted to assess whether there was any differentiation between options based on workforce. For this to be the case, we therefore needed to assess qualitative rather than quantitative measures and these were based on whether staff performed well and were satisfied in their work. Evaluation question: Which options will provide the best workplace for staff using staff satisfaction surveys? The measures used for the assessment were: Staff turnover rates Staff sickness rates Staff recommendation as a place to work or receive treatment Staff job satisfaction Page 31

32 Staff satisfied with the quality of work and patient care Expected time to deliver The main criterion that clinicians used to determine that five major hospitals were needed for NW London (rather than the clinically acceptable range of three to five) was the length of time it would have taken to deliver a three or four major hospital model. Within the remaining options with five hospitals, there are still a range of delivery timescales that are all acceptable. However, we would want to assess the likelihood of delivery for each option (a proxy for the management case in business case terms) and whether this is different for each option. Evaluation question: How long will it take to deliver the proposed changes in each option? A shorter delivery time means that benefits can be delivered earlier. The measures used for the assessment were: The number of sites already delivering the services The new capacity required The volume of acute beds that would be moved The volume of maternity beds that would be moved Fitting in with other strategies Clinical practice and demands on the health system are constantly evolving. Therefore we prefer options that are most likely to enable us to respond to future changes. Evaluation question: How well does each option fit with what is happening, or may happen, nationally or in London? The measures used for the assessment were: Fit with previous Major Trauma designation Fit with previous stroke designation for Hyper-Acute Stroke Units and Stroke Units Fit with national initiatives: o Transparency agenda o Enhancing and improving out of hospital care o Integrated care o Driving improvements in acute services, particularly out of hours o National QIPP challenge. Fit with broader London initiatives: o Primary care o Integrated care. Fit with local strategies in place or in development: o Inner NW London Integrated Care Pilot (ICP) o Mental Health ICP o Pathology modernisation programme o Ongoing work by cancer, cardiac and other networks. Page 32

33 Disruption Minimising disruption to research and education during any service change is important. The assessment of this criterion depends as much on what happens to sites not designated as major hospitals as much as those that are. As Hammersmith would be a specialist hospital if not a major hospital, this limited the differentiation that was possible with this criterion. Evaluation question: Which options best fit with current research and education to minimise disruption in these areas? The measures used for the assessment were: Research spend at non-major hospital and non-specialist hospital sites Education spend at non-major hospital and non-specialist hospital sites Support current and developing research and education delivery The final sub-criterion was to look at the degree to which each option supported current research. Education was not used as this can move with clinical activity and therefore would not help differentiate between options. Evaluation question: Which options best support what is happening in research and education? The measure used for the assessment was: Space allocated to research on each site Rating each option for each criterion For each criterion, each option is given a rating. Rather than an absolute score, a relative evaluation is given to differentiation between the eight options. The ratings are shown in Figure Figure 13.6: Ratings used in evaluations Page 33

34 14. Appraisal of options for locating major hospital sites This chapter describes the process for evaluating the list of eight options for locating major hospital sites. By applying the evaluation criteria, each of the eight options were evaluated across a range of factors. The evaluation showed that three options were rated better than the others and should be considered further. In all these three options St Mary s would be designated as a major hospital and Hammersmith would remain as a specialist hospital. Option 5 (Hillingdon, Northwick Park, St Mary s, West Middlesex and Chelsea and Westminster) emerged as the strongest, preferred option. Across the remaining two options for major hospitals there is still a choice between Ealing and West Middlesex, and between Charing Cross and Chelsea and Westminster. The chapter sets out: The assessment of the eight options against the evaluation criteria o Quality of care o Access to services o Value for money o Deliverability o Impact on research and education. As described in Chapter 13, a list of eight options was recommended by local clinicians to undergo further evaluation against the agreed evaluation criteria. It was important to note that the first two evaluation criteria, Quality of Care and Access to Care, were essentially used as hurdle criteria to determine the eight options, as set out in Chapters 11 and 12. However the next stage of assessment used all five evaluation criteria. For all evaluation criteria, it was noted that for some multi-site Trusts it was not possible to obtain specific information for given sites. In those instances it was assumed that that each site was equivalent to the overall Trust position. Northwick Park and Hillingdon have been included throughout the evaluation analysis to make sure that the full costs of implementing the reconfiguration have been taken into account. Figure 14.1 sets out the eight options that were evaluated against the criteria. Figure 14.1: Summary of eight options to undergo further evaluation Page 34

35 14.1 Review of eight potential options against Quality of Care Two criteria, Clinical Quality and Patient Experience, contributed towards the evaluation of options under the Quality of Care heading. The analysis agreed by local clinicians for both of these criteria was as follows: Clinical Quality Review whether or not the option can deliver against the clinical standards - assessment of ability of option to deliver access to experienced, skilled staff and specialist equipment Comparison of current clinical quality of sites which are expected to deliver future inpatient activity under each option Patient experience For each configuration option, which are expected to maintain the highest quality, as assessed by reviewing patient experience surveys Three CQC patient experience measures used by Dr Foster Trust awards (overall care, respect and involvement in decisions) Number of sites with better quality of estate expected to be a site in future configuration Clinical Quality The importance of quality should not be underestimated and it was variation in quality standards which provoked the reviews in Emergency and Paediatric Care, the results of which have informed the development of the reconfiguration proposals. Throughout the analysis described so far, local clinicians have always required that each option under consideration must be of sufficient quality to meet the standards they have set out and indeed the first step in the first part of the process (determining the number of hospitals) stated that major hospitals were required to ensure high quality care. At this stage of the evaluation local clinicians agreed to review further analysis on clinical quality data to see whether this would provide more detailed information to differentiate between the eight options being evaluated. Page 35

36 Figure 14.2: Clinical Quality Data 13 The data set out by in the Dr Foster reports (as shown in Figure 14.2) suggests that out of the NW London PCTs, Chelsea & Westminster achieved statistically better than expected results for all four mortality rates during 2010/11. Imperial also fared well with mortality rates better than the national average in three out of the four measures. However Hillingdon did not do better than expected against any of the measures and Ealing only did better against one. This set of data suggests that clinical quality is better, due to better performance of mortality rates, at Chelsea & Westminster and Imperial hospitals. However, another useful source of data on clinical quality are the Quality Dashboards produced by the East Midlands Quality Observatory. A summary of the outputs of these Quality Dashboards is set out in Figure Notes from Figure 14.2: 1. Dr Foster standardised mortality rate for 3 years. The current three year HSMR is for the financial years 2008/09, 2009/10 and 2010/11; Summary Hospital-level Mortality Indicator (SHMI), and Deaths After Surgery, 2010/11, National average is Includes Charing Cross, Hammersmith and St. Mary s hospitals 3. Includes Northwick Park and Central Middlesex 4. Dr Foster Mortality Ratios 2010/11, Deaths in low risk conditions per 1,000 patients Page 36

