CUREd+ Centre for Urgent and Emergency Care Research Database

The University of Sheffield requires access to NHS England data for the purpose of the following research programme:

Centre for Urgent and Emergency Care Research Database Refresh (CUREd+)

The following is a summary of the aims of the research programme provided by or on behalf of the University of Sheffield:

In England, demand for urgent and emergency care (UEC) services has increased significantly over the past two decades, partly due to an ageing population, but also due to service users attending services which are not the most appropriate for their health needs. This increase in demand has impacted on patient flow, patient safety, effective discharge, response times and waiting times within the UEC services. To this end, research into the UEC system is vital to help improve these areas as well as relieving pressure and demand on the emergency departments, improving the standards and quality of care, and delivering an improved experience for patients and staff.

Previous work by the University of Sheffield (UoS) led to the creation of a large, unique database, titled the Centre for Urgent and Emergency Care database (“CUREd”) Research Database. This database contained NHS data from a number of UEC providers including Yorkshire Ambulance Service (YAS), NHS hospital Trusts, and Sheffield Health and Social Care NHS Foundation Trust between 2011 and 2017. Using NHS England data, the Centre for Urgent and Emergency Care Research (CURE) group now wish to create a new and improved database, the CUREd+ database, to ensure that the data held reflects changing practices and continues to be useful for analyses. The new CUREd+ research database is a longitudinal linked database that will expand the hospital data to cover all of England, update the linked ambulance service data, which will now include electronic patient records data (supplied by YAS), add death registration data, reduce variation of type and format within the hospital data, enabling more robust research analysis, reduce the amount of patient identifiable data held and processed by the UoS, and enable further research on a number of UEC-related topics. It has also become clear that there is a need for more in-depth research into the use of the UEC system. To this end, the CUREd+ research database will provide access to data extracts for use in research projects aligning with one or more of the following research aims:

> Research Aim 1: Identifying and characterising the pathways of specific cohorts of patients prior to joining the UEC system and their journey through the UEC system;

> Research Aim 2: Identifying and characterising factors which influence when, how and why individuals within specific cohorts of patients join the UEC system in the first place;

> Research Aim 3: Analysing performance indicators and measures of quality of care, such as response times, waiting times and risk-adjusted mortality rates;

> Research Aim 4: Investigating methods for managing demand for urgent care, such as NHS 111 other interventions aiming to reduce avoidable attendances and admissions

> Research Aim 5: Analysing the use of the urgent care workforce, including nurse practitioners and advanced paramedic practitioners.

Some examples of planned and possible projects that will make use of the NHS England data contained within the CUREd+ database, and aligning with one or more of these research aims, are outlined below:

> Project example 1- Avoidable conveyances to hospital. Identifying and characterising 999 calls suitable for care closer to home

Demand for the ambulance service is rising every year, however very little is known about the factors that are driving this continuous increase in demand. To ensure that ambulance services can continue to operate a safe and effective service for patients, it is important not only to understand this demand, but work on methods that could reduce the number of ambulances dispatched to incidents, as well as the number of ‘low-acuity’ (semi-urgent/non-urgent) patients who are conveyed to Emergency Departments (ED).

> Project example 2 - Avoidable ED attendances: Identifying and characterising ED attendances that could reasonably be managed in alternative settings

The pressures of patient demand on UEC services continue to be reported in England, evidenced by declining performance of Emergency Departments against the national four-hour performance target and increased crowding, evidenced by higher numbers of 12-hour trolley waits and diversions of ambulances. The ambulance service is also under considerable pressure with calls to the ambulance service doubling in the last 10 years.

It has been reported for some time that a proportion of attendances to EDs are amenable to management in settings providing a lower level of care such as primary care, walk-in centres and urgent care centres. These attendances (variously described as non-urgent, avoidable or inappropriate) are an indicator of emergency and urgent care systems that could perform better, particularly if such attendances are conveyed by ambulance.

> Project example 3 - Measuring ED performance and solutions for improvement

Another method for dealing with high levels of demand for services and overcrowded EDs while ensuring efficiency and high quality is for the services to be performance managed. However, when such measures are introduced, effective service improvements can only be achieved if the performance measures that are being used to assess performance are suitable to the task and have been shown to correspond with an improvement of care.

By evaluating current and potential performance measures, as well as creating, examining, and comparing new ones, it would become easier to assess EDs in the country and highlight areas that need improvement, as well as creating benchmarks for measuring interventions aimed at improving ED services.

> Project example 4 - Identifying and characterising avoidable admission to hospital

Similar to the rising rates of ED admissions, and the proportion of these that are considered avoidable, and the pressure this puts on EDs, there has also been a year-on-year increase in emergency hospital admissions, growing by roughly 40% in the decade. While a lot of these admissions are complex cases that require long stays in hospital, particularly for older patients with multiple health conditions, around a third of these admissions result in zero-day stays, where a patient does not stay overnight.

Being able to distinguish between complex admissions, necessary zero-day stays, and avoidable admissions would help reduce the pressures that are currently being felt by acute hospitals and could help in them deliver more reliable elective care, lower the constantly high bed occupancy rates, and reduce the number of cancelled or postponed elective procedures.

