NIHR programme grant: The Design, Development, Commissioning and Evaluation of Patient Focused Vascular Services

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Objective for processing

In providing any clinical service there is tension between maximising efficiency, cost effectiveness and other desirable features of the service. There is currently enormous pressure for the reconfiguration of vascular services due to many conflicting requirements. Vascular services are changing rapidly, having emerged as a new specialty with its own training and specialised techniques. This has resulted in the need for reconfiguration of services to provide adequate specialist provision and accessible and equitable services. An NIHR Programme Grant has been secured by The University of Sheffield and the title of the grant is “The Design, Development, Commissioning and Evaluation of Patient Focused Vascular Services”. The programme grant has four workstreams. The first workstream is Objective 1: Current Service Arrangements (https://scharr.dept.shef.ac.uk/vascular-research/our-research/workstream-1-analysis-of-routinely-collected-nhs-data/). A key element of this first workstream involves the examination of vascular service activity and outcomes using pseudonymised HES data. The data will be used to identify trends and variation in activity and aspects of case mix and outcome that can be established from routinely collected data sources. A five-year £1.86 million NIHR Programme Grant for Applied Research (PGfAR) on “The Design, Development, Commissioning and Evaluation of Patient Focused Vascular Services” commenced in June 2013 and was undertaken to inform redesign of vascular services in England. The grant was due to end in May 2018 but the deadline was extended to May 2019. Workstream 2 is carrying out a thorough evaluation of the various outcome measures that are already available to vascular services in order to assess the quality of the service and outcomes. Where necessary it is developing new methods of evaluating outcome that can be used to assist clinicians and patients in managing their condition and can also be used to evaluate the overall quality of services that are being provided. The overall aim and purpose of this work conducted by the University of Sheffield was to inform the design, development, commissioning and evaluation of vascular services in England. Workstream 3 is focusing on evaluating the strength of patients preferences for various aspects of service that are likely to be affected by re-organisation, including both clinical aspects of the service and non-clinical aspects of the organisation, such as travelling distances and locality. The programme had four workstreams: Workstream 4 will bring together the evidence from the first three workstreams in creating computer models of the ways in which patients with vascular disease are treated, This will allow a prediction of the effects of re-organising services on workload, outcomes and the use of resources. Workstream 1 – The identification of current service arrangements and analysis of Hospital Episode Statistics (HES): This workstream comprised a series of systematic reviews principally examining the relationship between activity and outcome, HES data analysis and use of results from a survey carried out by the Vascular Society of Great Britain and Ireland (VSGBI) to characterise current service arrangements. A supporting document has been provided with further, in-depth information about each workstream. Workstream 2 – The identification and development of outcome measures: The output from this workstream is a prototype electronic, web-based data collection instrument for use with patients with vascular disease (the ePAQ-VAS) to obtain patient-reported outcome measures. The Programme Grant will feed into the process of reconfiguration, initially based upon data analysis and the development of organisational models for care and subsequently through the development of quality indicators that will become available and validated over the course of the research programme. Workstream 3 – The evaluation of non-health service attributes: This workstream evaluated societal preferences for aspects of service other than clinical outcomes that were important to decision making. Trade-off exercises found that people would be prepared to trade expected QALY gain for the less invasive (endovascular) procedures and would expect some additional QALY benefit for increased travel distance following centralisation of services. Quality Adjusted Life-year (QALY) is a measure of the value of health outcomes. As part of the NIHR programme grant work, the initial aim is to characterise existing vascular service arrangements and establish their relationship to workload, case mix and outcomes. This aim will be achieved by analysing routinely collected sources of data, primarily Hospital Episode Statistics (HES) and Civil registration Mortality data. Workstream 4 – The development of vascular service models: This workstream produced a prototype mathematical simulation model which can be used to predict activity, costs, resource use and outcomes for different centralisation scenarios. The main results of the programme grant were published in 2021 (DOI: 10.3310/pgfar09050). However, NIHR widened the scope of their Programme Development Grant (PDG) to include a “Stream B - Post-Programme Grant”. The scope for this stream was defined as “For researchers to develop and enhance the quality and value of an existing or ongoing Programme Grant for Applied Research (PGfAR) award. Applications for such PDGs are only permitted from applicants whose PGfAR contracts are finishing within 12 months or have finished in the past 18 months.” Further funding was secured through Stream B to extend the work by carrying out further secondary analysis of the HES data already obtained for the main programme grant. This aim of the