Economic Analyses of Health and Social Care -Evaluation of differences in the performance of health care providers in terms of the amount and cost of provision and in patient outcomes including mortality and self-reported morbidity;

The Centre for Health Economics (CHE), based at University of York, requires access to pseudonymised data to support a number of projects conducted under the Policy Research Unit (PRU) in the Economics of Health and Social Care Systems. For each of the projects described below, CHE staff analyse pseudonymised, record level data from the below datasets:

• Hospital Episode Statistics (HES)

• Emergency Care Data (ECDS)

• Civil Registration (Deaths) Data

• Patient Reported Outcome Measures (PROMS)

• Mental Health Data Sets (Mental Health Minimum Data Set (MHMDS), Mental Health and Learning Disabilities Data Set (MHLDDS), Mental Health Services Data Set (MHSDS))

Following use of the NHS Digital data under any of the projects, only aggregated results with small numbers suppressed as per the HES Analysis guide will be published and disseminated to third parties.

The following projects are conducted by the Centre for Health Economics under the PRU:

Project 1 - Measurement of efficiency, effectiveness, and productivity in the delivery of health care system nationally, sub-nationally and among hospitals.

Project 2 - Evaluation of differences in the performance of health care providers in terms of the amount and cost of provision and in patient outcomes including mortality and self-reported morbidity.

Project 3 - Evaluation of the impacts of health care policy, organisation, finance and delivery of NHS services and quantification of differences in health care utilisation, expenditure, morbidity and mortality over time, across geographic regions, health providers, and among different patient groups.

Project 4 - Investigation of inequalities in access, outcomes, and costs of health services in England.

Project 5 - Evaluation of the interface between the different sectors of the health care system, including the effects of quality and access of primary care on patient use and outcomes in secondary care; and the relationship between long term care, social care and secondary care utilisation.

Most of these projects are funded, at least in part, by the National Institute for Health Research (NIHR), via a major programme of work funded as a Policy Research Unit (PRU) in the Economics of Health and Social Care Systems (http://eshcru.ac.uk/). This programme was re-awarded for a further five years starting from January 2019. The aim of the PRU is to inform and guide policy-making in the health and social care sectors by undertaking high quality, robust and policy-relevant research, based on the discipline of economics, thereby helping to improve the health and well-being of the population, reflecting distributional concerns and population diversity. A detailed work programme for the next phase of programme funding is developed in advance in collaboration with both a Department of Health and Social Care (DHSC) Stakeholder Group and the PRU’s Advisory Group, meetings being held with each Group every six months.

ESHCRU II is one of 15 PRUs and is a collaboration between the Centre for Health Economics (CHE) at the University of York and the Care Policy and Evaluation Centre (CPEC) at the London School of Economics and Political Science (LSE). The ESHCRU II team works with a Patient and Public Involvement (PPI) panel and with 6 PPI Advisory Group members who participate in programme advisory group and workstream advisory group meetings. PPI Advisory Group members have provided insightful contributions at meetings and are invaluable to the discussions with the research team. They have suggested new datasets, identified patient-level factors affecting recovery from surgery, encouraged the team to examine impacts by ethnicity and deprivation and to consider whether and how patient perspectives and outcomes could be incorporated into payment systems. The PPI panel has reviewed the ESHCRU II website (https://eshcru.com/). Six PPI panel members submitted detailed reviews. In response to the feedback received, the layout, language and content were refined to appeal to a broad audience. Wider benefits of the PPI Panel include identifying an individual with relevant experience for a separate York proposal on health inequalities and putting the Department of Health and Social Care (DHSC) in contact with two Panel members to speak at a networking/engagement event for the NIHR call Artificial Intelligence for Multiple Long-Term Conditions (AIM) (https://www.nihr.ac.uk/documents/nihr-artificial-intelligence-for-multiple-long-term-conditions-aim-clusters-call-research-specification-finalised/24646).

When it comes to individual projects in the PRU, either the University of York or the LSE take the lead. The LSE have a separate Data Sharing Agreement with NHS Digital (DARS-NIC-354497-V2J9P). The LSE has no involvement in the research projects carried out by CHE, and LSE will not have any access to NHS Digital data that the University of York holds for ESHCRU. Similarly, if LSE leads a project that requires NHS Digital data access, LSE will be responsible for all aspects related to the data, without any sharing of that responsibility with the University of York and without the University of York having any access to or involvement in their use of NHS Digital data. If an ESHCRU project is led by the University of York and requires the researchers to access NHS Digital datasets, the University of York will be solely responsible for determining the purposes and means of the processing of personal data.

The University of York is the sole Data Controller who also processes the NHS Digital data held under this Agreement to be processed for the purposes outlined in this section. The University of York is solely responsible for the data provided under this Agreement and for all decision-making related to the data, its analysis, and any outputs. The NIHR / DHSC / NHS England are funders of the University of York's research and as such are not directly involved nor responsible for decision-making related to the data, their analysis, and outputs.

The University of York was founded "for the advancement of learning and knowledge by teaching and research", as established by the University of York Charter. The University of York is a public authority under the Freedom of Information Act 2000. The University's Strategy 2030 builds on the founding principles of the Charter, with the overriding aim to be a university 'for public good'.

In line with the University of York's charter which states that the University of York advance learning and knowledge by teaching and research, the University of York (CHE) has considered the legal basis for processing data under General Data Protection Regulation (GDPR) in respect of the projects covered in this agreement. CHE processes data under the following legal bases:

Article 6(1)(e) of the GDPR (lawfulness of processing): 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

Article 9(2)(j) of the GDPR (Processing of special categories of personal data): processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes

The processing of personal data, including special category data, is necessary to carry out research that serves the public interest, to inform policy and practice with the goal of improving health and well-being.

The research undertaken using NHS Digital data informs health and care policy and practice by identifying the effectiveness, efficiency, distribution, and quality of a wide range of services provided to the population. It produces insights that allow the maximisation of health gain from limited healthcare budgets, along with information on how health and health care is/can be distributed equally to meet the health needs of varying demographics. NHS Digital data provides a view of health care utilisation for CHE to understand how effective delivery of care is distributed both nationally and locally, contributing to the delivery of new healthcare policy aimed at improving the quality of care.

Approximately 20% of funding is reserved for the PRU to respond to short-term responsive requests for research. This process ensures that the work programme can be shaped to reflect enduring and emerging policy concerns. For some projects, CHE have been able to secure additional funding, enabling the University to extend or deepen the analyses of the research topic. NIHR also funds other projects, as mentioned below.

The University of York process both the pseudo-consultant code and the underlying clear text consultant code. Some research projects will analyse variation in case-mix adjusted hospital inpatient activity and their related outcomes at the consultant level. For these projects, it is necessary to assign activity to consultants but it is not necessary to identify who they are. The pseudo-consultant code is therefore sufficient for these projects. Other research projects, such as the European Training Network “Improving Quality of Care in Europe (IQCE)”, Horizon 2020 Marie Sklodowska-Curie Innovative Training Network, will need access to the underlying clear text consultant code field to link to it other publicly available data on consultants, such as the General Medical Council (GMC) medical register (https://www.gmc-uk.org/registration-and-licensing/the-medical-register). The GMC medical register provides details on, for example, the type of registration held by a doctor and the date at which they completed their primary medical qualification. This linkage is achieved through the clear text consultant code field, which corresponds to doctors’ GMC numbers. The pseudo-consultant code is insufficient for these projects. The underlying clear text consultant code and the pseudo-consultant code will be stored as two separate files and stored in two separate folders. The files can be linked to the HES Admitted Patient Care (APC) dataset through the epikey variable. Access to these files will be managed on a researcher and project-by-project basis. In this way, the University of York will ensure that no researcher has access to both data fields at the same time for the same project, thereby eliminating the risk of re-identification.

The University will not publish any information at identifiable consultant level under any of the work projects described in this agreement.

All of the work involves analysing the data in different ways. For example, an analysis under project 1 may focus on particular specialities, comparison of productivity across hospitals, or may be a broader assessment of national productivity. Many of the statistical methods to be employed require longitudinal data to investigate how changes in patient outcomes (including morbidity, mortality, emergency re-admissions, length of stay, admissions for conditions that could be managed in primary care, inpatient admission rates after Accident & Emergency (A&E) attendance) are related to changes in policy (including payment policies and incentives), changes in market configurations, changes in organisational structure, and changes in patient characteristics. Pseudonymised patient level information is required to allow for the influence of past utilisation, for demographic factors, for socio-economic factors (e.g. deprivation) linked to the small area in which patients live, and patient distance from hospitals, social care providers, and general practices. It is also essential in investigating the equity implications of policies, market structure, and organisational arrangements. The Principal Investigators and Project Leads are responsible for determining what analyses will be undertaken and what data will be used for each analysis in support of the objectives agreed with the funding organisations.

Different data dissemination frequencies are required under different projects. Under DARS-NIC-84254-J2G1Q-v3, the University of York requested the addition of monthly data disseminations under certain projects, due to the coronavirus pandemic precipitating the DHSC's need for more timely evidence to inform recovery from the initial crisis. Under DARS-NIC-84254-J2G1Q-v4, the University of York have revised the monthly disseminations to quarterly disseminations.

CHE confirms that the data under this Agreement would only be used for the five projects listed, and any additional project (whether as part of the DHSC programme or otherwise) would require a separate approval. Equally individuals working on each project will only be permitted to access the data relating to that project, as identified within this Agreement. Access is granted for each project only to the named individuals associated with that project under authorised user names. Such access is password controlled (with a password reset required on a regular refresh).

The controls enable a single copy of the data to be held, reducing security risk associated with multiple copies being provided per project.

Project 1 - Measurement of efficiency, effectiveness and productivity in the delivery of health care system nationally, sub-nationally and among hospitals;

The purpose of this project is to produce information for the Department of Health and Social Care (DHSC) and Secretary of State for Health on efficiency, effectiveness and productivity. In the current economic climate it is particularly important that changes in efficiency and productivity can be identified and monitored. This helps ensure accountability to the public for how the annual NHS budget is spent and to identify opportunities for better use of resources devoted to the NHS. This project provides numerical answers and context for, among others, House of Commons Health Committee, the Public Accounts Committee, Public Expenditure Inquiries, and DHSC submissions in support of annual Spending Reviews. The work also contributes to the measurement of productivity of the health service in the national accounts, compiled by the Office of National Statistics.

Funders:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems.

• NHS Productivity (National Assessment Project) (Ref NIHR 200687).

This project will use only the following data supplied under this Agreement: HES APC 1998/99-2021/22 A&E 2007/08 - 2020/21; Critical Care 2011/12 – 2021/22; Outpatient 2011/12-2021/22; Patient Reported Outcome Measures (PROMs) 2009/10 –2021/22; Civil Registration (Mortality) data 1998/99 - 2021/22; Emergency Care Dataset (ECDS) 2017/18 – 2021/22; Mental Health Services Data Set 2016/17 - 2020/21

Project 1 will only process data provided through Annual dissemination for the datasets listed above.

Data has been assessed and minimised to only utilise the required data from that disseminated under this Agreement for this project and cannot be minimised further.

The University of York will estimate survival models that relate the timing of death (based on the exact date of death) for individual patients to the relevant unit of assessment (provider, region) after adjusting for relevant case-mix differences across the populations under study.

Most of the work undertaken under this project involves measurement of productivity over time, hence the need to hold the data from 1998/99. It is also necessary to construct aggregated measures of NHS output and quality based on what has happened to each individual patient in whatever setting care has been delivered, hence the need for patient-level information. The project also requires use of the sensitive PROMs data as measures of the quality of health care.

Project 2 - Evaluation of differences in the performance of health care providers in terms of the amount and cost of provision and in patient outcomes including mortality and self-reported morbidity;

The purpose of this project is to produce information for National and local decision makers, such as the Department of Health and Social Care (DHSC), commissioning bodies and Local Authorities (LAs), to assist decisions regarding the provision of services that offer the greatest value for money according to the benefits achieved. Delivering appropriate, high quality, health care services to patients, in the most cost-effective way, are important priorities in any health care system. Advancing these priorities requires the analyses of such things as variations in practice and of the relationship between patient outcomes and hospital and consultant workload; which dimensions of performance are most important to patients; and the extent to which financial incentives motivate best practice. Ultimately this project informs the assessment of the most efficient and cost-effective way of delivering a particular service. This helps ensure accountability to the public for how the annual NHS budget is spent and helps to identify opportunities for better use of resources devoted to the NHS. The project is designed to develop a more systematic evidence base that will allow policy-makers, providers and commissioners to develop policies to achieve efficiency targets and outcome-based commissioning, publish information on performance in formats that are most useful for the intended stakeholders, and to redeploy resources to produce more efficient mixes of services both within and across the health and social care sectors.

Funders:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems.

• NIHR Applied Research Collaboration Yorkshire and Humber ARC (Ref NIHR 200166)

• NHS England (Ref R1790901) Economic evaluation of the Fragility Hip Fracture Best Practice Tariff.