> Project example 5 - Assessing the impact of the pandemic on UEC use and identifying strategies for recovery and resilience

The COVID-19 pandemic is an ongoing severe global health challenge, the impact of which had an immediate effect and will continue to impact health, as well as health care services and systems for a long time to come. During this time, EDs went through rapid change in order to cope with the shift in priorities, enacting changes much quicker than safeguards would normally allow.

> Project example 6 – Identifying and characterising households containing frequent users of ED

Understanding patterns of attendance, and being able to identify unusual patterns, has implications for safeguarding of children and for identifying families where interventions to support childcare may be particularly appropriate. Children in Yorkshire and Humber who are frequent users of ED can be selected within the CUREd+ database based on ED attendance records, and records relating to their household contacts found using the household grouping information provided by pseudonymised UPRN data. It will then be possible to look at the ED attendance numbers, reasons and timing for household contacts of frequent ED users and understand how patterns of attendance cluster within households/families.

The CUREd+ research database will present a unique and powerful resource enabling in-depth research into aspects of the UEC system and how it is used. It will also allow researchers to assess data quality, completeness, relevance and volumes of data prior to and during undertaking research analysis, as well as developing quality indicators in each area of research, to inform the measurement of variation and identification of best practice.

Two cohorts of patients will be defined, for which varying data on individual patient episodes of care between 1st April 2011 and 31st March 2023 will be collected.

[Cohort A] – Cohort identified by the University of Sheffield and YAS; includes patients who:

1) contacted or received care from the emergency ambulance service provided by Yorkshire Ambulance Service (YAS) NHS Trust, or

2) contacted the NHS 111 telephone triage service provided by YAS

[Cohort B] – Cohort defined by NHS England; includes patients who:

1) received unscheduled care at a Walk-in Centre, Minor Injuries Unit, Urgent Care Centre or Emergency Department in England, or,

2) received inpatient or outpatient NHS hospital care in England, or

3) received care from Mental Health Services in England

This Agreement permits the University of Sheffield to use the data for the purposes of projects aligning with the research aims described above, and which are conceived, planned, approved and initiated through the following process:

Projects will be conceived and planned by project teams, including a Chief/Principal Investigator who is member of the University of Sheffield School of Health and Related Research (ScHARR). It will be the project teams’ responsibility to ensure that:

1. Projects have clearly defined objectives and operational plans;

2. The aims of the project align with at least one of the CUREd+ research aims (as stated above), and achieving those aims through one or more of the methods listed above;

3. An analysis plan is prepared for each project, setting out the data requirements and methods;

4. In each case, the use of the data is necessary and proportionate to the purpose of the project and that the minimum amount of data necessary is used - this will include consideration of the necessity for use of each individual dataset, and field; the number of years of data; the sizes of any cohorts or control cohorts derived from the data, and the inclusion and exclusion criteria (such as presence of specific diagnostic or procedure codes);

5. Appropriate safeguards are in place to protect confidentiality; minimise risks of re-identification and use of excessive data beyond necessity;

6. The necessary ethics are in place for the project from the UoS University Research Ethics Committee.

A CUREd+ Data Access Request Form (DARF) will be completed and submitted to the CUREd+ Data Release Committee (DRC). The DRC will be made up of a Data Release Chair, a Data Release Deputy, a UEC (Emergency Department) representative, a UEC (Ambulance Service/NHS111) representative, a YAS representative, a ScHARR representative, a ScHARR Information Governance Representative, a UoS Representative, and at least two public contributors. The DRC provides a forum for the discussion, in depth and expert assessment and recommendation of project ideas, drawing on expertise and knowledge focused on both the projects aims, objectives, and methodology, as well as information governance, and data minimisation and anonymisation. The DRC are responsible for receiving a complete DARF, and that this includes the relevant data showing that every proposed project:

1. Aligns with the strategic aims of CUREd+ and has original objectives;

2. Is methodologically sound;

3. Has the necessary study governance (ethical approval, funding, evidence of Information Governance Training);

4. Has given sufficient due thought to data minimisation;

5. Has given sufficient due thought to data anonymisation (for publishing purposes)

All co-applicants who wish to access the data are personnel at the UoS (employees, agents, and visiting academic staff members), or postgraduate students supervised by University of Sheffield personnel. No individuals from other organisation (not University of Sheffield personnel) can request the data via this Data Access Request facility.

The CUREd+ DRC will convene to consider applications for data extracts from the CUREd+ research database. Members of the DRC involved in a study making an application will be excluded from making a decision on that application and will not be present for deliberations on that application. The DRC will recommend projects for approval based on the criteria listed above. Only projects that are recommended for approval by the DRC will be progressed. The likely number of successful applications to the DRC is expected to be around ten in the first year, and six to eight per year in following years.

Each project that is reviewed and approved by the DRC will be defined and bound by the information provided in the DARF, including an analysis plan detailing what data is permitted for use in the project and how it shall be processed, how it shall be stored and accessed and not removed from the secure data environment, and that no attempts should be made to re-identify the individuals within the data.