extension work is to add value by investigating in more detail the socioeconomic, geographical and ethnic disparities in utilisation and outcomes of vascular services. The aim of this extended Data Sharing Agreement is being achieved by analysing routinely collected sources of data, primarily the pseudonymised Hospital Episode Statistics (HES) and linked pseudonymised Civil Registration Mortality data. This additional work is an extension of the analytical work previously carried out in the HES data analysis element of Workstream 1 in the main programme grant. An extension to the funding has been agreed by NIHR. Vascular case mix groups being used in this extension work were previously defined using a previous Health Technology Assessment (HTA) study (in 2000). An expert working group of specialists in general practice, vascular surgery, vascular radiology, nursing and other relevant specialties was established for the main programme grant to review the decision-making rules used to analyse the data. The group only had access to the aggregated outputs, with small numbers suppressed in line with the HES Analysis Guide. The case-mix categories (groupings of codes) were therefore previously established with advice from this expert working group and are being used in the current analyses. However, the expert working group no longer exists and no longer has any ongoing involvement in this project. The legal basis for processing is: Article 6(1) (e): processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller, as this research project has been designed and funded with the premise of being in the public interest; and Article 9 (j): processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject, as it is scientific research. The research is expected to inform vascular service design, development, commissioning and evaluation and is expected to be of benefit to all patients with vascular disease in England. The analysis for this extension work is using pseudonymised HES data with linked mortality data supplied by NHS Digital on all patients admitted with vascular conditions from 1998/99 to 2017/18 and is focussing on four main conditions (abdominal aortic aneurysms, peripheral arterial disease, carotid artery disease, varicose veins). Variations in case mix and comorbidity are being examined. Outcomes for which data are available are being examined (e.g. extended hospital stay, readmissions, surgical revision procedures, subsequent cardiovascular events, early and late mortality after aortic aneurysm surgery, diagnosis of stroke after carotid surgery, amputation of a lower limb after arterial surgery). Linked mortality data are the key indicator in relation to longer term outcome. Data analysis is considering: trends in activity; identification of appropriate outcome measures; variations in practice, particularly in relation to new technologies (e.g. endovascular repair of aortic aneurysms, endovascular treatment of peripheral arterial disease, new treatment modalities for varicose veins) and differences in patient selection; travel distances to treatment centres and transfers between hospitals; relationships between service configuration, volume and outcomes; resource use; costs and benefits; and socioeconomic, ethnic and geographical disparities in activity and outcomes. The datasets requested and supplied are needed to comprehensively analyse all the aspects described above. The patient record-level pseudonymised data are needed to examine individual-level effects of surgical procedures, comorbidities, and confounders on outcomes. The research requires data for the number of years provided to include sufficient periods of time before and after reconfiguration and to examine longer term time trends. The geographical spread the data has been supplied for, i.e., England, is needed as the research is examining differences in practice, outcomes, geographical access and area-level socioeconomic deprivation across the country. There are no alternative, less intrusive ways of achieving the aims and purpose of this project. Efforts taken to minimise the data required include: only the use of pseudonymised data; the two stage data extraction to obtain only records relevant to vascular patients; restricting Healthcare Resource Groups (HRGs, which are standard groupings of clinically similar treatments which use comparable levels of healthcare resource) relating to vascular procedures and intracranial procedures except trauma to only 20% to 69% of available fields requested; and restricting Civil Registration (Deaths) data to date of death between 01/04/1998 and 31/03/2018 and patient is present in the HES dataset. The University of Sheffield is the sole Data Controller who also process the data. Sheffield Teaching Hospitals NHS Trust is involved as the lead organisation for the overall project for contractual purposes with NIHR because only NHS bodies and other providers of NHS services in England may apply for NIHR Programme Grant awards, in collaboration with an appropriate academic partner. Sheffield Teaching Hospitals NHS Trust has a formal Collaboration Agreement with the University of Sheffield for the University to carry out the research project. Sheffield Teaching Hospitals NHS Trust does not have any access to, or any control over processing of, the data supplied by NHS Digital. There is currently one employee of Sheffield Teaching Hospitals NHS Trust involved in the project, who contributes by providing comments from a clinical perspective on results of statistical analyses and is involved in project meetings. This individual does not have access to any record-level NHS Digital data and only has access to the data in tabulated format with small numbers suppressed. Similarly, NIHR critically reviewed the grant application and is the funder for the project but does not have any access to, or any control over processing of, the data supplied by NHS Digital.