• NIHR Health Services and Delivery Research (HS&DR) Programme (Ref DRF-2016-09-097): Doctoral Research Fellowship - "Providers' response on the Pay for Performance incentives".

• European Commission, European Training Network. “Improving Quality of Care in Europe (IQCE)”, Horizon 2020 Marie Skłodowska-Curie Innovative Training Networks.

The work for all these funders will require the sensitive PROMs data to measure patient outcomes.

The project will use only the following data: HES APC 1989/90 – 2021/22 Sensitive field: Consultant Code; HES Outpatient 2002/03 – 2021/22; PROMs 2009/10 – 2021/22; Civil Registration (Deaths) 1998/99 - 2021/22.

Project 2 will utilise both Annual and Quarterly disseminations of the datasets indicated above.

Project 3 - Evaluation of the impacts of health care policy, organisation, finance and delivery of NHS services and quantification of differences in health care utilisation, expenditure, morbidity and mortality over time, across geographic regions, health providers, and among different patient groups.

The purpose of this project is to produce evidence to inform NHS England and the Department of Health and Social Care’s decisions on resource allocation, funding models, and the design and direction of future policy regarding the health, mental health, and social care sectors, with CHE’s advice and analyses being regularly sought to feed into White papers and specific government reviews.

This project includes understanding which types of budgeting, organisation, structure and contracting arrangements for health, incl. mental health, and social care services best achieves strategic goals. It also includes evaluations of payment policies (including financial incentive schemes) and changes to the organisation of services (e.g. co-location of general practitioners alongside emergency departments, mergers of providers, vertical integration of providers, care pathways) that seek to encourage good quality, cost-effective care and/or facilitate access to timely care. The main aims are to: analyse the potential for use of different organisational structures and payment mechanisms in health and social care to improve overall performance; analyse the impact that different payment policies and service configurations can have on prices, outputs, quality and outcomes; explore how the best payment systems and service configurations could be implemented in practice; and establish the effect of innovative organisational forms on costs and quality of care.

Funders:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems.

• NIHR Health Services & Delivery Research (HS&DR) 10/1011/22 and NIHR HS&DR 13/54/40: Relationships between quality of primary care and secondary care outcomes for people with mental illness.

• Wellcome Trust [ref: 105624] through the Centre for Chronic Diseases and Disorders (C2D2) at the University of York: Finance and organisation of mental health services.

• Health Foundation [ref: 57151] Efficiency, cost and quality of mental healthcare provision.

• NIHR HS&DR (Ref DRF/2014-07-055): Doctoral Research Fellowship - Measuring & explaining variations in general practice performance.

• NIHR HS&DR (Ref 15/145/06): General Practitioners and Emergency Departments (GPED): Efficient Models of Care.

The project will use only the following data: HES APC 1998/99 – 2021/22; A&E 2007/08 – 2020/21; Outpatient 2002/03 – 2021/22; PROMs 2009/10 – 2021/22; Emergency Care Dataset 2017/18 – 2021/22; MHMDS 2011/12 – 2013/14; MHLDS 2014/15 – 2015/16; MHSDS 2016/17-2020/21; Civil Registration (Deaths) 1998/99-2021/22; HES APC Sensitive Psychiatric Fields: Detention category (DETNCAT), Legal group of patient (psychiatric) (LEGALGPC), Legal status classification (LEGLSTAT).

Project 3 will utilise both Annual and Quarterly disseminations of the datasets listed above.

The work for several funders will require the use of the PROMs data to measure morbidity over time.

This project will also require use of MHMDS/MHLDS/MHSDS data linked to HES data in order to carry out analyses into the economics around mental health and mental health care provision. CHE is requesting sensitive MHMDS/MHLDS/MHSDS fields and sensitive HES psychiatric fields (Legal group of patient, Legal status classification, and Detention category). These relate to the legal category / legal status of the patient which is an important indicator of patient severity. CHE will need these sensitive data items to accurately control for the impact of detention on resource use and utilisation. CHE needs to check data consistency between HES and the MHMDS/MHLDS/MHSDS and therefore requires sensitive data on legal status in both datasets.

Project 4 - Investigation of inequalities in access, outcomes and costs of health services in England.

The purpose of this project is to produce information that NHS England and commissioning bodies will use to address the NHS’ duty under the Health and Social Care Act 2012 to consider reducing health inequalities, and that Public Health England, Local Authorities and a variety of other public and third sector organisations will use to inform decision making and quality assurance around health and social care and wider public policies with impacts on health. CHE has developed new methods of local health equity monitoring for health care quality assurance, which NHS England adopted in 2016. In collaboration with colleagues at the Department of Health Sciences, University of York, and analysts at NHS England, CHE will refine and use these methods and related measures to monitor the progress of national and local NHS organisations in reducing inequalities in healthcare access and outcomes, to gain insight into the determinants of inequalities and which local areas show sustained improvements and deteriorations in health inequality and why, and to evaluate the equity impacts of local new models of care and other health policies. The work will also assist the ONS to conduct distributional analyses of NHS spending for use in constructing statistics about in-kind social transfers.

The work for the Department of Health and Social Care will investigate why providers respond differently to policy incentives and quantify the associated impact on inequalities in care quality and/or access, including how these change over time.

Funders:

• NIHR Training Co-ordinating Centre (TCC) (Ref SRF-2013-06-015) Health equity impacts: evaluating the impacts of organisations and interventions on social inequalities in health.

• Wellcome Trust. ‘Re-Engineering Health Policy Research for Fairer Decisions and Better Health’.

• Department of Health to the Policy Research Unit in the Economics of Health and Social Care Systems

The project will use only the following data: HES APC 1989/90 – 2021/22; A&E 2007/08 – 2020/21; Outpatient 2002/03 – 2021/22; Critical Care 2011/12 – 2021/22; PROMs 2009/10 – 2021/22; Civil Registration (Deaths) 1998/99-2021/22; Emergency Care Dataset 2017/18 – 2021/22.

Project 4 will utilise both Annual and Quarterly disseminations of the datasets listed above.

The work requires the use of sensitive PROMs data to measure patient outcomes in secondary care.

The following is a separately funded piece of research work which fits within the scope of project 4 (DARS-NIC-84254-J2G1Q-v3):

Funder:

• NIHR Policy Research Programme (grant number PR-X06-1014-22005), Partnership for Responsive Policy Analysis and Research (PREPARE), a collaboration between the University of York and the King's Fund.

Though PREPARE is a collaboration between the University of York and the King’s Fund, the King’s Fund has no involvement in this research project. The research request was considered by the co-leads in the University of York and the King’s Fund who agreed that the University of York would carry out the research. The lead within the University of York has been solely responsible for all decisions on how this research will be carried out including all decisions in respect of what data processing is required.

Description of additional use of data:

This area of the project will explore the links between child health and child poverty, in particular the NHS hospital utilisation of children born into deprivation (using the indices of deprivation (ID) as a proxy for poverty) in comparison with children who are not born in deprived areas. The study will also explore whether any difference in hospital utilisation over the early life course has changed over time. To do this the project will create a birth cohort of children born in NHS hospitals in England in specific financial year (2000, 2005, 2010, 2015, 2018), and track their use of NHS services (inpatient, outpatient and A&E) over their life course (up to age 18 for those born in 2000). The analysts will then test whether the age-sex adjusted differential use of hospital services across children born into rich and poor neighbourhoods has changed over time.

The work will use the following data:

HES APC 2000/01 – 2020/21; HES A&E 2007/08 - 2018/19; HES OP 2002/03 – 2020/21; Emergency Care Dataset 2017/18 – 2020/21.

The work will utilise both Annual and Monthly disseminations of the datasets listed above.

Linkages:

There is a need to link the HES APC data to Index of Multiple Deprivation (IMD) via the child’s Lower Super Output Area (LSOA) of residence. An older version of the IMD (2004) is already included in HES but the study will have to link to subsequent versions of the ID.

This Agreement permits the use of the data for this research but does not permit the use of the data for other research projects funded by PREPARE. Use of the data for any other projects funded by PREPARE would require a formal amendment to this Data Sharing Agreement.

*** NEW FOR VERSION 4 - The following is a separately funded piece of research work which fits within the scope of project 4:

Funder:

NIHR Policy Research Programme (grant number NIHR200417), Evidence to support efficient and effective reduction of health inequality. The duration of the project is: 1st June 2019 to 31st May 2022.

Description of additional use of data:

This project will investigate the extent to which the introduction of the Best Practice Tariff (BPT) for fragility hip fracture in English NHS hospitals in April 2010, and subsequent changes to the tariff design, have affected health inequalities in this patient population.

In previous work conducted under this Data Sharing Agreement (DSA) and funded by NHS England (see Project 2: NHS England (Ref R1790901) Economic evaluation of the Fragility Hip Fracture Best Practice Tariff), CHE researchers quantified how the introduction of the BPT changed care delivery on eight incentivised process quality standards, and linked these changes to long-term health outcomes and healthcare resource use via decision-analytical modelling techniques. However, this study did not test whether the benefits and opportunity costs of the BPT are equally distributed across socioeconomic groups, and, consequently, how the BPT introduction affected health inequalities across the population of England. This new study will build on the earlier research, using data from the National Hip Fracture Database (NHFD) linked to HES via a bridging file, and will examine whether the BPT introduction and subsequent amendments to its design had different effects on incentivised clinical behaviours across socioeconomic groups. The University of York have a DSA with the NHFD (ref: FFFAP/DSA/2020/004), and a separate DSA with NHS Digital (DARS-NIC-50329-G1L1P) which details the bridging file to be produced, enabling linkage of the HES data held under DARS-NIC-84254-J2G1Q and the NHFD data held under FFFAP/DSA/2020/004. All DSAs will be amended to include the above specified use of the data.

The work will use the following data:

HES APC, Outpatient, A&E/ECDS, Civil Registration (Deaths) Secondary Care Cut: 2011/12 - 2019/20 (period of bridge file: 01 April 2011 to 31st March 2020)

Linkages:

There is a need to link the HES data to data from the National Hip Fracture Database (NHFD; part of the Falls and Fragility Fracture Audit Programme). Access to NHFD is granted under an existing DSA with the Royal College of Physicians (FFFAP/DSA/2020/004). DARS-NIC-50329-G1L1P-v4 is the DSA between the University of York and NHS Digital under which NHS Digital will create a bespoke bridging file to enable the University of York to securely link NHFD data to HES data. ***

Project 5 - Evaluation of the interface between the different sectors of the health care system, including the effects of quality and access of primary care on patient use and outcomes in secondary care; and the relationship between long term care, social care and secondary care utilisation.

It has long been understood that health and social care services frequently provide treatment and care for the same individuals, so ensuring that these are ‘joined up’ or well co-ordinated has been an important and long-standing policy objective. In practice, however, both the services and approaches to monitoring these have developed separately, with potential implications for the efficiency and effectiveness of both health and social care. The purpose of this project is to produce evidence that will be used by the Department of Health and Social Care and commissioners to inform discharge arrangements and the design of integrated care arrangements and to identify opportunities for substitution of different types of health and social care services. CHE has also developed an online web tool to inform patients about their likely outcome of surgery to impact on shared decision making in primary care in York.

Funder:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems

The project will use only the following data: HES APC 1989/90– 2021/22; A&E 2007/08 – 2020/21; Emergency Care Dataset 2017/18 – 2021/22; MHMDS 2011/12 – 2013/14; MHLDS 2014/15 – 2015/16; MHSDS 2016/17 - 2020/21; Outpatient 2002/03 – 2021/22; Critical Care 2011/12 – 2021/22, PROMs 2009/10 – 2021/22; Civil Registration (Deaths) 1998/99-2021/22.

Project 5 will utilise both Annual and Quarterly disseminations of the datasets listed above.

This project requires the sensitive PROMs data to measure patient outcomes in secondary care.

The outputs from all of the projects are intended to include:

• Reports and slide decks to funding bodies;

• peer reviewed papers in academic journals;

• lay summaries such as newsletters and blogs;

• conference and seminar presentations to a variety of audiences, such as academic, policy, professional and public audiences.

The aim is to produce reports containing aggregate results that show trends over time, differences across providers, commissioners, geographical areas and by patient subgroups and patient characteristics. The results are intended to contain estimated correlations showing associations between patient outcomes and patient characteristics, hospital, institutional, geographic and environmental factors. Statistical results are intended to be presented in interactive spreadsheets or “Dashboards” (e.g. similar to http://health-inequalities.blogspot.co.uk/ which uses Quality Outcomes Framework (QOF) data and only contains aggregated data which can be interrogated), tables and maps of aggregate statistics summarising patient characteristics

All outputs will contain only data that are aggregated with small numbers suppressed in line with the HES Analysis Guide.