37 Figure 14.3: Quality Dashboard data 14 In stark comparison to the Dr Foster data, according to the Quality Dashboard scores, Hillingdon is the top performer in NW London with better than national average performance against 62 of the metrics. Ealing has the lowest number of metrics that are above national average, however, the range between Ealing and Hillingdon is fairly small, with all the other Trusts having similar scores. It is not clear whether these scores help to sufficiently differentiate between the levels of clinical quality that are being achieved between the different Trusts. As mentioned at the start of this section, local clinicians have been clear since the start of Shaping a healthier future that clinical quality is at the heart of the programme and that it is the driving force behind all the proposals and recommendations. The reconfiguration is being pursued to achieve the clinical standards and the improved clinical quality through the reshaped clinical service delivery models set out in Chapter 8. After reviewing the data available on clinical quality, local clinicians agreed that all the eight options under consideration had been designed to achieve the highest levels of clinical quality and that the additional data reviewed at this stage of the evaluation did not provide any significant information that allowed them to differentiate between options on this basis. All eight options were evaluated to provide high quality of clinical care. A summary is provided in Figure East Midlands Quality Observatory: Note from Figure 14.3: 1. The number of metrics that scored better for the Trust than national average, out of a total of 104 metrics (this takes into account metrics where a lower score is better (e.g. mortality) and metrics where a higher score is better (e.g. % patients seen within X weeks) Page 37

38 Figure 14.4: Quality of Care Clinical Quality Evaluation Patient Experience The second part of evaluating Quality of Care was around patient experience. The first measurement of patient experience was on quality of estate. There is a growing body of evidence to support the view that the quality of the hospital or clinic where a person is treated is associated with their experience as a patient and the outcomes that are produced. This view was supported by the Clinical Board. In addition, there are a number of articles in the literature supporting this hypothesis, and Appendix I contains some abstracts from these articles. Figure 14.5 sets out an overview of the quality of the estate of the current nine acute hospital sites in NW London. Page 38

39 Figure 14.5: Quality of acute estate in NW London 15 Chelsea & Westminster, West Middlesex and Central Middlesex are considered the sites with the highest quality of estate. These sites consist of very recently built buildings with space that is suitable for the current and future requirements. Ealing, St Mary s and Northwick Park are considered to be sites with poor quality estate. St Mary s and Northwick Park have high levels of floor space that is considered not functionally suitable NHS space. In addition 56% of St Mary s estate was built pre-1964 and therefore suffers from issues typical of older buildings. In Chapter 15, consideration is given of the capital implications for each of these options to ensure that the quality of estate reaches the necessary levels to achieve the clinical standards. As evidence has shown, there will always be limitations when using estate that is older. A second indicator of patient experience is the data sourced from the National NHS Patient Survey Programme, carried out annually by the Care Quality Commission (CQC). The latest data available is from ERIC Site-level data, HEFS, 2010/11 ( Notes from Figure 14.5: 1. ERIC Site-level data, HEFS, 2010/11 ( 2. Qualitative assessment Page 39

40 Figure 14.6: Patient Experience Data 16 Figure 14.6 shows that some Trusts in NW London had results from the National NHS Patient Survey that were statistically worse than the national average. Both Ealing and North West London Hospitals (Northwick Park and Central Middlesex) had two measurements that were worse than the national average. West Middlesex performed poorly against one indicator. However, the difference between all the scores is minimal and indeed the national scores have a very small range. Local clinicians did not feel that using this data in isolation gave them sufficient basis to differentiate between the options. Combining these different elements of patient experience, local clinicians agreed that some options were likely to offer patients a better experience than others. This is summarised in Figure Notes from Figure 14.6: 1. Standardised Patient Satisfaction Scores, National NHS Patient Survey Programme, 2010 Survey of Inpatients, CQC Page 40

41 Figure 14.7: Quality of care Patient Experience evaluation Options 1 and 5 achieved the highest evaluations. Both of these options included Chelsea & Westminster and West Middlesex, therefore achieving high evaluations for the quality of estate. Neither of these options contained Ealing which would have lowered its evaluation due to its poor quality of estate and the lower scores it achieved on the National Patient Survey. Options 4 and 8 were given the lowest evaluations. Both of these options include Ealing but do not include one or both of Chelsea & Westminster and West Middlesex. The other hospitals do not impact any of the evaluations between the options. All options include Northwick Park as a major hospital and exclude Central Middlesex as a major hospital (as set out in Chapter 12), therefore they are all considered to be equally impacted by the low rated quality of estate at Northwick Park and the slightly lower than national average scores on the National Patient Survey for West Middlesex. As described in Chapter 12, Central Middlesex has been designated as an elective hospital. This will ensure that the quality of its buildings will be utilised in all options Review of eight potential options against Access to Care Two criteria, Distance and time to access services and patient choice, contributed towards the evaluation of options under the Access to Care heading. The analysis agreed by local clinicians for both of these criteria was as follows: Distance and time to access services Impact on population weighted average travel times for each option due to reconfiguration, based on activity volume and travel time estimations Blue light travel times Off-peak car times Peak car times Public transport times Page 41

42 Patient choice Reduction in the number of sites delivering specific services Emergency care Obstetrics Elective care Outpatients and diagnostics Number of Trusts with major hospital sites Distance and time to access services Understandably, all patients want to access excellent NHS services as close to their homes as possible, therefore the distances and time taken to access services was key to the development of the proposals. Figure 14.8: Time to major hospital services 17 Due to the importance of travel time and access to the proposed major hospitals, this analysis was one of the main factors in determining the list of eight options that clinicians recommended for further evaluation (see Chapters 11 and 12). All eight options have been rated the same in recognition that this analysis has been used in the development of the options and that the analysis has not enabled any differentiation between the options. All eight options have been given a slightly negative evaluation as by reducing the number of sites that provide services as a major hospital, and as demonstrated in the earlier analysis (see chapter 12), we know that for some residents in certain boroughs, moving to a 17 Transport for London: HSTAT travel time model. Notes from Figure 14.8: 1. Population Weighted average travel time to nearest Major hospital within NWL; population weighted travel times by postcode area (LSOA), morning peak 7-10am; hospitals outside of NWL included in analysis = Barnet, Guy s, Kingston, Royal Free, St George s, St Thomas, St Peter s, UCLH, Watford, Wexham Park and Wycombe 2. Blue Light travel time estimated as 67% of TfL HSTAT off-peak Private Car travel times 3. Longest travel time for Blue Light, Private Car and Public Transport is for a postcode area in Hillingdon Borough, which is unchanged in all options (including the current configuration Page 42