Processing activities

Over the past two years detailed analysis and modelling protocols have been developed to define case mix groups and outcomes and incorporate these into models of service configuration. The University of Sheffield are currently awaiting NHS Digital data so that these models can be updated with more recent data. This data will be used to identify trends and variation in activity across vascular service provision areas in England. The analysis will focus on four main conditions (abdominal aortic aneurysms, peripheral arterial disease, carotid artery disease, varicose veins). Variations in case mix and comorbidity will be examined. Outcomes for which data are available will be examined (e.g. readmissions, surgical revision procedures, subsequent cardiovascular events). Linked mortality data will be a key indicator in relation to longer term outcome. There is no flow of any data into NHS Digital as part of this project. Under a previous iteration of this Agreement, NHS Digital sent the University of Sheffield HES Admitted Patient Care; HES Critical Care; HES Outpatients; and HES:Civil Registration (Deaths) bridge data. There are no subsequent flows of the data expected for this project. The data flow that has occurred from NHS Digital to the University of Sheffield for this project comprises pseudonymised patient record-level data. Data analysis will consider: trends in activity; identification of appropriate outcome measures; variations in practice, particularly in relation to new technologies (e.g. endovascular repair of aortic aneurysms, endovascular treatment of peripheral arterial disease, new treatment modalities for varicose veins) and differences in patient selection; travel distances to treatment centres and transfers between hospitals; relationships between service configuration, volume and outcomes. The data is being analysed using a combination of descriptive and analytical statistical techniques and mathematical modelling to achieve the aims of the project stated under “Objective for processing”. Vascular case mix groups will be defined using a previous Health Technology Assessment (HTA) study (in 2000). An expert working group of specialists in general practice, vascular surgery, vascular radiology, nursing and other relevant specialties will be established to review the decision-making rules used to analyse the data. The group only have access to the aggregated outputs, with small numbers suppressed in line with the HES Analysis Guide. . It will review and modify the classification of case mix groups to take account of changes in coding and new procedures and investigations. It will also advise on potential quality indicators and outcomes from aggregated outputs. These outputs will be small numbers suppressed in line with the HES Analysis Guide. There is no data linkage to any other individual record level dataset. The only linkage that is being undertaken is at the area-level (lower super-output area (LSOA) level). LSOAs are census areas that were created in the 2001 national census with an average of approximately 1500 people in an LSOA. LSOA-level variables such as the Index of Multiple Deprivation (IMD) and the urban/rural flag (which indicates whether an LSOA is in an urban or rural area) can be attached, using the LSOA code that is already available within the HES dataset. This then allows the dataset to be collapsed (aggregated) to counts of procedures by LSOA so that population-based rates can be calculated using published Office for National Statistics (ONS) mid-year population estimates. By doing this, operation rates in different areas can be compared e.g. rates in urban areas can be compared with rates in rural areas, and rates in deprived areas can be compared with rates in affluent areas. Potential indicators and outcomes include extended hospital stay (criteria specified for individual procedural or diagnostic groups), readmissions and repeated procedures, early and late mortality after aortic aneurysm surgery, diagnosis of stroke after carotid surgery and amputation of a lower limb after arterial surgery. It is believed that the risk of reidentification as a result of the processes being carried out is minimal. There will be no attempt to re-identify individuals. The research requires data over a long time-period, to include periods of time before and after reconfiguration. The data also needs to be for the whole country in order to assess differences in practice and outcomes across the country. Data will only be accessed and processed by substantive employees of the University of Sheffield. The data will be stored on the University of Sheffield’s secure central computer machines managed by the University’s Corporate Information and Computing Services Department (CiCS). A firewall is operated and maintained by CiCS to protect the entire University of Sheffield campus network. The machine room is secure with shutters, access control, intrusion detection and early fire detection with fire suppressant facilities. CiCS carries responsibility for the physical and other security of the networked systems of the University, a requirement of the data handling commitments under data protection legislation. Staff are required to comply with training in data protection and confidentiality before being granted access to the data. Training must be repeated annually, and compliance is closely monitored by the University of Sheffield (via the Information Governance Committee of the School of Health and Related Research). Access is removed from staff who are not compliant. Data will not be accessed or processed by any third parties. Access to the NHS Digital data will be password protected and restricted to researchers working directly on the project. The data is stored and processed only on the University of Sheffield’s secure central computer machines managed by the University’s IT Services Department. A firewall is operated and maintained to protect the entire University of Sheffield network. The machine room is secure with shutters, access control, intrusion detection and early fire detection with fire suppressant facilities. IT Services carries responsibility for the physical and other security of the networked systems of the University, a requirement of the data handling commitments under data protection legislation. Remote access is via a secure University of Sheffield VPN requiring two-factor authentication and through a virtual machine which also requires two-factor authentication. The controls in place ensure that only the staff trained and approved for carrying out the analyses have access to the data. The volume of data is large and processing is carried out on fast encrypted computers housed in University premises and connected to the University’s campus network. The data are stored and processed only on premises owned by the University of Sheffield. Data will only be accessed and processed by substantive employees of the University of Sheffield and will not be accessed or processed by any other third parties not mentioned in this agreement. All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract ie: employees, agents and contractors of the Data Recipient who may have access to that data). There will not be data linkage undertaken with NHS Digital data provided under this agreement that is not already noted in the agreement.