The dissemination and communication strategy will vary between projects and will have been previously agreed with the funders. It would normally include the oral presentations; interim and final reports; and published reports and journal articles. Examples are provided below:

Oral presentations / knowledge exchange

• Presentation of interim and emerging findings to the various project study advisory groups and/or steering committees. Members - who typically include policy makers, clinicians, academics and public contributors - provide feedback and advice.

• Interactive workshops with policy analysts from DHSC and NHSE/I to discuss emerging findings and ensure policy relevance

• Presentations to commissioning bodies, NHS trusts, and PPI groups

• Open lectures and invited talks at Universities/Research Centres both in the UK and abroad

• Oral or poster presentations at national and international conferences, such as Health Economists’ Study Group, International Health Economics Association (iHEA), and European Health Economics Association (EuHEA). Delegates may include international organisations such as The World Bank, Organisation for Economic Co-operation and Development (OECD) and World Health Organisation (WHO), alongside members of the international academic community.

• End of project workshops or conferences to present research findings to key stakeholders and policy makers

Unpublished reports

• Draft reports with preliminary findings to advisory groups

• Interim reports for funders and policy analysts

• Draft final reports for funders. These are usually peer reviewed externally by academics and internally by policy analysts

Publications

• Published reports containing full, detailed findings, with an accompanying lay summary to make key messages more accessible.

• Press releases to accompany the publications of reports (full or short), via the University of York Press Office as well as through the CHE website and social media platforms, as well as funders own Press Release Offices and social media platforms;

• Peer reviewed scientific papers in academic and policy journals

• Short articles in CHE annual reports and CHE newsletters

Target dates for outputs

The outputs from each project will be delivered in accordance with CHE’s funding contracts, which run to different timelines with various milestones for each. Below are examples of key milestones and timelines for some of the projects undertaken.

Project 1 - The primary output from this ongoing project is the production each year of an annual update to national NHS productivity figures that incorporates the most recent financial year of data. Under this project, CHE has demonstrated that NHS productivity growth is meeting the requirements of the Five Year Forward View and outpaces that of the economy as a whole. CHE’s figures are widely used to inform policy discourse, with the DHSC relying on the information for internal monitoring purposes and for external reporting and response purposes, such as to inform annual Spending Reviews. Under this project, CHE also provides data about the quality of NHS care to the Office of National Statistics that are used in the construction of the national accounts.

In addition to the annual update of national figures, CHE also undertakes analyses of variation in hospital productivity and produces short reports, memorandum or slide decks for the DHSC to address specific questions about NHS productivity. CHE presents the work regularly to various audiences, including politicians, policy makers, academics, health professions and the general public through seminars, conference presentations and media appearances.

Project 3 – Policy makers need timely evidence on the impact of health care policies. These policies concern both the supply side of health care - the organisation, finance and delivery of services - and the demand side, such as utilisation, morbidity and mortality. Under this project, analyses may investigate variations over time and/or across geographic regions, providers, or patient groups. Empirical research frequently relies on analyses of HES. Outputs include open access reports and scientific papers in academic journals. Interim and final reports are submitted to funders as part of our contractual requirements. Emerging findings are presented and discussed at regular advisory and stakeholder group meetings, as well as at invited seminars and workshops.

*** NEW TO VERSION 4 - Project 4 - NIHR Policy Research Programme (grant number NIHR200417), Evidence to support efficient and effective reduction of health inequality - planned outputs include: scientific report to the funder (target date 31/05/2022); two article in scientific journals (target date 01/03/2023); in addition, we plan to disseminate the findings from the work through lay summaries such as newsletters and blogs, and conference and seminar presentations to academic, policy, professional and public audiences. ***

Project 5 – The focus of this project is the evaluation of the interface between the different sectors of the health care system, such as the effects of quality and access of primary care on patient use and outcomes in secondary care; and the relationship between long-term care, social care and secondary care utilisation. Published outputs from this project include open access reports and scientific papers in academic journals. Interim and final reports to funders will be provided in line with contractual requirements. Emerging findings will be presented at conferences, advisory and stakeholder group meetings, or at DHSC workshops or seminars. This ongoing engagement helps ensure the work remains policy relevant. An example is a departmental working paper examining the effects of changes in social care funding on health care utilisation (Pace et al. Understanding the interdependencies between health and social care resources and arrangements. CHE Research Paper, December 2020). Early findings from this work were shared with NHSE (May 2018), discussed with DHSC (March 2019), with further analyses undertaken in response to their questions about the impact of the Better Care Fund. The report was revised following written comments from the Department (May 2019), and published in the Health Economics journal (Liu et al 2020, https://doi.org/10.1002/hec.4175).

The DSA includes as funders the European Commission, European Training Network. “Improving Quality of Care in Europe (IQCE)”, Horizon 2020 Marie Skłodowska-Curie Innovative Training Networks, which funds two PhD research projects. The outputs for this specific project is the compilation and publication of a doctorate thesis and related research articles to be published in international research journal. Both PhD students have now completed their PhD thesis, however access to the data is still required to finalise outputs and respond to peer review.

The Centre for Health Economics has a long-established track record in the delivery of policy research that utilises HES data, as recognized by the award of the Queens Anniversary Prize in 2007. Examples of recent publications arising from the above projects that have employed the HES data can be found at the links below:

https://eshcru.com/publications/

http://www.york.ac.uk/che/publications/in-house/ https://www.york.ac.uk/che/publications/all/

Project 1 - The NHS spends over £120 billion a year and the budget provided for the NHS is influenced by the evidence on how well the money is spent. The annual estimates of productivity that we produce for the Department of Health and Social Care are a key part of their negotiations with the Treasury about the size of the budget and the research is cited in exchanges with ministers and government committees that scrutinise NHS spending. In particular, the quality adjustment method devised by York is important in ensuring that productivity is not under-stated which may have a negative impact on the size of budget settlements and hence lead to lower investment in the NHS. The most recent update to the NHS productivity measure (https://www.york.ac.uk/media/che/documents/papers/researchpapers/CHERP171_NHS_productivity_update2017_18.pdf) shows a steady positive increase and that NHS productivity has increased substantially faster than the overall economy.

The methods developed by the York research team have also caught interests of Governments and health ministries in other parts of the world (eg Japan, Italy, Malaysia, Sweden), which have resulted in a number of workshop and collaborative work.

The primary output from this ongoing project is the production each year of an annual update to national NHS productivity figures that incorporates the most recent financial year of data. In 2017, the annual update was produced for the DHSC and, as in other years, was used by them externally and internally in monitoring, informing policy debate, the annual spending review and negotiations on budget setting. Under this project, CHE also provided data about the quality of NHS care to the Office of National Statistics that are used each year in the construction of the national accounts. In 2017, additional analyses on hospital level productivity were produced for the DHSC, examining the factors underlying variation in productivity, which assists the DHSC in exploring how to get the best value from NHS resources.

Project 2 - Examples of outputs produced from this project include journal articles on the determinants of health care expenditure. For example, an article by Howdon and Rice (Health care expenditures, age, proximity to death and morbidity: Implications for an ageing population, 2017, https://doi.org/10.1016/j.jhealeco.2017.11.001) investigates the implications of an ageing population on health care expenditure, showing that the latter is primarily determined by proximity to death, rather than age. Chalkley et al. (Elective hospital admissions: secondary data analysis and modelling with an emphasis on policies to moderate growth, 2017, https://doi.org/10.3310/hsdr05070) investigating growth in elective hospital admissions in England find that these are mainly due to increases in both demand for and supply of hospital care, with health system reforms and ageing population less important.

This project has produced a range of evidence that allows the DHSC and other organisations to understand and plan expenditure on different aspects of NHS care, to account for changes in activity and also to understand aspects of quality of care, First, it has investigated the drivers of health care expenditure, disentangling the influence of age, morbidity and proximity to death on the level of expenditure. This has important implications for the DHSC in terms of planning and for the design of the resource allocation formulae used to distribute the healthcare budget, Second, it has explored several aspects of activity in the secondary care sector, for instance, policies for moderating growth in elective admissions and how the provision of specialist rehabilitation services affect the duration and costs of hospital care. It has also looked at the use of patient valuation of the care they receive as a way of measuring the quality of hospital care. Third, it has investigated in-hospital mortality trends over time. These insights allow policy makers to appreciate the costs and the benefits of NHS provision and to plan for more effective and efficient care.

Project 3 - Outputs from this project include journal articles on payment systems, providing evidence on intended and unintended impacts for policy makers designing new approaches to paying for care. For example, one study examined the effects of two incentive schemes in primary care that aimed to tackle underdiagnosis in dementia, with HES data used to control for a parallel hospital scheme. The schemes were effective in increasing dementia diagnoses (Mason, et al 2018, Investigating the impact of primary care payments on underdiagnosis in dementia: a difference-in-differences analysis, IJGP. https://doi.org/10.1002/gps.4897). However, patient satisfaction was lower in practices that achieved higher rates of dementia diagnoses (Liu, et al 2019, Incentive schemes to increase dementia diagnoses in primary care in England: a retrospective cohort study of unintended consequences. BJGP https://doi.org/10.3399/bjgp19X701513).

The research in this project has evaluated whether the way in which the NHS is organised can affect the costs and outcomes of services to help inform policy decisions. In 2017, the research has produced evidence on a number of issues in mental health, including the relationship between costs and quality and socio-economic inequalities in access to care. Other examples include the production of evidence on the impact of the electronic booking system on referrals and non-attendances and the investigation of reasons for delayed discharges from hospital, both of which are important in terms of planning future organisation of care.

Project 4 - Access to health services in England is generally lower for disadvantaged groups. This is a key policy concern because in the NHS access is ideally based on need. One study that used HES data to investigate inequalities in access to health services in England is by Moscelli, et al 2017 (Socioeconomic Inequality of Access to Healthcare: Does Choice Explain the Gradient? JHE https://doi.org/10.1016/j.jhealeco.2017.06.005). The authors investigated the extent of waiting time inequalities for elective coronary revascularisation procedures by socioeconomic status within hospitals. Using sophisticated statistical techniques, they showed that patients from the most deprived areas waited around one-third longer than the least deprived groups. Inequalities were not primarily due to differences across patients in exercising patient choice of hospital or procedure, suggesting that waiting lists may require more proactive and robust management.

This project has continued to investigate inequities in access, costs and outcomes and in 2017, further evidence relating to access, avoidable emergency hospital admissions and in-hospital mortality has been produced. This has added to knowledge of how to improve health and healthcare for vulnerable populations.

Project 5 – An example of an output from this project examined drivers of long waits in Emergency departments (EDs). EDs are seen as a ‘barometer’ of the NHS, acting as a safety net when other services are unavailable. Delays in receiving urgent treatment are a major policy concern, and may be due to factors within the hospital or to external factors such as the availability of social care. An analysis of HES A&E data identified factors driving long waits, but also highlighted much unexplained variation across providers. At the time of publication, the policy on tackling long waits was under review and the four-hour wait target operational during the study has now been dropped. However, the finding that bed occupancy levels and the age profile of attendees are independently associated with long waits remains relevant. Policy interest is evidenced by two invitations from NHS England to discuss findings (July 2017, Jan 2020). The work was presented at academic conferences (HESG (Aberdeen) 2017, International Health Congress (Oxford) 2018) and published in the CHE newsletter (March 2020) and an academic journal (Gaughan et al. Why are there long waits at English emergency departments? EJHE 21, 209–218 (2020). https://doi.org/10.1007/s10198-019-01121-7).

In 2017, previous research on patient-assessed outcomes was extended (by working with Vale of York commissioning body) to generate a web tool to support discussions between patients and their GPs about whether to undergo planned surgery. CHE developed the online tool, aftermysurgery.org.uk to inform patients about their likely outcome of hip and knee surgery and groin hernia repair. This online tool uses PROMs data to present for each user of the tool information on health outcomes experienced by other patients that have similar pre-operative characteristics. The intention of this tool is for it to be used in primary care to facilitate shared decision making between general practitioners and patients.

Other strands of the research have established that incentive payments made to GPs to increase the numbers of individuals diagnosed with dementia have been effective, but that there may have been unintended consequences in terms of adversely affecting patients’ experience of care and access to care.

The Centre for Health Economics (CHE), based at University of York, requires access to pseudonymised data to support a number of projects conducted under the Policy Research Unit (PRU) in the Economics of Health and Social Care Systems. For each of the projects described below, CHE staff analyse pseudonymised, record level data from the various datasets including:

• Hospital Episode Statistics (HES)

• Emergency Care Data (ECDS)

• Civil Registration Data

• Patient Reported Outcome Measures (PROMS)

Following use of the data under any of the projects, only aggregated results with small numbers suppressed will be published and disseminated.

The following projects are conducted by the Centre for Health Economics under PRU:

Project 1 - Measurement of efficiency, effectiveness, and productivity in the delivery of health care system nationally, sub-nationally and among hospitals.

Project 2 - Evaluation of differences in the performance of health care providers in terms of the amount and cost of provision and in patient outcomes including mortality and self-reported morbidity.