Experienced data management personnel will prepare the approved projects data extracts following a strict protocol for effective de-identification, including hashing of pseudo-identifier fields, and limiting fields and datasets to those stipulated in the project DARF. These extracts will be transferred onto a separate secure environment (on a per-project basis) to which the project researchers will be granted access to carry out their analysis without accessing the main health database.

A publicly available asset register of all the extracts that have been approved by the DRC is created by the data management team. This will include the full title of each project, the name of the CI/PI, the sponsor, and the date of the approval by the DRC and the date the extract was created.

The following NHS England data will be accessed:

> Hospital Episode Statistics

- Admitted Patient Care – necessary to identify patients admitted to hospital after a 999-phone call, an NHS 111 phone call, or an ED attendance etc. University of Sheffield will also examine the inpatient care received to identify potentially avoidable hospital admissions from ED. University of Sheffield will additionally use this dataset to create patient cohorts (e.g., patient groups who have previously received hospital care for specific conditions or undergone specific procedures) to understand different cohorts' usage of the UEC system. Admitted Patient Care activity is also an important factor in health economic evaluations (e.g., health outcomes such as re-admission, bed days, and economic costs).

- Accident & Emergency and Emergency Care Data Set (ECDS) – necessary to look at numerous factors that affect the UEC system. Examples include looking at the frequency and timing of admissions to Emergency Departments (ED), and the variation that can be seen depending on time of day, day of the week (weekday vs. weekend), and geographical location, while also taking into consideration local services that are also available. These outcomes will also be used to investigate the impact that the COVID-19 pandemic has had on services.

- Critical Care – necessary to look at the whole spectrum of care, specifically seriously ill patients. Looking at the whole journey for patients who enter critical care, e.g., care standards in ED for head injury or sepsis patients later admitted to critical care – to what extent could care be improved early on in order to have an impact on later patterns of critical care usage for these patients?

- Outpatients – necessary to enable a fuller picture of patients’ health care contacts. This will enable the identification of cohorts (i.e., attendance at outpatient services under one or more specialties) and outpatient activity (and associated costs) which are important in health economic assessments when evaluating changes to the UEC system.

> Mental Health – necessary to look at frequent users of the UEC network, (many of whom are affected by mental health conditions) and understand the contacts patients diagnosed with one or more mental health issues have with UEC services. Mental health services data will also allow the investigation of variation in delivery across services to identify good practice. This will be used to investigate potential areas where improvement could be done to better patient care across England.

> Civil Registration Mortality – necessary because selecting/omitting specific causes of death or disease that were present would allow more precise cohorts to be chosen for each project and are also key outputs when studying any system change or comparing/creating performance measures. Mortality is a key outcome for projects examining UEC services.

> Medicines dispensed in Primary Care (NHSBSA data) – Necessary to enable identification of cohorts of patients (e.g., identifying patients with asthma) and stratification of patients into various, more detailed risk groups. Dataset to be disseminated for cohort A only. This data will enable investigation of the extent to which medications are leading to an increase or decrease in the use of Urgent and Emergency Care (UEC): looking at the service use patterns of patients on different treatment plans to see if correctly medicating patients is effective in e.g., reducing ED use; if patients’ medication is safe and effective, this may reduce their use of UEC services such as the ambulance service and ED, by avoiding escalations of their condition etc. If the medication or combination of medications they are taking is not effective, or has safety issues that cause other problems, then some types of UEC service use may increase. Linking medicines data to UEC service use data therefore enables assessment of the safety and effectiveness of different medications, through the impact they may have on use of UEC services by patients who are taking those medications.

> Demographics – necessary to link data supplied by the original CUREd project, YAS and NHS England. The Demographics dataset will be used as a Single Source of Truth (SSoT) table for understanding and identifying the union of all patients present in CUREd+. This will enable different cohorts to be identified in the data, for example split by age and gender, perform case-mix adjustments, as well as allowing projects to produce descriptive statistics for their chosen cohort. Two demographics datasets have been requested; one for cohort B (whole of England), including all requested demographics pseudonymised data and excluding the identifiable address data, and another for cohort A, which contains both the pseudonymised data and the identifiable address data.

All data associated with the patient record of care, for example presenting complaint, investigations, treatments, disposal /discharge, diagnoses and disposal status (alive /dead) are key to the main objectives of the research database which is to build an accurate picture of patients accessing the different urgent and emergency care (UEC) pathways. By detailing patterns of access and use by different clinical groups, researchers will be able to determine which patients should be the focus of interventions to reduce inappropriate use of services.

The level of the data will be:

> Identifiable – For Demographics data (provided for cohort A only), full address and postcode data for current and historic addresses have been requested. This data is required to enable grouping of patients who live in the same household and for identifying residence types (e.g., care homes, houses of multiple occupation, flats, etc.). Identifying care homes within the CUREd+ research database enables exploration of how residents of care homes are using the UEC system, investigation of variation between care homes and identification of hotspots of good practice. Addresses will be linked to Unique Property Reference Numbers (UPRNs). These UPRNs will then be pseudonymised (using a hashing algorithm) and only pseudonymised UPRNs will be made available in the CUREd+ database. Identifying patients who share a household is essential for investigating the clustering of frequent ED attendance within households/families. Repeated ED attendance of children can be an indicator of difficulty managing illness within a family.