Expected output

The results of the overall project have been, and will continue to be, submitted to NIHR in the form of project reports and presented to committees overseeing national vascular service provision and strategic development. The findings will be disseminated through publication in peer reviewed journals and presentation at national and international conferences. Project reports have been submitted to NIHR annually through the five-year duration of the project (2013-2018) and will continue to be submitted through the extension period. The findings of the overall project have been, and will continue to be, disseminated through publication in peer reviewed journals and presentation at national and international conferences. The benefits will begin to be realised from 2019 onwards. Examples of outputs are described below. The programme (including all work streams) has produced a number of outputs, which are available on the website: Use of complex administrative datasets to monitor clinical activity and outcomes poses challenges. Part of this project work was to develop standardised algorithms with the support of a clinical consensus group to identify all hospital activity for aortic aneurysm, classify the management into clinically meaningful case mix groups and define outcome measures that could be used to compare outcomes among service providers. Robust methods were developed to help examine outcomes associated with previous and current service provision and aid future reconfiguration of aortic aneurysm surgery services. The methodology was published in 2019. Academic research papers: https://scharr.dept.shef.ac.uk/vascular-research/outputs-and-resources/academic-researchpapers/ Hospitals carrying out a higher volume of surgical procedures might be expected to have better outcomes. Investigation of the volume-outcome relationship for abdominal aortic aneurysm surgery confirmed that hospitals with higher annual volumes had significantly lower in-hospital mortality for open surgical repair. The paper reporting this finding was published in 2021. Posters and other media: https://scharr.dept.shef.ac.uk/vascular-research/outputs-and-resources/posters-and-othermedia/ Work that is currently being completed on varicose veins is showing very marked regional variation in access and treatment policies, which does not appear to comply with the NICE guideline for varicose veins (CG 168) that was introduced in 2013. The paper that includes these results is expected to be published in mid-late 2022. Presentations: https://scharr.dept.shef.ac.uk/vascular-research/outputs-and-resources/presentations/ The work on abdominal aortic aneurysms is the largest, most rigorous and most comprehensive examination to date of socioeconomic disparities in aneurysm repair rates, modes of presentation, methods of treatment and long-term survival following surgery. It has revealed clear socioeconomic disparities in all these aspects. The paper presenting these findings is expected to be published in mid-late 2022. The full report from completed work has been published and is publicly available through the NIHR website. The report includes results from all the completed HES analyses and also describes the development of a simulation model. The model has a web-based interface and development was informed by detailed HES analyses as well as other elements of the project. It incorporates disease-specific models for abdominal aortic aneurysm, peripheral arterial disease and carotid artery disease. This predicts the effects of specified reconfigurations on workload, resource use, outcomes and cost-effectiveness. [1 paragraph unchanged]