Project 3 - Evaluation of the impacts of health care policy, organisation, finance and delivery of NHS services and quantification of differences in health care utilisation, expenditure, morbidity and mortality over time, across geographic regions, health providers, and among different patient groups.

Project 4 - Investigation of inequalities in access, outcomes, and costs of health services in England.

Project 5 - Evaluation of the interface between the different sectors of the health care system, including the effects of quality and access of primary care on patient use and outcomes in secondary care; and the relationship between long term care, social care and secondary care utilisation.

Most of these projects are funded, at least in part, by the National Institute for Health Research (NIHR), via a major programme of work funded as a Policy Research Unit (PRU) in the Economics of Health and Social Care Systems (http://eshcru.ac.uk/). This programme was re-awarded for a further five years starting from January 2019. The aim of the PRU is to inform and guide policy-making in the health and social care sectors by undertaking high quality, robust and policy-relevant research, based on the discipline of economics, thereby helping to improve the health and well-being of the population, reflecting distributional concerns and population diversity. A detailed work programme for the next phase of programme funding is developed in advance in collaboration with both a Department of Health and Social Care (DHSC) Stakeholder Group and the PRU’s Advisory Group, meetings being held with each Group every six months.

The University of York is the sole Data Controller for data held under this Agreement to be processed for the purposes outlined in this section.

The University of York (CHE) has considered the legal basis for processing data under General Data Protection Regulation (GDPR) in respect of the projects covered in this agreement. CHE processes data under the following legal bases:

Article 6(1) of the GDPR (lawfulness of processing), the reference in point (e) to processing necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller.

Article 9(2)(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)

The research undertaken using NHS Digital data informs health and care policy and practice by identifying the effectiveness, efficiency, distribution, and quality of a wide range of services provided to the population. It produces insights that allow the maximisation of health gain from limited healthcare budgets, along with information on how health and health care is/can be distributed equally to meet the health needs of varying demographics. NHS Digital data provides a view of health care utilisation for CHE to understand how effective delivery of care is distributed both nationally and locally, contributing to the delivery of new healthcare policy aimed at improving the quality of care.

Approximately 20% of funding is reserved for the PRU to respond to short-term responsive requests for research. This process ensures that the work programme can be shaped to reflect enduring and emerging policy concerns. For some projects, CHE have been able to secure additional funding, enabling the University to extend or deepen the analyses of the research topic. NIHR also funds other projects, as mentioned below.

The University of York process both the pseudo-consultant code (not sensitive) and the underlying clear text consultant code (sensitive) to minimise the amount of sensitive data used by researchers across different projects. Some research projects will analyse variation in case-mix adjusted hospital inpatient activity and their related outcomes at the consultant level. For these projects, it is necessary to assign activity to consultants but it is not necessary to identify who they are. The pseudo-consultant code is therefore sufficient for these projects. Other research projects, such as the European Training Network “Improving Quality of Care in Europe (IQCE)”, Horizon 2020 Marie Sklodowska-Curie Innovative Training Network, will need access to the sensitive consultant code field to link to it other publicly available data on consultants, such as the General Medical Council (GMC) medical register (https://www.gmc-uk.org/registration-and-licensing/the-medical-register). The GMC medical register provides details on, for example, the type of registration held by a doctor and the date at which they completed their primary medical qualification. This linkage is achieved through the clear text consultant code field, which corresponds to doctors’ GMC numbers. The pseudo-consultant code is insufficient for these projects. The sensitive consultant code and the non-sensitive pseudo-consultant code will be stored as two separate files and stored in two separate folders. The files can be linked to the HES APD dataset through the epikey variable. Access to these files will be managed on a researcher and project-by-project basis. In this way, the University of York will ensure that no researcher has access to both data fields at the same time for the same project, thereby eliminating the risk of back-engineering.

The University will not publish any information at identifiable consultant level under any of the work projects described in this agreement.

The University holds a variety of NHS Digital Mental Health data sets, comprising of historic versions and newer replacements.

All of the work involves analysing the data in different ways. For example, an analysis under project 1 may focus on particular specialities, comparison of productivity across hospitals, or may be a broader assessment of national productivity. Many of the statistical methods to be employed require longitudinal data to investigate how changes in patient outcomes (including morbidity, mortality, emergency re-admissions, length of stay, admissions for conditions that could be managed in primary care, inpatient admission rates after A&E attendance) are related to changes in policy (including payment policies and incentives), changes in market configurations, changes in organisational structure, and changes in patient characteristics. Pseudonymised patient level information is required to allow for the influence of past utilisation, for demographic factors, for socio-economic factors (e.g. deprivation) linked to the small area in which patients live, and patient distance from hospitals, social care providers, and general practices. It is also essential in investigating the equity implications of policies, market structure, and organisational arrangements. The Principal Investigators and Project Leads are responsible for determining what analyses will be undertaken and what data will be used for each analysis in support of the objectives agreed with the funding organisations.

CHE confirms that the data under this Agreement would only be used for the five projects listed, and any additional project (whether as part of the DHSC programme or otherwise) would require a separate approval. Equally individuals working on each project will only be permitted to access the data relating to that project, as identified within this Agreement. Access is granted for each project only to the named individuals associated with that project under authorised user names. Such access is password controlled (with a password reset required on a regular refresh).

The controls enable a single copy of the data to be held, reducing security risk associated with multiple copies being provided per project. This model is aligned with similar arrangements for other sizeable research institutions.

Project 1 - Measurement of efficiency, effectiveness and productivity in the delivery of health care system nationally, sub-nationally and among hospitals;

The purpose of this project is to produce information for the Department of Health and Social Care (DHSC) and Secretary of State for Health on efficiency, effectiveness and productivity. In the current economic climate it is particularly important that changes in efficiency and productivity can be identified and monitored. This helps ensure accountability to the public for how the annual NHS budget is spent and to identify opportunities for better use of resources devoted to the NHS. This project provides numerical answers and context for, among others, House of Commons Health Committee, the Public Accounts Committee, Public Expenditure Inquiries, and DHSC submissions in support of annual Spending Reviews. The work also contributes to the measurement of productivity of the health service in the national accounts, compiled by the Office of National Statistics.

Funder:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems.

This project will use only the following data supplied under this Agreement: HES APC 1998/99-2020/21 A&E 2007/08 - 2017/18; Critical Care 2011/12 – 2020/21; Outpatient 2011/12-2020/21; PROMs 2009/10 –2020/21; Civil Registration (Mortality) data 1998/99 - 2020/21; Emergency Care Dataset (ECDS) 2017/18 – 2020/21;

Project 1 will only process data provided through Annual dissemination for the datasets listed above.

Data has been assessed and minimised to only utilise the required data from that disseminated under this Agreement for this project and cannot be minimised further.

The University of York will estimate survival models that relate the timing of death (based on the exact date of death) for individual patients to the relevant unit of assessment (provider, region) after adjusting for relevant case-mix differences across the populations under study.

Most of the work undertaken under this project involves measurement of productivity over time, hence the need to hold the data from 1998/99. It is also necessary to construct aggregated measures of NHS output and quality based on what has happened to each individual patient in whatever setting care has been delivered, hence the need for patient-level information. The project also requires use of the sensitive PROMs data as measures of the quality of health care.

Project 2 - Evaluation of differences in the performance of health care providers in terms of the amount and cost of provision and in patient outcomes including mortality and self-reported morbidity;

The purpose of this project is to produce information for National and local decision makers, such as the Department of Health and Social Care (DHSC), Clinical Commissioning Groups (CCGs) and Local Authorities (LAs), to assist decisions regarding the provision of services that offer the greatest value for money according to the benefits achieved. Delivering appropriate, high quality, health care services to patients, in the most cost-effective way, are important priorities in any health care system. Advancing these priorities requires the analyses of such things as variations in practice and of the relationship between patient outcomes and hospital and consultant workload; which dimensions of performance are most important to patients; and the extent to which financial incentives motivate best practice. Ultimately this project informs the assessment of the most efficient and cost-effective way of delivering a particular service. This helps ensure accountability to the public for how the annual NHS budget is spent and helps to identify opportunities for better use of resources devoted to the NHS. The project is designed to develop a more systematic evidence base that will allow policy-makers, providers and commissioners to develop policies to achieve efficiency targets and outcome-based commissioning, publish information on performance in formats that are most useful for the intended stakeholders, and to redeploy resources to produce more efficient mixes of services both within and across the health and social care sectors.

Funders:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems.

• National Institute for Health Research (NIHR) Collaboration for Leadership in Applied Health Research and Care Yorkshire and Humber (CLAHRC YH) (Ref NIHR CLARHC YH II 14653)

• NHS England (Ref R1790901) Economic evaluation of the Fragility Hip Fracture Best Practice Tariff.

• NIHR HS&DR (Ref DRF-2016-09-097): Doctoral Research Fellowship - "Providers' response on the Pay for Performance incentives".

• European Commission, European Training Network. “Improving Quality of Care in Europe (IQCE)”, Horizon 2020 Marie Skłodowska-Curie Innovative Training Networks.

The work for all these funders will require the sensitive PROMs data to measure patient outcomes.

The project will use only the following data: HES APC 1989/90 – 2020/21 Sensitive field: Consultant Code; HES Outpatient 2002/03 – 2020/21; PROMs 2009/10 – 2020/21; Civil Registration (Deaths) 1998/99 - 2020/21.

Project 2 will utilise both Annual and Monthly disseminations of the datasets indicated above.

Project 3 - Evaluation of the impacts of health care policy, organisation, finance and delivery of NHS services and quantification of differences in health care utilisation, expenditure, morbidity and mortality over time, across geographic regions, health providers, and among different patient groups.

The purpose of this project is to produce evidence to inform NHS England and the Department of Health and Social Care’s decisions on resource allocation, funding models, and the design and direction of future policy regarding the health, mental health, and social care sectors, with CHE’s advice and analyses being regularly sought to feed into White papers and specific government reviews.

This project includes understanding which types of budgeting, organisation, structure and contracting arrangements for health, incl. mental health, and social care services best achieves strategic goals. It also includes evaluations of payment policies (including financial incentive schemes) and changes to the organisation of services (e.g. co-location of general practitioners alongside emergency departments, mergers of providers, vertical integration of providers, care pathways) that seek to encourage good quality, cost-effective care and/or facilitate access to timely care. The main aims are to: analyse the potential for use of different organisational structures and payment mechanisms in health and social care to improve overall performance; analyse the impact that different payment policies and service configurations can have on prices, outputs, quality and outcomes; explore how the best payment systems and service configurations could be implemented in practice; and establish the effect of innovative organisational forms on costs and quality of care.

Funders:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems.

• NIHR HS&DR 10/1011/22 and NIHR HS&DR 13/54/40: Relationships between quality of primary care and secondary care outcomes for people with mental illness.

• Wellcome Trust [ref: 105624] through the Centre for Chronic Diseases and Disorders (C2D2) at the University of York: Finance and organisation of mental health services.

• Health Foundation [ref: 57151] Efficiency, cost and quality of mental healthcare provision.

• NIHR HS&DR (Ref DRF/2014-07-055): Doctoral Research Fellowship - Measuring & explaining variations in general practice performance.

• NIHR HS&DR (Ref 15/145/06): General Practitioners and Emergency Departments (GPED): Efficient Models of Care.

The project will use only the following data: HES APC 1998/99 – 2020/21; A&E 2007/08 – 2017/18; Outpatient 2002/03 – 2020/21; PROMs 2009/10 – 2020/21; Emergency Care Dataset 2017/18 – 2020/21; MHMDS 2011/12 – 2013/14; MHLDS 2014/15 – 2015/16; MHSDS 2016/17-2020/21; Civil Registration (Deaths) 1998/99-2020/21; HES APC Sensitive Psychiatric Fields: Detention category (DETNCAT), Legal group of patient (psychiatric) (LEGALGPC), Legal status classification (LEGLSTAT).

Project 3 will utilise both Annual and Monthly disseminations of the datasets listed above.

The work for several funders will require the use of the PROMs data to measure morbidity over time.

This project will also require use of MHMDS/MHLDS/MHSDS data linked to HES data in order to carry out analyses into the economics around mental health and mental health care provision. CHE is requesting sensitive MHMDS/MHLDS/MHSDS fields and sensitive HES psychiatric fields (Legal group of patient, Legal status classification, and Detention category). These relate to the legal category / legal status of the patient which is an important indicator of patient severity. CHE will need these sensitive data items to accurately control for the impact of detention on resource use and utilisation. CHE needs to check data consistency between HES and the MHMDS/MHLDS/MHSDS and therefore requires sensitive data on legal status in both datasets.

Project 4 - Investigation of inequalities in access, outcomes and costs of health services in England.