Sensitive fields required:

> Date of Death (Civil Registration of Deaths): required to generate derived binary indicator variables for whether death occurred within a specified timeframe from an index event, e.g., death within 30 days of first contact with the ambulance service; or seven days of first presentation at A&E.

> General practice codes (Demographics): used with medicines dispensed data to look at patterns of ED and hospital attendances that could have been avoided. An example is the care of patients with epilepsy, in which GPs play a prominent role. The linkage of medicines, emergency and urgent care data, provides data for a natural experiment to test the effects of expert medical generalism on healthcare quality.

> Fields on self-harm, and restrictive (coercive) intervention and the Mental Health Act (Mental Health): important for identifying subgroups who are more likely to use UEC services, which tend to be recurrent and predictive of suicide attempts, and severity of illness as well as the likelihood of experiencing a mental health crisis. Analysis would also look at how coercive care is experienced unequally and explore the ethnic variation in use of UEC among people with serious mental illness. These fields will also be used to investigate patient cohorts with identified severe mental health conditions and their use of the UEC care pathway.

> Census Output Areas (HES A&E): used for segmentation into geodemographic classification and facilitate creation of further explanatory variables such as travel distance and time to hospital, and accessibility (travel distance/time) of other health services.

> A&E staff member codes (HES A&E): enable projects to look at variability in patient pathways by staffing

> Fields under the psychiatric category will be used in conjunction with other requested datasets to enable a fuller picture of a patient's health care contacts, such as with the Mental Health Services Data set. This will enable the identification of cohorts (i.e., patients with severe mental health issues) and their use and journey in the UEC system. They can also enable the examination of stratified outcomes for mental health patients. Fields that indicate if a patient was admitted and subjected to detention and restrictive practices will be used as an indicator for emergency and often life-saving treatment as well as information on the severity of mental health problems the patient has.

The data will be minimised as follows:

> Two cohorts of patients will be defined, cohort A (Cohort identified by the University of Sheffield and YAS) and cohort B (Cohort defined by NHS England) for which varying data on individual patient episodes of care between 1st April 2011 and 31st March 2023 will be collected.

Research will focus on specific cohorts of patients (by gender, age, presenting complaint, etc.), pathways through the UEC services, and trends over time-of-service use.

> Limited to data between April 2011 – March 2023; this is to match up with the YAS data that has already been received (2011-2017) and the new YAS data that will be requested (2017-2023). This will allow any analysis that was conducted on this original data (2011-2017) data to be evaluated and compared on the CUREd+ research database which will include more recent data and cover a larger geographical area. This will bring the CUREd+ research database up to date and ensure any results will be as informed and relevant as possible. Projects that want to inspect and incorporate the impact of the Covid-19 pandemic on the UEC system (e.g., change in use, strategies for recovery and resilience), would also require data from before the pandemic to study and compare with data collected during the pandemic.

> Data for all of England is required to allow analyses of the UEC system as a whole and the examination of regional variances, including examples of good practice that could be important to learn, adapt and roll out nationally. This will allow analysis of factors that influence when, how and why individuals join the UEC system, which will vary dramatically due to the services that are available in the individuals’ vicinity, and factors such as distance to the services.

NHS 111 data for the cohort A is required as the NHS 111 service is part of the UEC system.

> A comprehensive decision-making process has been undertaken to ensure only the minimum datasets and individual fields have been requested that are essential for the creation of the research database. All stakeholders in the CUREd+ research database were consulted in selecting the fields, with multiple meetings held to discuss the need and usage of the datasets and the fields requested.

> The data requested cannot be filtered by patients’ episodes as an overarching aim of the research database project is to analyse current performance measures used in ED service (which include outcomes further in the patient journey) and create new performance measures. To fully enable this, all patient episodes (elective, maternity, neonatal etc.) are needed to ensure that EDs have been evaluated on how they are managed and function, allowing the best possible feedback that development could be based on.

After project approval by the DRC, the minimisation per use will be reviewed and approved by experienced data management personnel, following a strict protocol for effective de-identification, including hashing of pseudo-identifier fields, and limiting fields and datasets to those stipulated in the project DARF.

The University of Sheffield is the controller as the organisation responsible for ensuring that the data will only be processed for the purpose described above.

The lawful basis for processing personal data under the UK GDPR is:

Article 6(1)(e) - processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller.

The lawful basis for processing special category data under the UK GDPR is:

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.

This processing is in the public interest because the CUREd+ research database will enable many research projects that will all aim to have an impact and positive outcome for the operation and improvement of the UEC system.

The funding is provided by National Institute for Health and Care Research (NIHR). The funding is for the programme of work and is not specifically limited to the project described.

Amazon Web Services (AWS) provides IT hosting services to the University of Sheffield and will store the data as contracted by the University of Sheffield.