43 configuration with five major hospitals rather than the current nine existing acute sites, will result in some additional distance and travel times to get to their nearest hospital. Distance and travel times are not the only important factors in terms of access. Other factors such as opening times and translation services are also important. However local clinicians agreed that these factors are either driven by Out of Hospital services or are able to provided equally under all the options Patient choice Figure 14.9: Review of patient choice 18 Giving patients choice over where they receive treatment or undergo diagnostics is a key principle for the NHS. Patient choice is generally assumed to be greater if there are several suitable options for them to consider where they would like to use services. However, it is also recognised that the priority for patients is to receive the best care possible with the highest chance of a positive clinical outcome and a good patient experience. Taking this into consideration local clinicians agreed that all the options will provide better access due to the increased number of high quality services as well as increased access to specialists and diagnostics outside of the hospital environment, as set out in the Out of Hospital strategies (see Chapter 7). Figure 14.9 shows the reduction in the number of sites delivering particular services according to each of the eight options. Each of the options would see a reduction of three 18 Notes for Figure 14.9: 1. Does not include CMH as the site is not currently delivering 24/7 emergency care 2. Assumes that if C&W was not an major hospital site that CHX would provide new maternity service 3. Assumes that all major hospital and specialist hospital sites will provide some complex elective care however all sites may not provide non-complex elective care; Assumes CMH is an elective hospital under all options Page 43

44 Emergency Care sites. All the options would also see a reduction in the number of sites providing Obstetrics and Elective Care services, however this would be slightly greater for Options 1 to 4, where they would lose two Obstetric sites and three Elective Care sites compared to a reduction of one Obstetric unit and two Elective Care sites under options 5 to 8. Under options 5 to 8, Hammersmith is not designated as a major hospital and would be designated as a Specialist Hospital. Under these options Hammersmith would retain Queen Charlotte s Hospital, providing an additional obstetric unit. It would also retain an elective care unit. There would be no change under any of the options in the number of sites providing outpatient and diagnostic services. A further consideration under patient choice is the range of Trusts with major hospital sites that patients could access under the different options. Those options that locate a major hospital at Chelsea & Westminster rather than at Charing Cross result in five Trusts having a major hospital. Where Charing Cross is designated a major hospital then only four Trusts have major hospitals, and Imperial Trust would contain two major hospitals instead of one. After reviewing this analysis on patient choice, local clinicians agreed that options 5 and 7 should be given the highest evaluation. These two options result in a lower reduction of sites in obstetric and elective care as well as leading to five Trusts having major hospitals. Options 2 and 4 were given the lowest evaluation as these options both lose two obstetric sites and three elective care sites, as well as the options leaving only four Trusts with major hospitals Review of eight potential options against Value for money Five criteria, Capital Costs, Transition Costs, Site viability, Total surplus / deficit and Net Present Value contributed towards the evaluation of options under the Value for money heading. Figure sets out a summary of the evaluation undertaken that was agreed by the Finance & Business Planning Group to understand how each of the options measured against the different elements of the criteria. Page 44

45 Figure 14.10: Summary of the evaluation undertaken on eight options `against Value for money Capital cost to the system Each of the options was assessed for how much capital cost would be incurred as a result of their implementation. It was agreed that there were five components of potential capital cost: Adding new capacity Relocating maternity and other services Meeting backlog of maintenance Receiving net receipts from selling land Costs to build new Local Hospitals. Capital costs have been calculated to ensure a valid comparison can be made between options. The overall quantum of capital costs ensures that the overall affordability can be appraised. However, in any business case completed after consultation it will be necessary to go into substantially more detail on site configuration. The capital expenditure estimates used to evaluate the different options can be summarised as: 19 Notes for Figure 14.10: 1. Capital required to move specific services that require significant infrastructure changes, e.g. maternity units (including NICU) or specialist paediatric services 2. NPV is calculated on future cash flows over 20 years discounted at 3.5%pa. The decision to assess NPV over 20 years is a matter of judgement, and was selected by the F&BP board to strike a balance between accounting for the ongoing benefits post-reconfiguration, while not giving undue weighting to the long term forecasts that are necessarily less accurate (which may be the case with a longer period.) Note that the out of hospital investment is assumed to be the same in the do nothing scenario and all reconfiguration options, so is excluded from the NPV calculations comparing the acute reconfiguration options 3. Evaluation focuses on acute business only excludes non-acute parts of Trusts (e.g. Ealing CHS) 4. Assessment against 1% net surplus has been used as the minimum requirement, however for ongoing viability as a Foundation Trust, hospitals would require higher surpluses an assessment against 2% net surplus has been undertaken in the sensitivity analysis Page 45

46 The Capex estimates used in the options evaluations focus on those investments that would be directly related to the reconfiguration options, and that will differentiate the options from each other and from the do nothing scenario. These estimates include capital required to: o Add capacity to accommodate the changes in activity due to the reconfiguration o o o o o Build new Local Hospital facilities Move services between sites due to the reconfiguration option (e.g. linacs at Charing Cross; NICU and specialist paediatric surgery at Chelsea & Westminster) Increase capacity for maternity services Dispose of estate when changing the site to a Local Hospital (net receipts from disposal) Cover high risk backlog maintenance (included here because this is assumed to be one off additional spend over and above the ongoing programmes to replace assets, and the requirement differentiates the option as it is not required for sites becoming Local Hospitals) The estimates exclude capital that does not directly relate to the reconfiguration options and so does not differentiate the options from each other or from the do nothing scenario, e.g. the capital required to: o o o o Continue the ongoing replacement of assets Make changes to the estate to support or enable the delivery of the Trust CIP programmes Make incremental improvements to the estate in line with Trusts visions for the future of their sites Changes to estate that are independent of this reconfiguration programme (e.g. any proposed movement of the Western Eye site) It is recognised that in any reconfiguration option, as well as the do nothing scenario, Trusts will need to source and spend capital for schemes that are not included in this comparative analysis (see above). These may be funded through a combination of: o o Land and asset disposal not included in the reconfiguration evaluation Business cases for capital funding. Any applications would be over-andabove the option-specific funding estimated here o Depreciation charge 20 Trust and site income & expenditure forecasts include the annual depreciation charge and the additional impact of the capital spend for the schemes listed as included (modelled through income & expenditure via depreciation (over 20 years), public dividend capital at 3.5% and 4% operating costs) All estimates are subject to review and changes as Trust and site strategies are developed, and post-consultation business cases are worked up in more detail Adding new capacity As set out in the base case scenario (described in Chapter 6), there will be an overall reduction of 979 beds, across general wards, plus 26 beds across paediatric and maternity services. However, according to the different options, different sites will need increases or decreases in their specific number of beds. The movement of beds between sites will drive the forecast capital costs. 20 For information, the total 3-year deprecation charge 12/13-14/15 included in the do nothing forecasts for Trusts are: Imperial ( 109m), West Middlesex ( 15m), Ealing ( 12m), NWLH ( 38m), Hillingdon (includes Mt Vernon) ( 22m), Chelsea & Westminster ( 38m) Page 46