The purpose of this project is to produce information that NHS England and Clinical Commissioning Groups will use to address the NHS’ duty under the Health and Social Care Act 2012 to consider reducing health inequalities, and that Public Health England, Local Authorities and a variety of other public and third sector organisations will use to inform decision making and quality assurance around health and social care and wider public policies with impacts on health. CHE has developed new methods of local health equity monitoring for health care quality assurance, which NHS England adopted in 2016. In collaboration with colleagues at the Department of Health Sciences, University of York, and analysts at NHS England, CHE will refine and use these methods and related measures to monitor the progress of national and local NHS organisations in reducing inequalities in healthcare access and outcomes, to gain insight into the determinants of inequalities and which local areas show sustained improvements and deteriorations in health inequality and why, and to evaluate the equity impacts of local new models of care and other health policies. The work will also assist the ONS to conduct distributional analyses of NHS spending for use in constructing statistics about in-kind social transfers.

The work for the Department of Health and Social Care will investigate why providers respond differently to policy incentives and quantify the associated impact on inequalities in care quality and/or access, including how these change over time.

Funders:

• NIHR TCC (Ref SRF-2013-06-015) Health equity impacts: evaluating the impacts of organisations and interventions on social inequalities in health.

• Wellcome Trust. ‘Re-Engineering Health Policy Research for Fairer Decisions and Better Health’.

• Department of Health to the Policy Research Unit in the Economics of Health and Social Care Systems

The project will use only the following data: HES APC 1989/90 – 2020/21; A&E 2007/08 – 2017/18; Outpatient 2002/03 – 2020/21; Critical Care 2011/12 – 2020/21; PROMs 2009/10 – 2020/21; Civil Registration (Deaths) 1998/99-2020/21; Emergency Care Dataset 2017/18 – 2020/21.

Project 4 will utilise both Annual and Monthly disseminations of the datasets listed above.

The work requires the use of sensitive PROMs data to measure patient outcomes in secondary care.

The following is a separately funded piece of research work which fits within the scope of project 4:

Funder:

• NIHR Policy Research Programme (grant number PR-X06-1014-22005), Partnership for Responsive Policy Analysis and Research (PREPARE), a collaboration between the University of York and the King's Fund.

Though PREPARE is a collaboration between the University of York and the King’s Fund, the King’s Fund has no involvement in this research project. The research request was considered by the co-leads in the University of York and the King’s Fund who agreed that the University of York would carry out the research. The lead within the University of York has been solely responsible for all decisions on how this research will be carried out including all decisions in respect of what data processing is required.

Description of additional use of data:

This area of the project will explore the links between child health and child poverty, in particular the NHS hospital utilisation of children born into deprivation (using the indices of deprivation (ID) as a proxy for poverty) in comparison with children who are not born in deprived areas. The study will also explore whether any difference in hospital utilisation over the early life course has changed over time. To do this the project will create a birth cohort of children born in NHS hospitals in England in specific financial year (2000, 2005, 2010, 2015, 2018), and track their use of NHS services (inpatient, outpatient and A&E) over their life course (up to age 18 for those born in 2000). The analysts will then test whether the age-sex adjusted differential use of hospital services across children born into rich and poor neighbourhoods has changed over time.

The work will use the following data:

HES APC 2000/01 – 2020/21; HES A&E 2007/08 - 2018/19; HES OP 2002/03 – 2020/21; Emergency Care Dataset 2017/18 – 2020/21.

The work will utilise both Annual and Monthly disseminations of the datasets listed above.

Linkages:

There is a need to link the HES APC data to Index of Multiple Deprivation (IMD) via the child’s Lower Super Output Area (LSOA) of residence. An older version of the IMD (2004) is already included in HES but the study will have to link to subsequent versions of the ID.

This Agreement permits the use of the data for this research but does not permit the use of the data for other research projects funded by PREPARE. Use of the data for any other projects funded by PREPARE would require a formal amendment to this Data Sharing Agreement.

Project 5 - Evaluation of the interface between the different sectors of the health care system, including the effects of quality and access of primary care on patient use and outcomes in secondary care; and the relationship between long term care, social care and secondary care utilisation.

It has long been understood that health and social care services frequently provide treatment and care for the same individuals, so ensuring that these are ‘joined up’ or well co-ordinated has been an important and long-standing policy objective. In practice, however, both the services and approaches to monitoring these have developed separately, with potential implications for the efficiency and effectiveness of both health and social care. The purpose of this project is to produce evidence that will be used by the Department of Health and Social Care and commissioners to inform discharge arrangements and the design of integrated care arrangements and to identify opportunities for substitution of different types of health and social care services. CHE has also developed an online web tool to inform patients about their likely outcome of surgery to impact on shared decision making in primary care in York.

Funder:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems CHE Lead: Anne Mason

The project will use only the following data: HES APC 1989/90– 2020/21; A&E 2007/08 – 2017/18; Emergency Care Dataset 2017/18 – 2020/21; MHMDS 2011/12 – 2013/14; MHLDS 2014/15 – 2015/16; MHSDS 2016/17 - 2017/18; Outpatient 2002/03 – 2020/21; Critical Care 2011/12 – 2020/21, PROMs 2009/10 – 2020/21; Civil Registration (Deaths) 1998/99-2020/21.

Project 5 will utilise both Annual and Monthly disseminations of the datasets listed above.

This project requires the sensitive PROMs data to measure patient outcomes in secondary care.

The Centre for Health Economics has a long-established track record in the delivery of policy research that utilises HES data, as recognized by the award of the Queens Anniversary Prize in 2007. Examples of recent publications arising from the above projects that have employed the HES data can be found at the links below:

https://eshcru.com/publications/

http://www.york.ac.uk/che/publications/in-house/ https://www.york.ac.uk/che/publications/all/

The outputs from all of the projects will include

• Reports and slide decks to funding bodies;

• peer reviewed papers in academic journals;

• lay summaries such as newsletters and blogs;

• conference and seminar presentations to a variety of audiences, such as academic, policy, professional and public audiences.

Reports will be produced containing aggregate results that show trends over time, differences across providers, commissioners, geographical areas and by patient subgroups and patient characteristics. The results will contain estimated correlations showing associations between patient outcomes and patient characteristics, hospital, institutional, geographic and environmental factors. Statistical results will be presented in interactive spreadsheets or “Dashboards” (e.g. similar to http://health-inequalities.blogspot.co.uk/ which uses QOF data and only contains aggregated data which can be interrogated), tables and maps of aggregate statistics summarising patient characteristics

All outputs will contain only data that are aggregated with small numbers suppressed in line with the HES Analysis Guide.

The dissemination and communication strategy will vary between projects and will have been previously agreed with the funders. It would normally include the oral presentations; interim and final reports; and published reports and journal articles. Examples are provided below:

Oral presentations / knowledge exchange

• Presentation of interim and emerging findings to study advisory groups and/or steering committees. Members - who typically include policy makers, clinicians, academics and public contributors - provide feedback and advice.

• Interactive workshops with policy analysts from DHSC and NHSE/I to discuss emerging findings and ensure policy relevance

• Presentations to CCGs, NHS trusts, and PPI groups

• Open lectures and invited talks at Universities/Research Centres both in the UK and abroad

• Oral or poster presentations at national and international conferences, such as Health Economists’ Study Group, iHEA, and EuHEA. Delegates may include international organisations such as The World Bank, OECD and WHO, alongside members of the international academic community.

• End of project workshops or conferences to present research findings to key stakeholders and policy makers

Unpublished reports

• Draft reports with preliminary findings to advisory groups

• Interim reports for funders and policy analysts

• Draft final reports for funders. These are usually peer reviewed externally by academics and internally by policy analysts

Publications

• Published reports containing full, detailed findings, with an accompanying lay summary to make key messages more accessible.

• Press releases to accompany the publications of reports (full or short), via the University of York Press Office as well as through the CHE website and social media platforms, as well as funders own Press Release Offices and social media platforms;

• Peer reviewed scientific papers in academic and policy journals

• Short articles in CHE annual reports and CHE newsletters

Target dates for outputs

The outputs from each project will be delivered in accordance with CHE’s funding contracts, which run to different timelines with various milestones for each. Below are examples of key milestones and timelines for some of the projects undertaken.

Project 1 - The primary output from this ongoing project is the production each year of an annual update to national NHS productivity figures that incorporates the most recent financial year of data. Under this project, CHE has demonstrated that NHS productivity growth is meeting the requirements of the Five Year Forward View and outpaces that of the economy as a whole. CHE’s figures are widely used to inform policy discourse, with the DHSC relying on the information for internal monitoring purposes and for external reporting and response purposes, such as to inform annual Spending Reviews. Under this project, CHE also provides data about the quality of NHS care to the Office of National Statistics that are used in the construction of the national accounts.

In addition to the annual update of national figures, CHE also undertakes analyses of variation in hospital productivity and produces short reports, memorandum or slide decks for the DHSC to address specific questions about NHS productivity. CHE presents the work regularly to various audiences, including politicians, policy makers, academics, health professions and the general public through seminars, conference presentations and media appearances.

Project 3 – Policy makers need timely evidence on the impact of health care policies. These policies concern both the supply side of health care - the organisation, finance and delivery of services - and the demand side, such as utilisation, morbidity and mortality. Under this project, analyses may investigate variations over time and/or across geographic regions, providers, or patient groups. Empirical research frequently relies on analyses of HES. Outputs include open access reports and scientific papers in academic journals. Interim and final reports are submitted to funders as part of our contractual requirements. Emerging findings are presented and discussed at regular advisory and stakeholder group meetings, as well as at invited seminars and workshops. Examples of forthcoming outputs include

- Siciliani et al. Paying for health benefits using PROMs data, CHE Research Paper: November 2020

- Gutacker et al, Evaluating the intended and unintended consequences of best practice tariffs on patient health outcomes and provider behaviour. Report to DHSC. October 2020.

- Rice et al, Characterising end-of-life hospital expenditure. Oral presentation to the DHSC Oversight Group (steering committee) October 2020.

Project 5 – The focus of this project is the evaluation of the interface between the different sectors of the health care system, such as the effects of quality and access of primary care on patient use and outcomes in secondary care; and the relationship between long-term care, social care and secondary care utilisation. Published outputs from this project include open access reports and scientific papers in academic journals. Interim and final reports to funders will be provided in line with contractual requirements. Emerging findings will be presented at conferences, advisory and stakeholder group meetings, or at DHSC workshops or seminars. This ongoing engagement helps ensure the work remains policy relevant. An example is a departmental working paper examining the effects of changes in social care funding on health care utilisation (Pace et al. Understanding the interdependencies between health and social care resources and arrangements. CHE Research Paper, December 2020). Early findings from this work were shared with NHSE (May 2018), discussed with DHSC (March 2019), with further analyses undertaken in response to their questions about the impact of the Better Care Fund. The report was revised following written comments from the Department (May 2019), and is currently under review with a health economics journal.

Targets dates and deliverables met

Project 1: The NHS spends over £120 billion a year and the budget provided for the NHS is influenced by the evidence on how well the money is spent. The annual estimates of productivity that we produce for the Department of Health and Social Care are a key part of their negotiations with the Treasury about the size of the budget and the research is cited in exchanges with ministers and government committees that scrutinise NHS spending. In particular, the quality adjustment method devised by York is important in ensuring that productivity is not under-stated which may have a negative impact on the size of budget settlements and hence lead to lower investment in the NHS. The most recent update to the NHS productivity measure https://www.york.ac.uk/media/che/documents/papers/researchpapers/CHERP171_NHS_productivity_update2017_18.pdf) shows a steady positive increase and that NHS productivity has increased substantially faster than the overall economy.

The methods developed by the York research team have also caught interests of Governments and health ministries in other parts of the world (eg Japan, Italy, Malaysia, Sweden), which have resulted in a number of workshop and collaborative work.

Project 2: Examples of outputs produced from this project include journal articles on the determinants of health care expenditure. For example, an article by Howdon and Rice (Health care expenditures, age, proximity to death and morbidity: Implications for an ageing population, 2017, https://doi.org/10.1016/j.jhealeco.2017.11.001) investigates the implications of an ageing population on health care expenditure, showing that the latter is primarily determined by proximity to death, rather than age. Chalkley et al. (Elective hospital admissions: secondary data analysis and modelling with an emphasis on policies to moderate growth, 2017, https://doi.org/10.3310/hsdr05070) investigating growth in elective hospital admissions in England find that these are mainly due to increases in both demand for and supply of hospital care, with health system reforms and ageing population less important.

Project 3: Outputs from this project include journal articles on payment systems, providing evidence on intended and unintended impacts for policy makers designing new approaches to paying for care. For example, one study examined the effects of two incentive schemes in primary care that aimed to tackle underdiagnosis in dementia, with HES data used to control for a parallel hospital scheme. The schemes were effective in increasing dementia diagnoses (Mason, et al 2018, Investigating the impact of primary care payments on underdiagnosis in dementia: a difference-in-differences analysis, IJGP. https://doi.org/10.1002/gps.4897). However, patient satisfaction was lower in practices that achieved higher rates of dementia diagnoses (Liu, et al 2019, Incentive schemes to increase dementia diagnoses in primary care in England: a retrospective cohort study of unintended consequences. BJGP https://doi.org/10.3399/bjgp19X701513).