The CUREd+ Data Release Committee (DRC) will act in an advisory capacity to advise, alongside the UoS, if and how the data will be used for projects in support of those priorities and programmes. Some projects may involve collaboration with individuals or organisations outside of the UoS, however in all cases, the University of Sheffield will retain sole autonomy for determining if and how the data under this Agreement will be processed and the UoS cannot be compelled by any third party to process the data for any purpose. The data will only ever be used for purposes that directly support the priorities of the CURE group as described in this Agreement, and only personnel stated below will have access to patient level or derived data.

Data will be accessed by:

> Substantive employees of the University of Sheffield

> Postgraduate students affiliated with the University of Sheffield. Any student working with the data held under this Agreement must have completed relevant data protection and confidentiality training and is subject to the University of Sheffield’s policies on data protection and confidentiality. Any students accessing the data will do so under the supervision of a substantive employee of the University of Sheffield. The University of Sheffield will be responsible and liable for any work carried out by students. These students will only work on the data for the purposes described in this Agreement. The likely number of postgraduate student projects using CUREd+ data is expected to be around four per year.

> Visiting Academic staff members to the University of Sheffield. Only individuals who are subject to the same policies, procedures and sanctions as substantive employees will have access to the data and only for the purposes described in this agreement.

Patient and Public Involvement and Engagement:

The UoS study team have actively involved Patient and Public Involvement and Engagement (PPIE) support at every stage of decision-making processes for the design of the CUREd+ database.

The Sheffield Emergency Care Forum (SECF) have been actively involved and supportive throughout the development of the existing CUREd dataset and its policies. The SECF offers formal PPIE support to researchers in order to help them ensure their research is relevant to their target audience. Their special interest is in emergency care, and the group have long-standing collaborations with The University of Sheffield (UoS) offering independent and impartial general public and patient perspectives on research proposals, patient information sheets and lay person summaries. The CUREd+ database project was discussed with the SECF at an early stage, at the December 2021 SECF meeting, which was attended by a representative of the study team who detailed the upcoming database project and sought feedback and support from the forum.

During development of the CUREd+ DARS application, the study team held three online focus groups in June 2022 with members of the public, to gain more insight into the public's perception and views around the use of sensitive data. Some of the topics discussed were:

> Public opinion on researchers using confidential patient information without consent and opinions on the opt-out process as a whole, and

> Public opinion on the use of sensitive data within research and specifically the CUREd+ database.

The focus groups comprised a lay audience with a range of ages (20-65), genders and residency around the country. These focus groups were very successful and discussions around the opt-out process and using patient information without consent were interesting. All individuals across the three groups agreed that there was good reason to use health data in research and were happy for their data to be included as it could benefit the NHS and services in the future.

The CUREd+ Data Release Committee (which oversees the management of the database, its policies and any data extracts produced using it) includes ongoing PPIE representation, to maintain patient and public involvement throughout every stage and aspect of establishing and running the database. The DRC will review new applications to use the database as well as review the running and general management of the research database, including any proposed amendments to policy, incidents and access rights for the research database. The DRC, including its PPIE representation, will ensure that the database is being driven in the correct direction, to deliver quality, high-impact results, whilst maintaining sound ethical and governance standards.

There is also PPIE input into specific projects using data from the CUREd database: for example, individual studies will often have PPIE representatives within the co-applicant team, and may recruit a larger PPIE reference group of users and carers with relevant experience of speci

University of Sheffield (cohort A1) and the Yorkshire Ambulance Service (YAS) (cohort A2) will transfer data to NHS England, which combined, will form cohort A. The data will consist of identifying details (specifically NHS Number, Date of Birth, Postcode, Forename, Surname, unique person ID) for the cohort to be linked with NHS England data.

NHS England data will provide the relevant records from the HES, ECDS, Mental Health, Death, Demographics and NHSBSA datasets to the University of Sheffield.

> Demographic data for cohort A will contain directly identifying data items including latest address and postcode, which are required to access information about the type of residence.

> All other data will contain no other direct identifying data items but will contain a unique person ID which can be used to link the data with other record level data already held by the recipient. Cohort A (YAS cohort) will also have a study ID unique to each individual. Cohort B (Whole of England cohort) will only have unique person ID. A mapping table will be disseminated to the University of Sheffield to enable mapping of individuals in cohort A - this does not risk identifiability.

For cohort A, NHS England will supply data from:

i. Medicines Dispensed in Primary Care data (NHSBSA),

ii. Demographic (including address data), and

iii. Civil Registration – death data

For cohort B, NHS England will supply data from:

i. Hospital Episode Statistics (HES);

1. Emergency Care Data Set (ECDS)

2. Accident & Emergency (A&E)

3. Outpatient (OP)

4. Admitted Patient Care (APC)

ii. Mental Health Services Data Set (MHSDS)

iii. Demographic, and

iv. Civil Registration – death data

Cohort A is largely a subset of Cohort B. The datasets provided for Cohort B are therefore expected to include data for patients identified in Cohort A who will have used the services in Cohort B. Demographics and Deaths data is also required separately for cohort A to capture demographic and death data on YAS patients who have used no other services besides NHS111 or the ambulance service, and would therefore not appear in cohort B.