47 Figure sets out the estimated number of additional adult beds that will be required by each site, and the associated capital costs of this, for each of the options. Figure 14.11: Estimated capital costs for additional new capacity 21 Options 1 to 4 require significant additional capacity compared to options 5 to 8. This is mostly driven by the inclusion of Hammersmith as a Major Hospital in options 1 to 4. Hammersmith currently has very little capacity to provide Major Hospital services and mainly comprises specialist activities and Local Hospital services. In options 2 and 4, less capacity is required at Hammersmith, however a very similar amount of additional capacity would also be required at Charing Cross in these options, again due to the current services being provided at Charing Cross being primarily specialist and elective. Options 5 to 8 require much less additional capacity, with capital costs at around 10% of the costs for options 1 to 4. There is only a 3 million range between these options. In options 5, 6 and 8, additional capacity would be required at 3 out of the 5 proposed Major Hospitals whereas in option 7 additional capacity would only be required at 2 of the 5 sites Relocating maternity and other services In addition to the new capacity required, some specialist services will need to be relocated in each option. These services have highly specialised requirements, for example relocating maternity services means ensuring adequate theatres, delivery rooms and specialist equipment to be relocated as well. Figure 8.12 sets out the estimated capital spend by site for each of the options to relocate maternity, paediatric and diagnostic services. 21 Notes for Figure 14.11: 1. Figures include new beds required at the site based on current capacity, future required beds, and requirement for 5% headroom in terms of available unused beds Page 47

48 Figure 14.12: Estimated capital costs for relocating specialist services, including maternity services Options 2, 4, 6 and 8 would incur over double the level of capital spend for relocating specialist services compared to options 1, 3, 5 and 7. This is driven by the location of a major hospital at Charing Cross as it currently has no maternity services at present and would need to provide these services as part of the major hospital configuration. Options that include Hammersmith and Ealing as major hospitals also see larger capital spends to relocate specialist services, again mainly driven by the need to relocate maternity services Meeting backlog of maintenance Some of the existing acute hospitals in NW London have significant outstanding maintenance requirements. The backlog of maintenance represents the essential spend on high risk areas to ensure that the site provides a safe environment. Further spending would be needed to bring some of the sites up to the highest quality standards. It has been assumed that where a site is designated as a major hospital, capital spend would be required to deal with the backlog in maintenance. For some sites that could be designated as a local hospital, any backlog maintenance cost will be avoided as the new build cost is included in cost to build new local hospitals within this capital cost evaluation. However, where West Middlesex and Hammersmith are designated as local hospitals the small amounts of backlog maintenance would be included due to the modern estate (where there is no requirement to re-build). Figure sets out the capital costs of completing the current outstanding high risk backlog maintenance items for each of the existing nine sites in NW London. Page 48

49 Figure 14.13: Estimated costs of completing backlog maintenance at each existing site Northwick Park requires the largest amount of capital spend to bring the site up to safe working standards as a Major Hospital. Northwick Park and Hillingdon have been designated as Major Hospitals for all options in the medium list and therefore these capital costs will be included in all the options Receiving net receipts from selling land If a current acute site is designated as a Local Hospital, some of the sites would be able to reduce in size, potentially leading to the sale of land. There would be some costs associated with the sale of estate and Figure sets out the estimated net receipts that would result from the sale of land once certain sites became Local Hospitals. Page 49

50 Figure 14.14: Estimated net receipts from land sale when sites become Local Hospitals 22 Under the all the reconfiguration options Central Middlesex and West Middlesex sites retain elective services within their PFI buildings consequently land sales would not be possible. The Hammersmith site retains either major hospital or specialist services under all options and consequently no land sales are to be expected. Any spare estate at Hammersmith, West Middlesex and Central Middlesex would be used for other services. Despite having the least amount of land available for disposal, any sale of land from St Mary s or Chelsea & Westminster would result in large receipts of around 30 million. Charing Cross and Ealing have larger amounts of land that would be disposed of if they became Local Hospitals, but the lower value of land in those areas would result in net receipts of 7.6 million and 9.9 million respectively Cost to build new Local Hospitals As described in chapter 11, determining the number of major hospitals was based upon an assumption that no new sites would be built to support the reconfiguration of services in NW London. However, capital spend would be required to adapt some sites if they were designated as local hospitals. Figure sets out the estimated capital costs of building new local hospitals where the sites have been designated as such. 22 Analysis by EC Harris. Notes for Figure 14.14: 1. Total estimated footprint for Local Hospital including clinical and non-clinical space, car park, circulation, etc. (EC Harris) 2. Assumes that disposal value per Hectare is independent of the area being sold this would need to be revisited based on the specific parts and configuration of the sites being sold 3. Exit costs assume new build reprovision of third party interests (e.g. University, other Trusts, etc.) currently operating from site 4. Demolition costs estimated at 100 per square metre for the current buildings 5. Ealing excludes any joint development with West London Mental Health Trust who are located at the same site Page 50

51 Figure 14.15: Estimated capital cost per site for building new local hospitals It has been estimated that the cost of building a local hospital at any of the current sites would be 20 million 23. Options 1 and 2 would see the highest level of capital spending as 3 sites would be designated as Local Hospitals. In options 3, 4, 5, and 6, only 2 sites are designated as local hospitals and options 7 and 8 have the lowest capital cost with only 1 site becoming a local hospital. During this initial evaluation no costs have been marked against West Middlesex and Central Middlesex as they are PFI sites and any spare estate could be used for other services Summary of estimated capital cost to the system A summary of the estimated capital costs for each of the options is set out in Figure The local hospitals assumed 4000m 2 at 5000 per m 2, i.e., acute new build costs. 24 Preliminary work on potential local hospitals in Chapter 16 suggests that there may be some additional refurbishment costs for Central Middlesex and West Middlesex (depending on the option). However, these costs do not affect the evaluation detailed in Chapter 14. After consultation, further work will be done as part of the next business case. Page 51