Project 4: Access to health services in England is generally lower for disadvantaged groups. This is a key policy concern because in the NHS access is ideally based on need. One study that used HES data to investigate inequalities in access to health services in England is by Moscelli, et al 2017 (Socioeconomic Inequality of Access to Healthcare: Does Choice Explain the Gradient? JHE https://doi.org/10.1016/j.jhealeco.2017.06.005). The authors investigated the extent of waiting time inequalities for elective coronary revascularisation procedures by socioeconomic status within hospitals. Using sophisticated statistical techniques, they showed that patients from the most deprived areas waited around one-third longer than the least deprived groups. Inequalities were not primarily due to differences across patients in exercising patient choice of hospital or procedure, suggesting that waiting lists may require more proactive and robust management.

Project 5: An example of an output from this project examined drivers of long waits in Emergency departments (EDs). EDs are seen as a ‘barometer’ of the NHS, acting as a safety net when other services are unavailable. Delays in receiving urgent treatment are a major policy concern, and may be due to factors within the hospital or to external factors such as the availability of social care. An analysis of HES A&E data identified factors driving long waits, but also highlighted much unexplained variation across providers. At the time of publication, the policy on tackling long waits was under review and the four-hour wait target operational during the study has now been dropped. However, the finding that bed occupancy levels and the age profile of attendees are independently associated with long waits remains relevant. Policy interest is evidenced by two invitations from NHS England to discuss findings (July 2017, Jan 2020). The work was presented at academic conferences (HESG (Aberdeen) 2017, International Health Congress (Oxford) 2018) and published in the CHE newsletter (March 2020) and an academic journal (Gaughan et al. Why are there long waits at English emergency departments? EJHE 21, 209–218 (2020). https://doi.org/10.1007/s10198-019-01121-7).

The DSA includes as funders the European Commission, European Training Network. “Improving Quality of Care in Europe (IQCE)”, Horizon 2020 Marie Skłodowska-Curie Innovative Training Networks, which funds two PhD research projects. The outputs for this specific project is the compilation and publication of a doctorate thesis and related research articles to be published in international research journal. Both PhD research projects are nearing completion by the end of 2020, but with access to NHS Digital data required to prepare research articles and respond to potential comments/revision requests from journal reviewers.

Project 1 - The primary output from this ongoing project is the production each year of an annual update to national NHS productivity figures that incorporates the most recent financial year of data. In 2017, the annual update was produced for the DHSC and, as in other years, was used by them externally and internally in monitoring, informing policy debate, the annual spending review and negotiations on budget setting. Under this project, CHE also provided data about the quality of NHS care to the Office of National Statistics that are used each year in the construction of the national accounts. In 2017, additional analyses on hospital level productivity were produced for the DHSC, examining the factors underlying variation in productivity, which assists the DHSC in exploring how to get the best value from NHS resources.

Project 2 - This project has produced a range of evidence that allows the DHSC and other organisations to understand and plan expenditure on different aspects of NHS care, to account for changes in activity and also to understand aspects of quality of care, First, it has investigated the drivers of health care expenditure, disentangling the influence of age, morbidity and proximity to death on the level of expenditure. This has important implications for the DHSC in terms of planning and for the design of the resource allocation formulae used to distribute the healthcare budget, Second, it has explored several aspects of activity in the secondary care sector, for instance, policies for moderating growth in elective admissions and how the provision of specialist rehabilitation services affect the duration and costs of hospital care. It has also looked at the use of patient valuation of the care they receive as a way of measuring the quality of hospital care. Third, it has investigated in-hospital mortality trends over time. These insights allow policy makers to appreciate the costs and the benefits of NHS provision and to plan for more effective and efficient care.

Project 3 - The research in this project has evaluated whether the way in which the NHS is organised can affect the costs and outcomes of services to help inform policy decisions. In 2017, the research has produced evidence on a number of issues in mental health, including the relationship between costs and quality and socio-economic inequalities in access to care. Other examples include the production of evidence on the impact of the electronic booking system on referrals and non-attendances and the investigation of reasons for delayed discharges from hospital, both of which are important in terms of planning future organisation of care.

Project 4 - This project has continued to investigate inequities in access, costs and outcomes and in 2017, further evidence relating to access, avoidable emergency hospital admissions and in-hospital mortality has been produced. This has added to knowledge of how to improve health and healthcare for vulnerable populations.

Project 5 – In 2017, previous research on patient-assessed outcomes was extended (by working with Vale of York CCG) to generate a web tool to support discussions between patients and their GPs about whether to undergo planned surgery. The on-line tool, aftermysugery.org.uk can be used by patients and their GPs, who input basic demographic data and fill in a pre-operative health status questionnaire. The webtool then returns a predicted post-operative health status, together with national comparator data, displayed in various visual formats. This information is designed to a) help patients decide whether they feel the expected health improvement is sufficiently high to make having the operation worthwhile, b) inform patients about the likelihood of a negative outcome, and c) provide information about which hospitals secure better outcomes for their patients.

Other strands of the research have established that incentive payments made to GPs to increase the numbers of individuals diagnosed with dementia have been effective, but that there may have been unintended consequences on patient experience and access.

The Centre for Health Economics (CHE), based at University of York is requesting data for the following projects involving economic analyses of health and social care. Please note that for each of the following projects CHE staff will analyse individual level data from the various datasets. Only aggregated results will be published and disseminated.

Most of these projects are funded, at least in part, by the National Institute for Health Research (NIHR), via a major programme of work funded as a Policy Research Unit (PRU) in the Economics of Health and Social Care Systems (http://eshcru.ac.uk/). This programme was re-awarded for a further five years starting from January 2019. The aim of the PRU is to inform and guide policy-making in the health and social care sectors by undertaking high quality, robust and policy-relevant research, based on the discipline of economics, thereby helping to improve the health and well-being of the population, reflecting distributional concerns and population diversity. A detailed work programme for the next phase of programme funding is developed in advance in collaboration with both a DHSC Stakeholder Group and the PRU’s Advisory Group, meetings being held with each Group every six months. Approximately 20% of funding is reserved for the PRU to respond to short-term responsive requests for research. This process ensures that the work programme can be shaped to reflect enduring and emerging policy concerns. For some projects, CHE have been able to secure additional funding, enabling the University to extend or deepen the analyses of the research topic. NIHR also funds other projects, as mentioned below.

The University of York request both the pseudo-consultant code (not sensitive) and the underlying clear text consultant code (sensitive) to minimise the amount of sensitive data used by researchers across different projects. Some research projects will analyse variation in case-mix adjusted hospital inpatient activity and their related outcomes at the consultant level. For these projects, it is necessary to assign activity to consultants but it is not necessary to identify who they are. The pseudo-consultant code is therefore sufficient for these projects. Other research projects, such as the European Training Network “Improving Quality of Care in Europe (IQCE)”, Horizon 2020 Marie Sklodowska-Curie Innovative Training Network, will need access to the sensitive consultant code field to link to it other publicly available data on consultants, such as the General Medical Council (GMC) medical register (https://www.gmc-uk.org/registration-and-licensing/the-medical-register). The GMC medical register provides details on, for example, the type of registration held by a doctor and the date at which they completed their primary medical qualification. This linkage is achieved through the clear text consultant code field, which corresponds to doctors’ GMC numbers. The pseudo-consultant code is insufficient for these projects. The sensitive consultant code and the non-sensitive pseudo-consultant code will be stored as two separate files and stored in two separate folders. The files can be linked to the HES APD dataset through the epikey variable. Access to these files will be managed on a researcher and project-by-project basis. In this way, the University of York will ensure that no researcher has access to both data fields at the same time for the same project, thereby eliminating the risk of back-engineering.

The university hold a variety of NHS Digital Mental Health data sets, comprising of historic versions and newer replacements.

All of the work involves analysing the data in different ways. For example, an analysis under project 1 may focus on particular specialities, comparison of productivity across hospitals, or may be a broader assessment of national productivity. Many of the statistical methods to be employed require longitudinal data to investigate how changes in patient outcomes (including morbidity, mortality, emergency re-admissions, length of stay, admissions for conditions that could be managed in primary care, inpatient admission rates after A&E attendance) are related to changes in policy (including payment policies and incentives), changes in market configurations, changes in organisational structure, and changes in patient characteristics. Pseudonymised patient level information is required to allow for the influence of past utilisation, for demographic factors, for socio-economic factors (e.g. deprivation) linked to the small area in which patients live, and patient distance from hospitals, social care providers, and general practices. It is also essential in investigating the equity implications of policies, market structure, and organisational arrangements. The Principal Investigators and Project Leads are responsible for determining what analyses will be undertaken and what data will be used for each analysis in support of the objectives agreed with the funding organisations.

CHE confirms that the data under this application would only be used for the five projects listed, and any additional project (whether as part of the DHSC programme or otherwise) would require a separate approval. Equally individuals working on each project will only be permitted to access the data relating to that project, as identified within this application. Access is granted for each project only to the named individuals associated with that project under authorised user names. Such access is password controlled (with a password reset required on a regular refresh).

The controls enable a single copy of the data to be held, reducing security risk associated with multiple copies being provided per project. This model is aligned with similar arrangements for other sizeable research institutions.

Project 1 - Measurement of efficiency, effectiveness and productivity in the delivery of health care system nationally, sub-nationally and among hospitals;

The purpose of this project is to produce information for the Department of Health and Social Care (DHSC) and Secretary of State for Health on efficiency, effectiveness and productivity. In the current economic climate it is particularly important that changes in efficiency and productivity can be identified and monitored. This helps ensure accountability to the public for how the annual NHS budget is spent and to identify opportunities for better use of resources devoted to the NHS. This project provides numerical answers and context for, among others, House of Commons Health Committee, the Public Accounts Committee, Public Expenditure Inquiries, and DHSC submissions in support of annual Spending Reviews. The work also contributes to the measurement of productivity of the health service in the national accounts, compiled by the Office of National Statistics.

Funder:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems.

This project will use only the following data: HES APC 1998/99-2017/18; A&E 2007/08 - 2017/18; Critical Care 2011/12 – 2017/18; Outpatient 2011/12-2017/18; PROMs 2009/10 – 2017/18;

The Civil Registration (Moratlity) data from 1998/99-2017/18 are needed to:

measure and compare the quality of care between healthcare providers, other healthcare organisation, regions and other geographical areas, where quality is defined as risk of mortality following treatment. The University of York will relate this variation to individual health policies or economic incentives to study the behaviour of healthcare providers and/or purchasers of care. The University of York will also relate this variation to socio-economic characteristics of the population served by public healthcare services to study inequalities in the quality of care received.

The University of York will estimate survival models that relate the timing of death (based on the exact date of death) for individual patients to the relevant unit of assessment (provider, region) after adjusting for relevant case-mix differences across the populations under study.

Most of the work undertaken under this project involves measurement of productivity over time, hence the need to hold the data from 1998/99. It is also necessary to construct aggregated measures of NHS output and quality based on what has happened to each individual patient in whatever setting care has been delivered, hence the need for patient-level information. The project also requires use of the sensitive PROMs data as measures of the quality of health care.

Project 2 - Evaluation of differences in the performance of health care providers in terms of the amount and cost of provision and in patient outcomes including mortality and self-reported morbidity;

The purpose of this project is to produce information for National and local decision makers, such as the Department of Health and Social Care (DHSC), Clinical Commissioning Groups (CCGs) and Local Authorities (LAs), to assist decisions regarding the provision of services that offer the greatest value for money according to the benefits achieved. Delivering appropriate, high quality, health care services to patients, in the most cost-effective way, are important priorities in any health care system. Advancing these priorities requires the analyses of such things as variations in practice and of the relationship between patient outcomes and hospital and consultant workload; which dimensions of performance are most important to patients; and the extent to which financial incentives motivate best practice. Ultimately this project informs the assessment of the most efficient and cost-effective way of delivering a particular service. This helps ensure accountability to the public for how the annual NHS budget is spent and helps to identify opportunities for better use of resources devoted to the NHS. The project is designed to develop a more systematic evidence base that will allow policy-makers, providers and commissioners to develop policies to achieve efficiency targets and outcome-based commissioning, publish information on performance in formats that are most useful for the intended stakeholders, and to redeploy resources to produce more efficient mixes of services both within and across the health and social care sectors.

Funders:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems.

• National Institute for Health Research (NIHR) Collaboration for Leadership in Applied Health Research and Care Yorkshire and Humber (CLAHRC YH) (Ref NIHR CLARHC YH II 14653)

• NHS England (Ref R1790901) Economic evaluation of the Fragility Hip Fracture Best Practice Tariff.

• NIHR HS&DR (Ref DRF-2016-09-097): Doctoral Research Fellowship - "Providers' response on the Pay for Performance incentives".

• European Commission, European Training Network. “Improving Quality of Care in Europe (IQCE)”, Horizon 2020 Marie Skłodowska-Curie Innovative Training Networks.