Experienced data management personnel prepare approved projects’ data extracts following a strict protocol for effective de-identification and minimisations, as stipulated in the project DARF. These extracts are transferred onto a separate secure environment (on a per-project basis) to which the project researchers will be granted access to carry out their analysis without accessing the main health database.

The data will not be transferred to any other location.

The University of Sheffield stores data on the Cloud provided by AWS.

The data will be accessed by authorised personnel via remote access using the University of Sheffield VPN only. The data will remain within the University of Sheffield secure computing environment at all times.

Personnel are prohibited from downloading or copying data to local devices.

The data will not leave England/Wales at any time.

Access is restricted to employees, agents, and Visiting Academic staff members of the University of Sheffield, and postgraduate students supervised by University of Sheffield personnel, who have authorisation from the DRC. Non-substantive employees of the University of Sheffield accessing the data are subject to the same information governance framework as UoS employees and would be required to meet the level required to access the secure computing environment.

Access to confidential patient identifiable data is restricted to select employees or agents of the University of Sheffield. Select University of Sheffield data management staff will produce subsets of the data that will be accessed by individuals who have had their project approved by the DRC. Access to the confidential patient identifiable data will be restricted to substantive employees or agents of the University of Sheffield.

All personnel accessing the data have been appropriately trained in data protection and confidentiality.

The data will be linked at person record level with data obtained from the original CUREd and Yorkshire Ambulance Service (YAS) data using study ID. The original CUREd database consists of routine clinical ambulance data and includes CAD and NHS111 data from April 2011 - March 2017 (inclusive). The YAS data consists of routine clinical data and includes electronic patient records (ePR), computer aided dispatch (CAD), and NHS111 data from April 2017 to present.

Mental health data contained within the original CUREd database, originally obtained from Sheffield Health and Social Care NHS Foundation Trust, will not be linked to the CUREd+ database which contains NHS England data.

The data will be linked/processed with reference datasets such as Ordnance Survey geographical data to create pseudonymised fields, including travel distance/time to care activity; geographic based deprivation information; output area classification; and (for Cohort A only) place of residence type, for use within individual projects. These examples represent just some of the reference datasets that may be used in conjunction with the CUREd+ data: there will be other reference datasets in addition to these. End users of project-specific, de-identified extracts from the processed CUREd+ database may, where appropriate, link or process the extract with publicly-available (e.g., geographical or organisational) data, subject to an assessment of the risk of re-identification carried out during the Data Release Committee's review of each application for a CUREd+ database extract.

The data will not be linked with any other data than is included in this Agreement.

The identifying details, including address details and date of death, will be stored in a separate database to the linked dataset used for analysis and will never be accessed by the CUREd+ study team.

The address data provided for cohort A will allow the generation of pseudonymised Unique Property Reference Numbers (UPRNs), which will enable the grouping of patients who share a household and can be used to access information about the type of residence, without the need to retain the patient address data itself. Once a pseudonymised Unique Property Reference Number (UPRN) has been generated by UoS for all non-institute addresses, this will be linked to the other data (ECDS, HES A&E, HES OP, HES APC, MHSDS, Demographics, Deaths and NHSBSA), and the original, non-institutional, historical address will be destroyed. Historic institutional addresses will continue to be kept on the separate secure computing environment to datasets (ECDS, HES A&E, HES OP, HES APC, MHSDS, Demographics, Deaths and NHSBSA). All analyses will use the pseudonymised dataset.

There will be no requirement and no attempt to reidentify individuals when using the pseudonymised dataset.

Analysts from or associated with the University of Sheffield will analyse the relevant subset of data for the purposes described above.

The medicines data is not deemed disclosive and information on a GP level is available in the public domain. However, should the published information pose a risk of re-identification, the following suppression methodology should be applied:

· Zeros should be shown.

· 1-7 to be rounded to 5.

· Any other numbers rounded to nearest 5.

· Rounding unnecessary for averages etc.

· Percentages calculated from rounded values.

· If zeros need to be suppressed, round to 5.

Outputs are expected to inform, amongst other areas, the pathway that patients take prior to, and through the Urgent and Emergency Care (UEC) system, and factors that influence when, how, and why individuals within specific cohorts of patients join the UEC system in the first place.

The expected outputs of the processing will be:

> Reports aimed at lay audiences, available annually through the University of Sheffield website, indicating the projects completed during the year.

> Submissions of manuscripts to peer-reviewed scientific and medical journals such as the British Medical Journal Quality & Safety and Annals of Emergency Medicine – after the first year, expected to be approximately two peer-reviewed publications per year.

> Presentations at appropriate national and international scientific and medical professional conferences (e.g., Health Services Research UK, European Society for Emergency Medicine Congress)

> Annual report for NHS England which will detail the outputs from all active and finished projects, which have been delivered during the year, and the planned outputs from new projects. The report will reference the associated strategic priorities and programme(s).

> This Agreement is not in direct support for a PhD or postgraduate research study but extracts from the CUREd+ research database may be used in PhD or postgraduate research studies by PhD candidates or members of the University of Sheffield.