52 Figure 14.16: Summary of capital costs to the system 25 Options 1 to 4 have a much higher level of capital cost compared with options 5 to 8. The main differentiator is the capital costs that would be incurred in adding the new capacity for options 1 to 4, which as discussed in Chapter 15, is driven by the additional capacity required where Hammersmith has been designated as a major hospital Transition costs It will not be possible for the reconfiguration of services to happen across NW London at a single moment in time. Services will need to be moved gradually to ensure safe working and to minimise disruption for patients. Cost estimates for the transition between the current configuration and the future set up have been made, based on 12 months of disruption at 250 cost per bed-day. These estimates focus on the transition of three categories; adult, specialist and critical care beds, maternity beds and paediatric beds, as these comprise the bulk of the movement of beds through reconfiguration. Estimates are intended to be indicative of the relative costs for options evaluation. Further work will be required for detailed plans on implementation and phasing for the preferred option. Figure sets out the estimates for these transition costs for each of the options. 25 Notes from Figure 14.16: 1. Evaluation: do nothing or less (++); <= 100m from do nothing (+); 101m- 200m from do nothing (-); > 200m from do nothing (--) 2. Risk adjusted backlog maintenance Page 52

53 Figure 14.17: Estimated transition costs 26 All options would have transition costs, however options 1 to 4 score worse than options 5 to 8, with estimated transition costs being about 30 million higher. Options 1-4 would involve greater disruption with larger numbers of beds moving between sites compared to options 5 to 8. This is mainly due to the expansion that would be required at Hammersmith if it were designated as a Major Hospital Site & Trust viability As set out in the Case for Change, it is crucial that the NHS system in NW London is financially sustainable going forward. One of the key factors in ascertaining whether a site or Trust is viable is whether it is able to achieve a minimum of a 1% net surplus. The base case scenario showed that at least four Trusts would not be able to achieve a 1% net surplus going forward in their current configuration. Figure summarises the modelling that forecasts the ability of each site to achieve 1% net surplus. 26 Does not include elective beds as these services are assumed to transfer directly without transition phases Page 53

54 Figure 14.18: Summary of estimated achievement of 1% net surplus by site 27 All of the options are an improvement on the do nothing scenario both in terms of the gap to achieve 1% net surplus and numbers of sites or trusts not achieving 1% net surplus. Options 1, 2, 5 and 6 are given a higher evaluation as these see the lowest gap to get sites to 1% net surplus. Options 3, 4, 7 and 8 are given strongly negative evaluations. All of these options would leave at least 3 sites (and for option 8 it would leave 4 sites) not achieving a 1% net surplus. These four options would have a substantial ongoing gap of million. In all options, Central Middlesex will struggle to achieve a 1% net surplus. Plans will need to be developed to address this issue Total surplus / deficit A further measure to evaluate whether Trusts are viable going forward is to look at their total surplus/deficit. 27 Notes for Figure 14.18: 1. Forecast surpluses in years after all reconfiguration changes have taken place and transition costs and income loss occurred in previous years. For modelling and evaluation purposes the steady state figures are based on 14/15 forecasts, but changes may take longer to be complete 2. > 30m subsidy or >=3 sites requiring subsidy (- -); 20m- 30m and <3 sites (-); < 20m and >=1 sites (+); all sites >1% (++) 3. Only sites and Trusts more with forecast deficits more than 1m less than 1% counted due to margin of error Page 54

55 Figure sets out the estimated annual net surplus in total for each option once reconfiguration has been fully implemented. Figure 14.19: Estimated total surplus / deficit for each option 28 Under the do nothing scenario, the total annual surplus / deficit across Trusts in NW London would be a deficit of 8 million which represents -0.4% of total income. Across all options there is a range of total annual surplus from million, representing 0.7%- 2.6% of income. Options 1, 2 and 5 have been given a positive evaluation as these see the highest surplus of greater than 2%. Options 3, 6 and 7 have a slightly negative evaluation with an estimated surplus of between 1% - 2%. Options 4 and 8 are given a strongly negative evaluation as they achieve the lowest levels of surplus of 1% or less. 28 Notes for Figure 14.19: 1. Forecast surpluses in years after all reconfiguration changes have taken place and transition costs and income loss occurred in previous years. For modelling and evaluation purposes the steady state figures are based on 14/15 forecasts, but changes may take longer to be complete the relative differences between options should still be indicative of the differences 2. Scores determined based on net surplus as percent of income: <1% (- -); 1-2% (-); 2-3% (+); >3% (++) Page 55

56 Net present value Several factors are important in calculating the economic benefits of the reconfiguration options through a Net Present Value figure, which allows the consideration of different payments profiles over time. Figure sets out the assumptions that have been made in this area of modelling. Figure 14.20: Assumptions on the components included in the NPV calculation The Net Present Value (NPV) brings together all of the financial evaluation issues through a discounts payment profile, calculated over 20 years (2011/12 to 2030/31) at 3.5% discount rate with no terminal value. Values are reported relative to the NPV of the do nothing scenario. Option 5 provides the highest NPV due to the most favourable combination of large savings and relatively low capital investment. Options 6 and 7 involve larger upfront investments and/or lower savings and thus provide lower NPVs. Options 1, 2 and 8 create positive NPV of less than 100 million. Options 3 and 4 all have negative NPVs as the investments required are not recovered through the savings generated over the 20 year period. Page 56

57 Figure 14.21: Summary of Net Present Value relative to the do nothing scenario Notes from Figure 14.21: 1. Evaluation scores based on 100m increments between above the do nothing scenario: < 0m (--); 0m- 100m (-); 101m- 200m (+); > 200m (++) Page 57

58 Summary of Value for Money evaluation The evaluation of each of the eight options using all the components of the Value for Money criteria has been summarised in Figure Figure 14.22: Summary of evaluation against Value for Money Options 5 and 6 are the only two options that have an overall positive evaluation, with option 5 being the strongest. The only negative evaluation for Option 5 was given for transition costs, but these will be required under all options. Option 6 received an additional negative evaluation for overall surplus/deficit. Option 5, 6, and 7 were evaluated positively for their Net Present Value. Option 5 has the strongest evaluation in all aspects of this part of the evaluation, with better numbers of viable sites, the highest surpluses and the best Net Present Value calculations. Option 5 provides a good balance of minimal up front investments and high subsequent returns. Page 58