The work for all these funders will require the sensitive PROMs data to measure patient outcomes.

The project will use only the following data: HES APC 1989/90 - 2017/18, Sensitive field: Consultant Code; HES Outpatient 2002/03 - 2017/18; PROMs 2009/10 – 2017/18; Civil Registration (Deaths) 1998/99-2017/18.

Project 3 - Evaluation of the impacts of health care policy, organisation, finance and delivery of NHS services and quantification of differences in health care utilisation, expenditure, morbidity and mortality over time, across geographic regions, health providers, and among different patient groups.

The purpose of this project is to produce evidence to inform NHS England and the Department of Health and Social Care’s decisions on resource allocation, funding models, and the design and direction of future policy regarding the health, mental health, and social care sectors, with CHE’s advice and analyses being regularly sought to feed into White papers and specific government reviews.

This project includes understanding which types of budgeting, organisation, structure and contracting arrangements for health, incl. mental health, and social care services best achieves strategic goals. It also includes evaluations of payment policies (including financial incentive schemes) and changes to the organisation of services (e.g. co-location of general practitioners alongside emergency departments, mergers of providers, vertical integration of providers, care pathways) that seek to encourage good quality, cost-effective care and/or facilitate access to timely care. The main aims are to: analyse the potential for use of different organisational structures and payment mechanisms in health and social care to improve overall performance; analyse the impact that different payment policies and service configurations can have on prices, outputs, quality and outcomes; explore how the best payment systems and service configurations could be implemented in practice; and establish the effect of innovative organisational forms on costs and quality of care.

Funders:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems.

• NIHR HS&DR 10/1011/22 and NIHR HS&DR 13/54/40: Relationships between quality of primary care and secondary care outcomes for people with mental illness.

• Wellcome Trust [ref: 105624] through the Centre for Chronic Diseases and Disorders (C2D2) at the University of York: Finance and organisation of mental health services.

• Health Foundation [ref: 57151] Efficiency, cost and quality of mental healthcare provision.

• NIHR HS&DR (Ref DRF/2014-07-055): Doctoral Research Fellowship - Measuring & explaining variations in general practice performance.

• NIHR HS&DR (Ref 15/145/06): General Practitioners and Emergency Departments (GPED): Efficient Models of Care.

The project will use only the following data: HES APC 1998/99 – 2017/18; A&E 2007/08 – 2017/18; Outpatient 2002/03 – 2017/18; PROMs 2009/10 – 2017/18; MHMDS 2011/12 – 2013/14; MHLDS 2014/15 – 2015/16; MHSDS 2016/17-2017/18; Civil Registration (Deaths) 1998/99-2017/18; HES APC Sensitive Psychiatric Fields: Detention category (DETNCAT), Legal group of patient (psychiatric) (LEGALGPC), Legal status classification (LEGLSTAT).

The work for several funders will require the use of the PROMs data to measure morbidity over time.

This project will also require use of MHMDS/MHLDS/MHSDS data linked to HES data in order to carry out analyses into the economics around mental health and mental health care provision. CHE is requesting sensitive MHMDS/MHLDS/MHSDS fields and sensitive HES psychiatric fields (Legal group of patient, Legal status classification, and Detention category). These relate to the legal category / legal status of the patient which is an important indicator of patient severity. CHE will need these sensitive data items to accurately control for the impact of detention on resource use and utilisation. CHE needs to check data consistency between HES and the MHMDS/MHLDS/MHSDS and therefore requires sensitive data on legal status in both datasets.

Project 4 - Investigation of inequalities in access, outcomes and costs of health services in England.

The purpose of this project is to produce information that NHS England and Clinical Commissioning Groups will use to address the NHS’ duty under the Health and Social Care Act 2012 to consider reducing health inequalities, and that Public Health England, Local Authorities and a variety of other public and third sector organisations will use to inform decision making and quality assurance around health and social care and wider public policies with impacts on health. CHE has developed new methods of local health equity monitoring for health care quality assurance, which NHS England adopted in 2016. In collaboration with colleagues at the Department of Health Sciences, University of York, and analysts at NHS England, CHE will refine and use these methods and related measures to monitor the progress of national and local NHS organisations in reducing inequalities in healthcare access and outcomes, to gain insight into the determinants of inequalities and which local areas show sustained improvements and deteriorations in health inequality and why, and to evaluate the equity impacts of local new models of care and other health policies. The work will also assist the ONS to conduct distributional analyses of NHS spending for use in constructing statistics about in-kind social transfers.

The work for the Department of Health and Social Care will investigate why providers respond differently to policy incentives and quantify the associated impact on inequalities in care quality and/or access, including how these change over time.

Funders:

• NIHR TCC (Ref SRF-2013-06-015) Health equity impacts: evaluating the impacts of organisations and interventions on social inequalities in health.

• Wellcome Trust. ‘Re-Engineering Health Policy Research for Fairer Decisions and Better Health’.

• Department of Health to the Policy Research Unit in the Economics of Health and Social Care Systems

The project will use only the following data: HES APC 1989/90 – 2017/18; A&E 2007/08 – 2017/18; Outpatient 2002/03 – 2017/18; Critical Care 2011/12 – 2017/18; PROMs 2009/10 – 2017/18; Civil Registration (Deaths) 1998/99-2017/18.

The work requires the use of sensitive PROMs data to measure patient outcomes in secondary care.

Project 5 - Evaluation of the interface between the different sectors of the health care system, including the effects of quality and access of primary care on patient use and outcomes in secondary care; and the relationship between long term care, social care and secondary care utilisation.

It has long been understood that health and social care services frequently provide treatment and care for the same individuals, so ensuring that these are ‘joined up’ or well co-ordinated has been an important and long-standing policy objective. In practice, however, both the services and approaches to monitoring these have developed separately, with potential implications for the efficiency and effectiveness of both health and social care. The purpose of this project is to produce evidence that will be used by the Department of Health and Social Care and commissioners to inform discharge arrangements and the design of integrated care arrangements and to identify opportunities for substitution of different types of health and social care services. CHE has also developed an online web tool to inform patients about their likely outcome of surgery to impact on shared decision making in primary care in York.

Funder:

• Department of Health and Social Care to the Policy Research Unit in the Economics of Health and Social Care Systems CHE Lead: Anne Mason

The project will use only the following data: HES APC 1989/90– 2017/18; A&E 2007/08 – 2017/18; MHMDS 2011/12 – 2013/14; MHLDS 2014/15 – 2015/16; MHSDS 2016/17 - 2017/18; Outpatient 2002/03 – 2017/18; Critical Care 2011/12 – 2017/18, PROMs 2009/10 – 2017/18; Civil Registration (Deaths) 1998/99-2017/18.

This project requires the sensitive PROMs data to measure patient outcomes in secondary care.

The outputs from all of the projects will include peer reviewed papers in academic journals, reports for funders, lay summaries such as newsletters and blogs, and conference and seminar presentations to academic, policy, professional and public audiences. The Centre for Health Economics has a long-established track record in delivery of policy research that utilises HES data, as recognized by the award of the Queens Anniversary Prize in 2007. Examples of recent publications arising from the above projects that have employed the HES data can be found here http://eshcru.ac.uk/publications/index.htm and http://www.york.ac.uk/che/publications/in-house/.

Reports will be produced containing aggregate results that show trends over time, differences across providers, commissioners, geographical areas and by patient subgroups and patient characteristics. The results will contain estimated correlations showing associations between patient outcomes and patient characteristics, hospital, institutional, geographic and environmental factors. Statistical results will be presented in interactive spreadsheets or “Dashboards” (e.g. similar to http://health-inequalities.blogspot.co.uk/ which uses QOF data and only contains aggregated data which can be interrogated), tables and maps of aggregate statistics summarising patient characteristics.

Publications produced in 2017 by project are listed below.

Project 1:

Aragón M.J., Castelli A., Gaughan J., Hospital Trusts productivity in the English NHS: Uncovering possible drivers of productivity variations, PLoS ONE 12(8): e0182253, 2017, https://doi.org/10.1371/journal.pone.0182253

Bojke, C, Castelli, A, Grasic, K, Howdon, DDH, Street, AD & Rodriguez Santana, IDLN 2017 'Productivity of the English NHS: 2014/15 Update' CHE Research Paper, no. 146, Centre for Health Economics, University of York, York, UK, pp. 1-81.

Project 2:

Duarte, A. I., Bojke, C., Cayton, W., Salawu, A., Case, B., Bojke, L. & Richardson, G. A. (2017) Impact of specialist rehabilitation services on hospital length of stay and associated costs, The European Journal of Health Economics https://doi.org/10.1007/s10198-017-0952-0French E, McCauley J, Aragon J, Bakx P,

Chalkley M, Chen S, Christensen BJ, Huang H, Cote-Sergent A, De Nardi M, Echevin D, Fan E, Geoffard G, Gastaldi-Menager C, Gortz M, Ibuka Y, Izumida N, Jones JB, Kallestrup-Lamb M, Karisson M, Klein T, de Lagasnerie G, Michaud P, O’Donnell O, Ohtsu Y, Rice N, Skinner J, van Doorslaer E, Ziebarth NR, Kelly E. End-of-Life Medical Spending in Last Twelve Months of Life is Lower Than Previously Reported. Health Affairs 2017;36(7):1211-1217.

Howdon D, Rice N. Health care expenditures, age, proximity to death and morbidity: implications for an ageing population. Journal of Health Economics, In press (accepted 31 October 2017).

Chalkley M, McCormick B, Anderson R, Aragon MJ, Nessa N, Nicodemo C, et al. Elective hospital admissions: secondary data analysis and modelling with an emphasis on policies to moderate growth. Health Serv Deliv Res 2017;5(7), https://www.journalslibrary.nihr.ac.uk/hsdr/hsdr05070/#/abstract, 2017.

(accepted for publication) Aragon MJ, Chalkley M, How do time trends in in-hospital mortality compare? A retrospective study of England and Scotland over 17 years using administrative data, BMJ Open

Patton, T., Gutacker, N. & Shah, K. 2016. Putting the P back into PROMs - using patient valuations of EQ-5D health states to improve hospital performance comparisons. Health and Quality of Life Outcomes, 15(Suppl 1):185.

Project 3:

Reichert, A. & Jacobs, R. (2017) Socioeconomic inequalities in duration of untreated psychosis: Evidence from administrative data in England, Psychological Medicine, doi: https://doi.org/10.1017/S0033291717002197.

Moran, V. & Jacobs, R. (2017) Investigating the relationship between costs and outcomes for English mental health providers: A bi-variate multi-level regression analysis, The European Journal of Health Economics, doi: 10.1007/s10198-017-0915-5.

Moran, V. & Jacobs, R. (2017) Costs and performance of English mental health providers, The Journal of Mental Health Policy and Economics, 20: 83-94.

Longo, F., Siciliani, L., Gravelle, H., Santos, R. Do hospitals respond to rivals’ quality and efficiency? A spatial panel econometric analysis. Health Economics. 2017; 26(S2) 38-62. DOI:10.1002/hec.356. See also CHE Research Paper 144

Dusheiko, M., Gravelle, H. Choosing and booking – and attending? Impact of an electronic booking system on outpatient referrals and non-attendances. Health Economics. DOI:10.1002/hec.3552. 2017.

Gaughan, J., Gravelle, H., Santos, R., Siciliani, L. Long term care provision, hospital length of stay and discharge destination for hip fracture and stroke patients. International Journal of Health Economics and Management. 2017. DOI 10.1007/s10754-017-9214-z

Gaughan, J., Gravelle, H., Siciliani, L. Delayed discharges and hospital type: evidence from the English NHS. Fiscal Studies. 2017; 38, 5, 495-519. See also CHE Research Paper 133.

Moscelli, G., Gravelle, H., Siciliani, L. The effect of hospital ownership on quality of care: evidence from England. CHE Research Paper 145.

Project 4:

Cookson, R, Mondor, L, Asaria, M, Kringos, DS, Klazinga, NS & Wodchis, WP 2017, 'Primary care and health inequality: Difference-in-difference study comparing England and Ontario' PloS One, 12(11), e0188560. https://doi.org/10.1371/journal.pone.0188560

Moscelli, G., Siciliani, L., Gutacker, N., & Cookson, R. (2017). Socioeconomic inequality of access to healthcare: Does choice explain the gradient? Journal of Health Economics. doi: https://doi.org/10.1016/j.jhealeco.2017.06.005

Barratt, H., M. Asaria, J. Sheringham, P. Stone, R. Raine and R. Cookson (2017). "Dying in hospital: socioeconomic inequality trends in England." Journal of Health Services Research & Policy: 1-6 10.1177/1355819616686807

Sheringham, J., Asaria, M., Barratt, H., Raine, R., & Cookson, R. (2017). Are some areas more equal than others? Socioeconomic inequality in potentially avoidable emergency hospital admissions within English local authority areas. Journal of Health Services Research & Policy. Vol. 22(2) 83–90. http://journals.sagepub.com/doi/abs/10.1177/1355819616679198

Project 5:

Gutacker N, Street AD. Multidimensional Performance Assessment of Public Sector Organisations Using Dominance Criteria. Health Economics,2017. doi: 10.1002/hec.3554.