> Collaborative research outputs with NHS organisations and commercial collaborators, focused on operational improvements and service delivery innovations.

> New machine learning applications to model patient pathways and predict UEC demand.

> Increased policy engagement, as research findings from NHS and commercial researchers will contribute to NHS operational improvements and system-wide decision-making.

The outputs will not contain NHS England data and will only contain aggregated information with small numbers suppressed as appropriate in line with the relevant disclosure rules for the dataset(s) from which the information was derived.

Projects that use a CUREd+ extract will typically have a dissemination plan for the results of the project that is regularly updated and discussed at project meetings.

The outputs will be communicated to relevant recipients through the following dissemination channels:

> Journals

> Social media

> Press/media engagement

> Reports aimed at lay audiences, available through the University of Sheffield

> Briefings for NHS commissioners and trusts and to parliamentary select committees.

The University of Sheffield is also dedicated to public engagement and supports academics to undertake public engagement with research and enhance the cultural vibrancy of the city. The University of Sheffield delivers a series of festivals and events which provide a platform for academics to engage the public with their research. These consist of university-organised cross-disciplinary events (such as Festival of the Mind and the Mobile University), and national or international festivals (such as Pint of Science, ESRC Festival of Social Science, Being Human, Medical Research Council Festival and International Clinical Trials Day). The University of Sheffield also seeks to help University of Sheffield researchers to achieve impact through public engagement.

It is expected that the first projects to utilise the CUREd+ research database would begin soon after the data has been received, and initial results and outputs coming towards the end of these projects, estimated to be 1/1.5 years. Outputs will continue to be produced throughout the lifespan of the CUREd+ research database, which has a current end date of December 2026.

The majority of projects that will use the CUREd+ research database will be focused around the provision and assessment of the Urgent and Emergency Care (UEC) system. The outputs of these projects are expected to deliver the capability to provide benefits to these services and the patients they treat. The outputs are also expected to seek to provide robust evidence that can influence changes to the way the UEC system operates and is designed to make it more efficient, safer and provide better patient outcomes.

The use of the data could:

> Provide accurate information about the direct and indirect costs to the NHS of the frequent use of UEC. At present, these data are estimated and based on ED attendance alone, without inclusion of other aspects of the UEC network and hospital admission.

> Provide detailed frequency and cost data about different sub-groups of frequent users, and how they use different aspects of the UEC network. At present, this is unknown.

> Help the NHS in planning future services for frequent users and optimising current services to improve outcomes for patients.

> Generate cost-efficiencies. It is estimated that the cost of frequent use of ED alone (i.e., without inclusion of the whole UEC network) is £53 million per annum.

> Address data can be used to access information about the type of residence, without the need to retain the patient address data itself. Being able to group together patients who share a household is essential when looking into the clustering of people who access the UEC system to enable better insight to those who attend within households/families and how they utilise services.

> Be used to help understand the impact that COVID-19 had on the UEC system and the changes that could be done to help the UEC system plan and cope better in future pandemics.

> Provide a greater understanding of the factors that cause an individual to be at a higher-risk for an unplanned UEC contact. The inclusion of mental health services data is hoped to allow the investigation of variation in delivery across services to identify good practice. This will be used to investigate potential areas for improvement to seek to improve patient care across England.

> Provide a more complete picture of patients' patterns of care and allow better identification of cohorts of patients (e.g., identifying patients with asthma, multiple medical conditions) and stratification of patients into various, more detailed risk groups.

> The CUREd+ research database is intended to evaluate the scale of non-urgent attendance at Emergency Departments (ED). Conditions that could be treated elsewhere, such as primary care, will be identified from diagnosis fields. Doing so could lead to benefits for both the ED and patients. Diverting these types of attendances to alternative settings may have a considerable ‘decongesting’ effect on crowded departments, allowing scarce ED resources to be concentrated on those patients in greatest need of the facilities and care provided by a major (type I) emergency department. In addition, patients presenting non-urgently to ED are likely to benefit more clinically from care in an alternative setting where continuity of care may be more appropriate for their type of problem.

> Accelerate innovation in UEC research, particularly in predictive modelling and AI-driven decision-support tools.

> Enhance knowledge-sharing between academic, NHS, and commercial researchers, ensuring research findings are rapidly translated into practice.

> Increase the impact of research, as NHS stakeholders will have direct access to data-driven insights to optimise resource allocation and patient care.

The proposed research outputs from this project are expected to help lead to these benefits by informing and shaping discussion and formation of relevant policy: the research will produce peer-reviewed publications, reports to funders, social media posts, website and blog posts, podcasts, conference and meeting attendances, and direct feedback to policy makers via the NIHR Yorkshire and Humber Applied Research Collaboration, all of which will feed back policy-specific findings to Integrated Care Systems and other relevant commissioning bodies, who can then shape policy in response to relevant research.

CUREd+ is a newly developed, large-scale, linked dataset that provides unparalleled insights into urgent and emergency care (UEC) pathways. It captures the full patient journey, from NHS 111 or 999 calls in Yorkshire and the Humber to emergency department (ED) attendances, hospital admissions and outpatient treatment, use of mental health services, and mortality outcomes nationwide.