59 14.4 Review of eight potential options against Deliverability Three criteria, Workforce, Expected time to deliver and Co-dependencies with other strategies, contributed towards the evaluation of options under the Deliverability heading. The analysis agreed by local clinicians for these criteria was as follows: Workforce Qualitative assessment based on the number of better performing Trusts expected to be part of the future configuration, as assessed by staff metrics Staff metrics: Overall turnover rate, sickness, staff satisfaction Expected time to deliver Qualitative assessment of ease of delivering option within 3-5 years Number of sites that are already delivering relevant services Additional capacity required Required movements of beds within the system Co-dependencies with other strategies Qualitative assessment of whether each configuration option is in line with other current and developing NHS strategies and provides a flexible platform for the future National initiatives 30 Broader London initiatives (e.g. major trauma and HASU designation) 31 Other cluster initiatives 32 Local strategies in place or in development Workforce In order to successfully implement the changes that local clinicians have proposed, one of the key success factors will be to have the changes delivered by staff that are performing well and are satisfied in their work. Good staff satisfaction is another crucial enabler of clinical quality. The Programme Board recommended that local clinicians use the outputs of the National NHS Staff Survey to understand current levels of staff satisfaction, as shown in Figures and Transparency agenda, Enhancing and improving OOH care, Integrated Care, Driving improvement in acute services, particularly out of hours, National QIPP challenge 31 Integrated cancer systems, Adult emergency services, Paediatrics, including emergency services 32 South West London Programme Better services, better value, Primary Care and Integrated Care 33 Ealing and NWLH merger programme, Individual Trust specific strategies (Imperial service strategy, Royal Brompton site changes, Chelsea & Westminster, West Middlesex and Hillingdon strategies etc.), Inner NWL Integrated Care Pilot (ICP), Mental Health ICP, Pathology modernisation programme, Ongoing work by networks e.g. Cancer network, Cardiac network Page 59

60 Figure 14.23: Workforce data 34 Local clinicians agreed that data on turnover and sickness rates did not provide sufficient basis to determine whether staff at any given site were more satisfied on others. The turnover rates in particular are affected by the presence of staff in training roles, and so teaching hospitals typically have much higher levels of staff turnover compared to the national average. London is also considered to have differing working patterns compared to the majority of the country and its population tends to be more transient than in other areas. The outputs from the National NHS Staff Survey (2011) indicate that Chelsea & Westminster achieves the best scores for two of the three metrics and is the second best placed hospital in NW London for the third metric. West Middlesex has the lowest scores across all the metrics. Excluding West Middlesex, there is little difference in the scores for all the other Trusts in NW London. 34 Notes for Figure 14.23: 1. NHS Information Centre, NHS Hospital & Community Health Service (HCHS) monthly workforce statistics turnover, between Oct 2010 and Oct 2011, % leaving rate. National average estimated based on Sep 2010 to Sep 2011 figures. 1 = unlikely to recommend and 5 = likely to recommend 2. NHS Information Centre (workforce section), July-September 2011 Sickness Absence Rates in the NHS National NHS staff survey National NHS Staff Survey Co-ordination Centre, DH, Staff recommendation of the trust as a place to work or receive treatment, where 1 = unlikely to recommend and 5 = likely to recommend National NHS staff survey National NHS Staff Survey Co-ordination Centre, DH, Staff job satisfaction, where 1 = unsatisfied and 5 = satisfied National NHS staff survey National NHS Staff Survey Co-ordination Centre, DH, Staff feeling satisfied with the quality of work and patient care they are able to deliver, where 1 = unsatisfied and 5 = satisfied Page 60

61 Figure 14.24: Staff survey confidence intervals 35 Only a small proportion of staff from each trust participate in the survey. However, confidence intervals can be calculated that take account of the sample size and distribution of responses. Chelsea & Westminster can be seen to have scores that are statistically better than the scores achieved by other Trusts, whilst scores for West Middlesex are statistically worse. Figure 14.25: Summary of workforce evaluation Local clinicians agreed that the analysis on workforce provided only small differentiating factors between each of the eight options. Options 1, 3, 4, 5, 7 & 8 were given a slightly more positive evaluation compared with options 2 and 6. Options 2 and 6 were the only options that included West Middlesex, which had the lowest set of staff satisfaction scores, but did not include Chelsea & Westminster, which had the highest set of staff satisfaction scores. Options 3 and 7 included Chelsea & Westminster but not West Middlesex and so scored slightly more highly. Options 4 and 8 do not include either Chelsea & Westminster or West Middlesex and so scored positively due to the good overall scores for staff satisfaction compared with the national average. Options 1 and 5 included both Chelsea & Westminster National NHS staff survey Page 61

62 and West Middlesex and so were rated more positively due to the better scores for Chelsea & Westminster Expected time to deliver Local clinicians and programme leaders have agreed that for Shaping a healthier future to be successful it needs to be delivered within the next 3-5 years. This recommendation was made by the Programme Board. Figure 14.26: Ease of delivering each option in 3-5 years 36 Options 3, 4, 7 and 8 have all been given a strongly negative evaluation. In all of these options, at least two Trusts are forecast to be in deficit. It is very difficult for Trusts facing such financial difficulties to make the changes in services as part of reconfiguration. Options 1 and 2 are given a slightly negative evaluation. Both these options include Hammersmith as a major hospital, which would require significant amounts of change to services to make this happen. It is likely that the scale of change to make Hammersmith a major hospital would be difficult to achieve in the 3-5 year timeframe. 36 Notes for Figure 14.26: 1. Includes Adult general and acute, adult daycare, critical care at Major hospitals and Elective Centres (not at Local hospitals). Excludes Pediatric, maternity, rehabilitation and other beds, and neonatal cots 2. Assumption that Elective activity would move based on travel time used to estimate Elective bed capacity requirements 3. Major hospital Critical Care beds numbers include Specialist Critical Care beds (numbers required from ICHT, C&W and WMUH) Page 62

63 Options 5 and 6 are given slightly positive evaluation as neither of these options face the same degree of financial difficulties or the scale of change to services as compared with the options. Both of these options require fewer beds to be moved and less additional capacity. Movement of maternity beds is also an indication of the degree of change required. Options 2, 4, 6 and 8 would need higher numbers of maternity beds to be moved compared to options 1, 3, 5 and 7. Options 1, 3, 5, and 7 all include Chelsea & Westminster which has a large obstetric unit. If Chelsea & Westminster was not designated as a major hospital, these beds would need to be moved elsewhere Co-dependencies with other strategies Local clinicians agreed that their recommendations needed to take account of other work and initiatives going on within the region as well as across London and in surrounding areas. They agreed that our proposals should not go against other changes being made that support improvements in clinical quality and patient experience. Figure 14.27: Evaluation of co-dependencies with other initiatives The key initiatives that the Clinical Board agreed needed to be taken into consideration alongside the proposed reconfiguration were: Changes to the designation of the Major Trauma Centre at St Mary s Current location of stroke units (Ealing is the only site in NW London without a stroke unit) Changes to the location of the HASU at Charing Cross. Local clinicians agreed that options 3 and 4 had the lowest scoring as both of these options would require the Major Trauma Centre at St Mary s to be moved and the lack of stroke unit (and therefore one would need to be moved from another site) at Ealing. Options 5 and 6 scored the most positively compared to the other options as both of these options designate Page 63