Expected measurable benefits

The outputs from the programme will be a set of tools of use to providers and commissioners of vascular services. There will be six specific outputs. Current work is focussing on socioeconomic, ethnic and geographical disparities in vascular surgery rates and outcomes in England (and completed work is listed under “Yielded benefits to date”). 1. The outputs from the HES analysis will provide a report of existing service configuration, workload and referral patterns, along with a set of standardised measures for vascular activity and those outcomes that are identifiable from routine datasets. It will also identify recent and planned changes to vascular service configuration. The current work is examining disparities in surgery for abdominal aortic aneurysm, surgery for carotid artery disease, surgery for peripheral artery disease including major amputation as a result of arterial disease, and varicose vein surgery. 2. A systematic review of the relationship of outcome to identified aspects of service configuration will provide an assessment of the existing literature that links outcomes within vascular services to aspects of service configuration. The outputs will be in the public interest because they will quantify the scale of disparities within the publicly funded English NHS. 3. A library of reported utility values for outcomes relating to arterial disease will be published based upon the literature review. These values will be updated when values obtained from the outcome measure assessment in the ePAQ-VAS become available. The results will inform policy makers and decision making organisations. This will in turn be expected to lead to changes in the way services are organised and delivered in order to reduce disparities. The scale of the disparities can be measured and monitored by NHS authorities based on methods used in this project. 4. A set of validated generic and condition specific outcome measures will be developed along with a standardised and validated electronic format for data collection that will be suitable for clinical assessment and service evaluation. Patients with vascular disease will be the main beneficiaries of this project. However, it may take 5-10 years for the yielded benefits of this project in terms of patient care and outcomes to be realised. The results from the work on socioeconomic, ethnic and geographical disparities in vascular services will inform further development of vascular health care to address disparities and contribute to the Government’s levelling-up agenda. 5. A report will identify the attributes, other than those included in calculation of health related quality of life, that are considered important in the provision of vascular services with an estimate of QALY equivalents that can be used to guide future service planning. 6. The final report from the programme will include a set of cost utility models covering the main pathways of care in five key disease areas within vascular disease. The report will include a consideration of the cost utility of a variety of changes in practice that are identified as being potential effects of service reconfiguration, as well as allowing future assessment of the cost utility of other service changes or introductions of new technologies.

Benefits reported

The Researchers are still in the process of analysing the data. The original data was only received two years into the five-year programme grant duration. The analysis is therefore well behind schedule and an extension from NIHR has been granted. A benefit yielded in the short-term from the simulation model that has been produced is the initial exploration which suggests that further reconfiguration of services in England to accomplish high-volume centres would result in improved outcomes, within the bounds of cost-effectiveness usually considered acceptable in the NHS. The Vascular Society of Great Britain and Ireland is the pre-eminent organisation in the country promoting vascular health. The publication from this project on volume-outcome relationships in open and endovascular repair of abdominal aortic aneurysm has contributed to the 2021 revision of the Vascular Society policy document on the “Provision of Services for People with Vascular Disease”. This seminal document was welcomed by commissioners when first published and has become a very important lever to drive change and encourage best practice in the UK and Ireland. This project has contributed to the analysis of clinical benefit, harms, and cost-effectiveness of screening women for abdominal aortic aneurysm, which helped to determine the national abdominal aortic aneurysm screening programme's policy on screening women for abdominal aortic aneurysm. The project is yet to yield other benefits because some of the key elements of work, particularly the work on disparities, are still ongoing. In addition, it can take time to influence national policy, which then has to be implemented before there is an impact on patient care.

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Processing activities

The original processing is continuing. Over the past two years detailed analysis and modelling protocols have been [103 words unchanged] data will be a key indicator in relation to longer term outcome. [10 paragraphs unchanged]

Expected output

[2 paragraphs unchanged] The following paper has been accepted for publication: Aber A, Tong TS; Chilcott J, Thokala P, Maheswaran R, Thomas SM, Nawaz S, Walters SJ, Michaels J. Sex differences in the rates of repair of emergency abdominal aortic aneurysm: A Nationwide Population-Based Cohort Study in England between 2002-2015. British Journal of Surgery 2018 (in press). The following paper will be resubmitted to BMC Health Services Research following revision in line with reviewers’ comments: Aber A, Tong TS; Chilcott J, Maheswaran R, Thomas SM, Nawaz S, Michaels J. Methodology to Identify Aortic Aneurysm Activity in England from Hospital Admissions Data. The Programme Grant will feed into the process of reconfiguration of NHS vascular services in England, initially based upon data analysis and the development of organisational models for care and subsequently through the development of quality indicators that will become available and validated over the course of the research programme. [6 paragraphs unchanged]

Unchanged: Objective for processing, Expected measurable benefits, Benefits reported.