While CUREd+ is still in its early days, its value is already evident. Four research studies have had data access approved, and have been receiving data extracts over the last few months, representing a major step towards real-world impact. These NIHR-funded studies will provide insights that directly inform NHS planning, clinical decision-making, and emergency care policy.

By facilitating such studies, CUREd+ is expected to yield benefits in three key areas:

1. Improving Emergency Resource Use and Reducing Unnecessary Admissions

2. Enhancing Patient Outcomes and Reducing Harm

3. Strengthening NHS Policy and Planning Through Evidence-Based Insights

1. Improving Emergency Resource Use and Reducing Unnecessary Admissions

CUREd+ is being used to optimise the use of emergency resources by identifying avoidable admissions, delays, and inefficiencies within the urgent and emergency care system.

The Same Day Emergency Care Study (SDEC)

Same Day Emergency Care (SDEC) services allow some patients to receive urgent treatment without an overnight stay, reducing pressure on hospitals.

This study is using CUREd+ to assess the impact of SDEC on hospital admissions, A&E waiting times, and NHS costs.

By identifying which patients benefit most from SDEC, the research will help optimise SDEC services across England, improving patient flow and reducing unnecessary admissions.

The Risk of Adverse Outcomes after a Suspected Seizure Study (RADOSS)

Ambulance crews often face uncertainty about whether patients with a suspected seizure require hospital admission.

This study will develop a risk prediction tool to help ambulance clinicians determine whether a patient can be safely treated at the scene.

By reducing unnecessary hospital conveyances, this tool will free up emergency resources and ensure that high-risk patients receive hospital care when needed.

Together, these studies will help improve decision-making at multiple points in the emergency care pathway, ensuring that hospital resources are reserved for those who need them most.

2. Enhancing Patient Outcomes and Reducing Harm

CUREd+ is enabling research that will identify factors leading to patient harm and support interventions to reduce mortality and long-term complications.

Reducing harm from long delays in emergency admissions (Mortality & Winter Pressures Study)

This study investigates whether increased demand during winter months leads to higher patient mortality due to admission delays.

Using CUREd+, researchers will analyse how seasonal variation in emergency care impacts waiting times, patient flow, and outcomes.

The findings will be used to support NHS decision-making on surge capacity planning, ensuring that hospitals can better manage demand spikes to prevent avoidable harm.

Minimising complications from prolonged time on the floor after a fall (Long Lies Study)

1 in 5 older adults who fall experience a “long lie” (over 1 hour on the floor), which can lead to dehydration, pressure injuries, muscle damage, and psychological harm.

This study is using CUREd+ to examine the scale of the problem and track the long-term health consequences of a long lie.

Findings will support the development of improved care strategies, including interventions to assist patients while waiting for an ambulance.

These studies will provide actionable insights into how delays and inefficient care pathways contribute to patient harm, leading to better planning, resource allocation, and clinical practice improvements.

3. Strengthening NHS Policy and Planning Through Evidence-Based Insights

CUREd+ is providing the NHS with robust evidence to support policy changes and service improvements, ensuring that emergency care is both effective and sustainable.

Providing evidence for NHS service redesign (SDEC Study)

NHS England has identified SDEC as one of its 10 key interventions to tackle winter pressures and hospital overcrowding.

This study will provide evidence on the effectiveness of SDEC services, ensuring that they are implemented in the most impactful way.

Supporting better NHS planning for winter pressures (Mortality & Winter Pressures Study)

Seasonal demand fluctuations place huge pressure on emergency care services, but there is limited evidence on how these variations impact patient mortality.

This study will provide the first national-scale analysis of how A&E demand changes in winter and whether delays in admission lead to increased deaths.

The findings will be used to help NHS policymakers make informed decisions on staffing levels, resource allocation, and surge capacity during peak periods.

Developing new clinical guidance for managing falls (Long Lies Study)

Current NHS advice for people who have fallen assumes that an ambulance will arrive quickly, which may not always be the case.

This study will provide evidence to inform new guidelines on how to manage a long lie, ensuring that patients receive the right advice and care while waiting for help.

Improving ambulance triage and decision-making (RADOSS Study)

By developing a risk prediction tool for suspected seizures, this study will support the development of national ambulance guidelines for seizure management.

The findings will contribute to better triage models, reducing unnecessary hospital visits and ensuring timely care for high-risk patients.

CUREd+ is still in its early days, but has already attracted strong research interest; the four major studies already approved, together with other studies currently at earlier stages of the approval pipeline, demonstrate the broad range of potential benefits that CUREd+ offers, from improving patient care and reducing harm to optimising emergency resource use and shaping NHS policy.

The last few months have marked a critical milestone as the first data extracts have been shared, setting the stage for transformative research that will directly inform NHS decision-making. As more research teams apply for access, CUREd+ is expected to become one of the most important data resources for urgent and emergency care research in the UK, driving evidence-based improvements in patient care, service efficiency, and health policy.