64 St Mary s as a major hospital, therefore requiring no change to the Major Trauma Unit (MTU), and have West Middlesex rather than Ealing as a major hospital, therefore there would be no requirement to move a stroke unit. As the Hyper-Acute Stroke Unit consultation designated the location of the unit to Charing Cross, but said that co-location with a MTU could be considered in future, then any potential re-location of the HASU from Charing Cross to St Mary s (where the MTU is located) is possible within the different options and does not differentiate between options 5 and 6 (option 5 including Chelsea & Westminster as a major hospital with option 6 having Charing Cross). The consultation document will need to be explicit about any potential move of the HASU within Imperial. Options 1, 2, 7 and 8 all have a slightly negative rating. Options 1 and 2 would see a change in the MTU from St Mary s as both of these options designate Hammersmith as a major hospital rather than St Mary s. Both options 7 and 8 designate Ealing as a major hospital rather than West Middlesex which would require changes to the stroke units. Clinicians agreed that other initiatives did not sufficiently impact or give cause for differentiation between the options. These initiatives included: National initiatives: Transparency agenda, Enhancing and improving OOH care, Integrated Care, Driving improvement in acute services, particularly out of hours, National QIPP challenge Broader London Initiatives: Primary Care and Integrated Care Local Strategies in place or in development: Inner NWL Integrated Care Pilot (ICP), Mental Health ICP, Pathology modernisation programme, Ongoing work by networks e.g. Cancer network, Cardiac network Review of eight potential options against Research and Education Two criteria, Disruption and Support current and developing research and education delivery, contributed towards the evaluation of options under the Research and Education heading. The analysis agreed by local clinicians for these criteria was as follows: Disruption Disruption to Research and Education Support current and developing research and education delivery Qualitative assessment of whether each configuration option supports current and developing research and education delivery Fit with government R&D strategy Support for Academic Health Science Partnership and Imperial College strategy Alignment with GMC trainee plans Fit with emerging Local Education and Training Boards (LETBs) strategy and plans Disruption to Research and Education The NHS in NW London has a strong reputation for Research and Education and local clinicians agreed that it was important that any proposals would allow for continued opportunities to grow and enable an innovative environment to support further work in this area. Page 64

65 Figure 14.28: Results from 2011 National Training Survey 37 All current 9 acute sites in NW London have scored well in the recent National Training Survey, with Hammersmith and St Mary s scoring particularly well being rated in the top quartile nationally. Local clinicians agreed that in options where Hammersmith was not designated as a Major Hospital, that it would be designated as a Specialist Hospital and would therefore be able to retain its teaching. It was agreed that those options with St Mary s as a Major Hospital and Hammersmith as a Specialist Hospital would be evaluated slightly higher to reflect this retention of teaching. Clinicians did not feel that these scores allowed for any further differentiation between the different options. 37 General Medical Council, National Training Surveys, 2011 Page 65

66 Figure 14.29: Evaluation of disruption to Research and Education 38 Figure sets out the planned levels of spending for 2014/15 that would be located at hospitals that had not been designated as either major hospitals or specialist hospitals. For example, in option 1, where West Middlesex, Hammersmith, Chelsea & Westminster have been designated Major Hospitals (alongside Northwick Park and Hillingdon), and all other sites are either local hospitals or elective hospitals, then 21 million of research spend would be located at the local or elective hospitals and 46 million of education spend would be located at the local or elective hospitals. This is in comparison with option 8, where only 6 million of research spend and 29 million of education spend would be located at the local or elective hospitals. Clinicians agreed that options 1, 2, 3 and 4 should be rated lower than options 5, 6, 7 and 8. Clinicians agreed that it is critical for research to be co-located with clinical delivery to optimise the transfer of treatments from the academic environment to real-life. Additionally in NW London, the majority of research is currently carried out at Hammersmith, St Mary s and Chelsea & Westminster (excluding Northwick Park and specialist hospitals which are the same in all options). All of options 1-4 designate Hammersmith as a Major Hospital and assumes that St Mary s would become a local or specialist hospital, and would therefore lose its research capability. Whereas in options 5-8, Hammersmith is assumed to become a specialist hospital whilst St Mary s is a Major Hospital, leading to the continuation of research at both Hammersmith and St Mary s, therefore minimising disruption. Chelsea & Westminster already carries out research and could do more, but the amount of research work is much smaller compared to the research undertaken at Hammersmith and St Mary s. It therefore has a negligible effect in being able to differentiate between options. 38 Notes for Figure 14.29: 1. ICHT site split provided by the Trust. Research split as 20% Charing Cross; 60% Hammersmith and 20% St Mary s; Education split as 38% Charing Cross; 23% Hammersmith and 39% St Mary s 2. Where Hammersmith is not an Major hospital site, it is assumed to retain all Research and Education as an Specialist centre (Specialist) site Page 66

67 Support current research and education delivery Figure 14.30: Evaluation of support to current research and education delivery 39 Options 1, 2, 3 and 4 have been rated slightly lower than options 5, 6, 7 and 8 for support to current research and education delivery. Imperial has the current strategy to concentrate its research activity onto the sites at Hammersmith and St Mary s. Clinicians agreed that options 5-8 would continue to support this strategy by designating St Mary s as a Major Hospital and Hammersmith as a Specialist hospital, thereby supporting the actions undertaken by Imperial. Clinicians agreed that education was not a differentiating factor as education can move with clinical activity. The ability to develop teaching will be maintained regardless of the option recommended Summary of evaluation The assessment across all five evaluation areas, including their sub-metrics, was brought together onto a single evaluation grid, shown in Figure Notes from Figure 14.30: 1. Hammersmith has ~60% of Imperial College space (39,000m2) and received the majority of major research infrastructure investment over recent years; St Mary s has 17,000m2 of research space and the ability to recruit patients for Imperial College trials through their outpatient clinics Page 67