Gutacker, N. & Street, A. 2017. Use of large-scale HRQoL datasets to generate individualised predictions and inform patients about the likely benefit of surgery, Quality of Life Research, 26(9): 2497-2505.

The outputs from each project will be delivered in accordance with CHE’s funding agreements, which run to different timelines with various milestones for each. The key milestones and timelines for each project are:

Project 1 - The primary output from this ongoing project is the production each year of an annual update to national NHS productivity figures that incorporates the most recent financial year of data. Under this project, CHE has demonstrated that NHS productivity growth is meeting the requirements of the Five Year Forward View and outpaces that of the economy as a whole. CHE’s figures are widely used to inform policy discourse, with the DHSC relying on the information for internal monitoring purposes and for external reporting and response purposes, such as to inform annual Spending Reviews. Under this project, CHE also provides data about the quality of NHS care to the Office of National Statistics that are used in the construction of the national accounts.

In addition to the annual update of national figures, CHE also undertakes analyses of variation in hospital productivity and produces short reports or memorandum for the DHSC to address specific questions about NHS productivity. CHE presents the work regularly to various audiences, including politicians, policy makers, academics, health professions and the general public through seminars, conference presentations and media appearances.

Project 2 - This project will produce a range of outputs, including reports to support policy decisions and peer reviewed publications. Where appropriate, analysis will also be disseminated to national and local decision makers at formal and informal meetings, including strategic commissioning groups.

In 2018/19 CHE will submit the results of the commissioning care hubs research to a peer reviewed journal.

Project 3 - During 2018/19, CHE will produce several papers on mental health funding and mental health waiting times; papers on quality of small hospitals, effects on patients of hospital closure, competition and quality in general practice, the effect of competition on hospital waiting times, and waiting time inequalities across and within hospitals.

Project 4 - During 2018/19, CHE expect to carry out extensive analyses with a report, co authored with NHS England Equality and Health Inequalities Team, on ‘Why some areas do better than others at reducing health inequalities’, to come out later in the year and further outputs to follow in 2019. Preparatory work for the Wellcome Trust project will be conducted for quasi-experimental evaluations of policy impacts on health equity, with published outputs not expected until 2019.

Project 5 - By early 2019, CHE will publish journal articles and working papers, and produce a number of final project reports for the DHSC.

Further updates on outputs are set out below;

Project 1:

Aragón M.J., Castelli A., Gaughan J., Hospital Trusts productivity in the English NHS: Uncovering possible drivers of productivity variations, PLoS ONE 12(8): e0182253, 2017, https://doi.org/10.1371/journal.pone.0182253

Bojke, C, Castelli, A, Grasic, K, Howdon, DDH, Street, AD & Rodriguez Santana, IDLN 2017 'Productivity of the English NHS: 2014/15 Update' CHE Research Paper, no. 146, Centre for Health Economics, University of York, York, UK, pp. 1-81.

Project 2:

Duarte, A. I., Bojke, C., Cayton, W., Salawu, A., Case, B., Bojke, L. & Richardson, G. A. (2017) Impact of specialist rehabilitation services on hospital length of stay and associated costs, The European Journal of Health Economics https://doi.org/10.1007/s10198-017-0952-0French E, McCauley J, Aragon J, Bakx P,

Chalkley M, Chen S, Christensen BJ, Huang H, Cote-Sergent A, De Nardi M, Echevin D, Fan E, Geoffard G, Gastaldi-Menager C, Gortz M, Ibuka Y, Izumida N, Jones JB, Kallestrup-Lamb M, Karisson M, Klein T, de Lagasnerie G, Michaud P, O’Donnell O, Ohtsu Y, Rice N, Skinner J, van Doorslaer E, Ziebarth NR, Kelly E. End-of-Life Medical Spending in Last Twelve Months of Life is Lower Than Previously Reported. Health Affairs 2017;36(7):1211-1217.

Howdon D, Rice N. Health care expenditures, age, proximity to death and morbidity: implications for an ageing population. Journal of Health Economics, In press (accepted 31 October 2017).

Chalkley M, McCormick B, Anderson R, Aragon MJ, Nessa N, Nicodemo C, et al. Elective hospital admissions: secondary data analysis and modelling with an emphasis on policies to moderate growth. Health Serv Deliv Res 2017;5(7), https://www.journalslibrary.nihr.ac.uk/hsdr/hsdr05070/#/abstract, 2017.

(accepted for publication) Aragon MJ, Chalkley M, How do time trends in in-hospital mortality compare? A retrospective study of England and Scotland over 17 years using administrative data, BMJ Open

Patton, T., Gutacker, N. & Shah, K. 2016. Putting the P back into PROMs - using patient valuations of EQ-5D health states to improve hospital performance comparisons. Health and Quality of Life Outcomes, 15(Suppl 1):185.

Project 3:

Reichert, A. & Jacobs, R. (2017) Socioeconomic inequalities in duration of untreated psychosis: Evidence from administrative data in England, Psychological Medicine, doi: https://doi.org/10.1017/S0033291717002197.

Moran, V. & Jacobs, R. (2017) Investigating the relationship between costs and outcomes for English mental health providers: A bi-variate multi-level regression analysis, The European Journal of Health Economics, doi: 10.1007/s10198-017-0915-5.

Moran, V. & Jacobs, R. (2017) Costs and performance of English mental health providers, The Journal of Mental Health Policy and Economics, 20: 83-94.

Longo, F., Siciliani, L., Gravelle, H., Santos, R. Do hospitals respond to rivals’ quality and efficiency? A spatial panel econometric analysis. Health Economics. 2017; 26(S2) 38-62. DOI:10.1002/hec.356. See also CHE Research Paper 144

Dusheiko, M., Gravelle, H. Choosing and booking – and attending? Impact of an electronic booking system on outpatient referrals and non-attendances. Health Economics. DOI:10.1002/hec.3552. 2017.

Gaughan, J., Gravelle, H., Santos, R., Siciliani, L. Long term care provision, hospital length of stay and discharge destination for hip fracture and stroke patients. International Journal of Health Economics and Management. 2017. DOI 10.1007/s10754-017-9214-z

Gaughan, J., Gravelle, H., Siciliani, L. Delayed discharges and hospital type: evidence from the English NHS. Fiscal Studies. 2017; 38, 5, 495-519. See also CHE Research Paper 133.

Moscelli, G., Gravelle, H., Siciliani, L. The effect of hospital ownership on quality of care: evidence from England. CHE Research Paper 145.

Project 4:

Cookson, R, Mondor, L, Asaria, M, Kringos, DS, Klazinga, NS & Wodchis, WP 2017, 'Primary care and health inequality: Difference-in-difference study comparing England and Ontario' PloS One, 12(11), e0188560. https://doi.org/10.1371/journal.pone.0188560

Moscelli, G., Siciliani, L., Gutacker, N., & Cookson, R. (2017). Socioeconomic inequality of access to healthcare: Does choice explain the gradient? Journal of Health Economics. doi: https://doi.org/10.1016/j.jhealeco.2017.06.005

Barratt, H., M. Asaria, J. Sheringham, P. Stone, R. Raine and R. Cookson (2017). "Dying in hospital: socioeconomic inequality trends in England." Journal of Health Services Research & Policy: 1-6 10.1177/1355819616686807

Sheringham, J., Asaria, M., Barratt, H., Raine, R., & Cookson, R. (2017). Are some areas more equal than others? Socioeconomic inequality in potentially avoidable emergency hospital admissions within English local authority areas. Journal of Health Services Research & Policy. Vol. 22(2) 83–90. http://journals.sagepub.com/doi/abs/10.1177/1355819616679198

Project 5:

Gutacker N, Street AD. Multidimensional Performance Assessment of Public Sector Organisations Using Dominance Criteria. Health Economics,2017. doi: 10.1002/hec.3554.

Gutacker, N. & Street, A. 2017. Use of large-scale HRQoL datasets to generate individualised predictions and inform patients about the likely benefit of surgery, Quality of Life Research, 26(9): 2497-2505.

Presentations given in 2017 by project:

Project 1

Presentation: Thinking about Selection and Productivity, DREAMS Workshop, University of York, December 2017 by Castelli A.

Project 2

Presentation: The role of EQ-5D value sets based on patient preferences in the context of hospital choice in the national PROM programme in England. 34rd EuroQol Group Scientific Plenary, Barcelona, Germany, Sep 2017.

Presentation: Putting the P back in PROMs. PROMs Research Conference, Oxford, UK, July 2017.

Project 3

Presentation: Quality in primary care: is it possible to identify peer effects among English GP practices? (work with Hugh Gravelle and Nigel Rice), Health Economics seminar series, Manchester, 22nd May 2017; Second Spatial Econometrics Advanced Institute Alumni Reunion, Rome, 26th May 2017 and Centre of Expertise Healthwise, Faculty of Economics and Business, Groningen, 1st June 2017.

Project 5

Presentation: Gaughan, J, Grasic, K, Gutacker, N, Siciliani, L, & Street, AD. The effects of Best Practice Tariffs on the provision of elective and emergency day cases. Health Economists' Study Group, Birmingham. 4-6 January 2017.

Presentation: Liu, D, Kasteridis, P, Goddard, M, Jacobs, R, Wittenberg, R, & Mason, A. Mind the gap! The impact of incentives to reduce underdiagnosis in people with dementia. Health Economists' Study Group, Birmingham. 4-6 January 2017.

Presentation: Moscelli, G, Jacobs, R, Gutacker, N, Mason, A, Aragon, MJ, Chalkley, M & Boehnke, J. Clustering of mental health patients for payment: does the hospital matter? Evidence from English mental health providers. International Health Policy Conference. LSE, London, 16-19 February 2017.

Presentation: Gaughan J, Kasteridis P, Mason A, Street A. Waits in A&E Departments of the English NHS. Health Economists' Study Group, Aberdeen, 28-30 June 2017.

Project 1 - The primary output from this ongoing project is the production each year of an annual update to national NHS productivity figures that incorporates the most recent financial year of data. In 2017, the annual update was produced for the DHSC and, as in other years, was used by them externally and internally in monitoring, informing policy debate, the annual spending review and negotiations on budget setting. Under this project, CHE also provided data about the quality of NHS care to the Office of National Statistics that are used each year in the construction of the national accounts. In 2017, additional analyses on hospital level productivity were produced for the DHSC, examining the factors underlying variation in productivity, which assists the DHSC in exploring how to get the best value from NHS resources.

Project 2 - This project has produced a range of evidence that allows the DHSC and other organisations to understand and plan expenditure on different aspects of NHS care, to account for changes in activity and also to understand aspects of quality of care, First, it has investigated the drivers of health care expenditure, disentangling the influence of age, morbidity and proximity to death on the level of expenditure. This has important implications for the DHSC in terms of planning and for the design of the resource allocation formulae used to distribute the healthcare budget, Second, it has explored several aspects of activity in the secondary care sector, for instance, policies for moderating growth in elective admissions and how the provision of specialist rehabilitation services affect the duration and costs of hospital care. It has also looked at the use of patient valuation of the care they receive as a way of measuring the quality of hospital care. Third, it has investigated in-hospital mortality trends over time. These insights allow policy makers to appreciate the costs and the benefits of NHS provision and to plan for more effective and efficient care.

Project 3 - The research in this project has evaluated whether the way in which the NHS is organised can affect the costs and outcomes of services to help inform policy decisions. In 2017, the research has produced evidence on a number of issues in mental health, including the relationship between costs and quality and socio-economic inequalities in access to care. Other examples include the production of evidence on the impact of the electronic booking system on referrals and non-attendances and the investigation of reasons for delayed discharges from hospital, both of which are important in terms of planning future organisation of care.

Project 4 - This project has continued to investigate inequities in access, costs and outcomes and in 2017, further evidence relating to access, avoidable emergency hospital admissions and in-hospital mortality has been produced. This has added to knowledge of how to improve health and healthcare for vulnerable populations.

Project 5 – In 2017, previous research on patient-assessed outcomes was extended (by working with Vale of York CCG) to generate a web tool to support discussions between patients and their GPs about whether to undergo planned surgery. The on-line tool, aftermysugery.org.uk can be used by patients and their GPs, who input basic demographic data and fill in a pre-operative health status questionnaire. The webtool then returns a predicted post-operative health status, together with national comparator data, displayed in various visual formats. This information is designed to a) help patients decide whether they feel the expected health improvement is sufficiently high to make having the operation worthwhile, b) inform patients about the likelihood of a negative outcome, and c) provide information about which hospitals secure better outcomes for their patients.

Other strands of the research have established that incentive payments made to GPs to increase the numbers of individuals diagnosed with dementia have been effective, but that there may have been unintended consequences on patient experience and access.