68 Figure 14.31: Summary of evaluation of eight options Page 68

69 The evaluation was discussed by the Clinical Board, which recommended that options 5 to 8 should go forward for further analysis. The Finance and Business Planning Working Group discussed the affordability criteria and recommended that Option 5 was significantly better than the other options, with options 6 and 7 the next strongest. The Programme Board reviewed the completed evaluation and analysis and considered the recommendations of the Clinical Board and the Finance & Business Planning Working Group. The Board noted the two recommendations and agreed with the assessment that option 5 was the strongest followed by options 6 and 7. The Programme Board agreed that these three options should go forward to the next stage of further analysis with option 5 as the preferred option. The Programme Board agreed that the options 1 to 4 and option 8 were not viable and should not be considered further. The programme also carried out additional analysis on the twelve options that were agreed as not viable in the last step of the production of the list of eight options for further evaluation. As a final consideration, the three options, 5, 6 and 7 that are being put forward for public consultation were tested against a list of 20 potential configuration options (these were 20 options that were developed for potential locations of major hospitals before the choices between Hammersmith and St Mary s, Charing Cross and Chelsea & Westminster and Ealing and West Middlesex were used as a further filter). This final piece of analysis showed that the three options that are going to consultation maintain their strong evaluation, with Option 5 remaining as the preferred option. Page 69

70 15. Confirmation of options This chapter describes the sensitivity analysis that was carried out on the remaining three options to ascertain how robust they are and whether any should not be proposed to be part of the consultation. The sensitivity analysis showed that option 5 remained the preferred option and that all three options are recommended to go forward to public consultation Further analysis for options As a final step in the evaluation, we conducted further analysis on the three options considered being taken to consultation, both to validate the options and check their sensitivity to changes in the assessment. Throughout the evaluation to assess the list of eight options, we used a set of assumptions that have driven the model. We wanted to test the effect of variation in these assumptions (as is expected in normal business). Using this, we can assess the likelihood that the options being considered for consultation would withstand the pressures of expected fluctuations and change. Any further information may be found in Appendix C. Figure 15.1 sets out the different sensitivity analyses that were carried out against the options considered for further analysis 40. Sensitivity analysis was conducted on the penultimate version of the financial models, prior to some final changes. As agreed by the Programme Board, full sensitivities were not re-run because the changes had no impact on the ranking of options and so the sensitivity would not have been affected by this change. The outputs of this analysis are shown in Figure The impact of not achieving the commissioner targets for savings outside acute hospitals has not been assessed as it is outside the scope of the programme. Page 70

71 Figure 15.1: Summary of sensitivity analyses (1/4) Page 71

72 Figure 15.1: Summary of sensitivity analyses (2/4) Page 72

73 Figure 15.1: Summary of sensitivity analyses (3/4) Page 73

74 Figure 15.1: Summary of sensitivity analyses (4/4) Page 74

75 Figure 15.2: Summary of outputs of sensitivity analyses on options undergoing further analysis Page 75

76 Sensitivity analysis for Out of Hospital transformation Several significant financial risks and key sensitivities have been identified for the transformation of Out of Hospital care, as shown in Figure Figure 15.3: Main sensitivities and financial risks relating to Out of Hospital transformation Summary of sensitivity analysis The following summary conclusions may be drawn from the sensitivity analysis: Options 5, 6 and 7 are robust and remain the top 3 options in the 17 sensitivities tested (there are four of the 17 sensitivities where option 7 is equal third) Option 5 comes first in all scenarios (joint first in one scenario) The highest financial risk sensitivities for the providers are: Sensitivity B demand growth greater than planned by Trusts not reimbursed for the additional activity Sensitivity E QIPP under delivers and Trusts not reimbursed for the additional activity Sensitivity N sites that become local hospitals retain legacy fixed costs. These risks need to be very carefully managed in order for any reconfiguration to be successful and for improvements in acute sector finances to be delivered. Under these sensitivities: Option 5 still has a positive NPV relative to do nothing of greater than 170 million, but does not fully resolve the acute sector finances (Sensitivity E results in an overall deficit and all three options leave more sites unviable) Options 6 and 7 result in very low NPVs under scenarios E and N and have overall deficits. The outputs of the 17 sensitivity analyses showed that Option 5 remains the highest evaluated option using the combined Value for Money metric in all sensitivity tests, and the NPV relative to the base case remains million. Out of the three options being Page 76

77 considered for public consultation, the Value for Money evaluation for Option 5 is reduced in five of the tests compared to Option 6, which reduces in 10 tests, and Option 7, which reduces in 12 tests. This indicates that the financial metrics for Option 5 are generally more robust to the sensitivity tests. Assessing viability against a higher bar by increasing the surplus requirement to 2% net surplus would lower the evaluations for Options 5 and 6 but would not change the order of the options overall. Specifically, a 2% net surplus requirement would mean that three sites do not reach the bar for site viability in Option 5 (compared to one site under a 1% net surplus requirement), three sites do not reach the bar under Option 6 (compared to two sites under 1% net surplus), and five sites do not reach the bar under Option 7 (compared to three sites under 1% net surplus). The highest financial risks for commissioners are: Sensitivity A demand growth greater than planned and Trusts are reimbursed Sensitivity D QIPP under delivers and Trusts are reimbursed for additional activity Reprovision costs are higher than expected Risk management analysis The sensitivity analysis has shown that the selection of the options is robust. This section considers the impact of a scenario where a number of downside sensitivities occur simultaneously The four sensitivities with the greatest impact in terms of NPV are: That length of stay reduces by 10% rather than the target of 15% (sensitivity c). This reduces NPV for Option 5 by 46m when compared to the base case That only 60% of QIPP savings are achieved (sensitivity d / e). This would reduce the NPV by 84m for Option 5 compared to the base case That only 50% of the consolidation savings are achieved (sensitivity f). This would reduce the NPV by 60m for Option 5 compared to the base case That only 75% of the savings in fixed costs are achieved (sensitivity n). This would reduce the NPV by 97m for Option 5 compared to the base case. Most other sensitivities have a relatively modest impact on the NPV, in most cases less than 10% from the base case. If all of the top four sensitivities were to occur at the same time then the proposals would deliver a negative NPV of 16m for Option 5 relative to the Base Case. Given this the focus would need to be on the mitigation of the risks to each sensitivity, but especially these top four. This will need to be achieved through a robust project management structure, ensuring that the clinical and overhead efficiencies that are essential drivers to the change are delivered in full by all commissioners and providers. Other issues for which handling strategies are required to be developed before final decision-making include: Trading deficits for Trusts pre and post reconfiguration Transition costs. Page 77

78 15.2 Conclusion Programme Board agreed to recommend to the JCPCT that it consults on options 5, 6 and 7. Option 5 has been designated as the preferred option from a commissioning perspective. Fig 15.4: Confirmation of three options for consultation Page 78