National Joint Registry Annual Extract

The National Joint Registry (NJR) is commissioned by the Healthcare Quality Improvement Partnership (HQIP) on behalf of NHS England (NHSE) as part of the Clinical Audit and Patient Outcomes Programme (NCAPOP). The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety.

This agreement has Joint Controllership between HQIP and NHSE.

The NCAPOP is a large programme of circa 35 projects consisting of National Clinical Audits. HQIP is commissioned by NHSE to commission and manage the NCAPOP. NHSE is a controller of the NCAPOP jointly with HQIP as together both organisations determine the purposes and means of processing. NHSE is responsible for determining which projects/topics are included as part of the NCAPOP. HQIP, as commissioner of the NCAPOP, is responsible for project specification development, procurement and extension activities, contract management and authorising data sharing requests. NHS England, as a funder of the NCAPOP, participates within specification development, procurement and project extension activities and authorises the publication of project outputs.

NHSE is involved with developing the scope and purpose of the NCAPOP projects through participation within specification development activities and may authorise (as chair of the specification development meetings) the final project specifications. These specifications set out the purpose of the project, the patient groups and clinical services to evaluate and the types of data to collect. NHSE are a representative upon the HQIP Data access request group which authorises data sharing applications from third parties.

LEGAL BASIS JUSTIFICATION:

HQIP and NHSE both rely on the Article 6 (1) (e) legal basis under GDPR - "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". This is justified through commissioning arrangements which link back to NHS England and other national bodies with statutory responsibilities to improve quality of health care services.

HQIP rely on Article 9 (2) (i) as the legal basis for processing under GDPR - "processing is necessary for reasons of public interest in the area of public health, such as protecting against serious cross-border threats to health or ensuring high standards of quality and safety of health care and of medicinal products or medical devices, on the basis of Union or Member State law which provides for suitable and specific measures to safeguard the rights and freedoms of the data subject, in particular professional secrecy". This is justified as all projects aim to drive improvements in the quality and safety of care and to improve outcomes for patients.

NHSE rely on Article 9(2)(h) of the GDPR as the legal basis for processing. "Processing is necessary for the purposes of preventive or occupational medicine, for the assessment of the working capacity of the employee, medical diagnosis, the provision of health or social care or treatment or the management of health or social care systems and services on the basis of Union or Member State law or pursuant to contract with a health professional and subject to the conditions and safeguards referred to in paragraph 3". NHS England are responsible for provision of health and social care, and management of systems and compliance.

The NJR was established in 2003 by the Department of Health following a National Audit Office (NAO) report into the higher than expected failure rate of the 3M hip replacement device. The NAO report concluded that, had a national register of hip replacements been in existence, the failure rate would have been detected earlier. Earlier identification would have meant less patients were affected and the costs of revision surgery for the NHS would have been considerably less. The NJR went live in April 2003. Since its establishment, the responsibility for delivering the NJR has passed to HQIP who deliver the National Joint Registry.

HQIP and NHSE are joint controllers for the NJR and contract to two other organisations for the delivery of the programme, both of which are processors:

• NEC Software Solutions (UK) Limited; (a software and outsourcing business) is responsible for the NJR’s data collection and processing activities, including data storage and the provision of stakeholder reporting services.

• The University of Bristol (UoB) is also a processor with responsibilities for statistical analysis and reporting.

HQIP and NHSE, have no access to the record level data. NEC Software Solutions (UK) Limited has access to record level data, including patient identifiers, in order to link the data provided to patient records held by the NJR. The linked dataset is pseudonymised by NEC Software Solutions (UK) Limited before it is made available to the statistical analysis team at the UoB. Although the UoB team has access to record level data, it does not have access to patient identifiers. The work undertaken by the NJR is to monitor the outcomes of hip, knee, shoulder, ankle, and elbow joint replacement surgery with regards to the performance of devices, surgical teams, and Trusts and hospitals. The monitoring is necessary to ensure patient safety and improve patient outcomes.

The overall purpose of the NJR is summarised in its mission statement:

‘The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety. In a continuous drive to improve the quality of outcomes and ensure the quality and cost effectiveness of joint replacement surgery, the NJR will monitor and report on outcomes, and support and enable related research.’

The strategic goals of the NJR are as follows:

• To monitor the outcomes achieved by brand of prosthesis, hospital and surgeon, and highlight where these fall below an expected performance in order to allow prompt investigation and to support follow-up action.

• To inform patients, clinicians, providers and commissioners of healthcare, regulators and implant suppliers of the outcomes achieved in joint replacement surgery.

• To evidence variations in outcome achieved across surgical practice in order to inform best practice.

• To enhance patient awareness of joint replacement outcomes to better inform patient choice and patients’ quality of experience through engagement with patients and patient organisations.

• To support evidence-based purchasing of joint replacement implants for healthcare providers to support quality and cost effectiveness.

• To support suppliers in the routine post-market surveillance of implants and provide information to clinicians, patients, hospital management and the regulatory authorities.

The NJR has been collecting data since 2003 and is an ongoing audit with no planned end date. The NJR’s Steering Committee reviews the NJR’s mission statement and strategic goals annually to ensure that they remain relevant to existing stakeholder priorities, clinical guidance, and legislation.

The NJR requires record level data from Hospital Episode Statistics (HES), Patient Reported Outcome Measures (PROMS) and Civil Registrations of Death data from NHS England (formerly NHS Digital) to help achieve some of NJR's strategic goals. The NJR creates annual, linked dataset comprising data from the NJR, HES, Patient Episode Database Wales (PEDW), PROMs (NHS England programme) and Civil Registration. This dataset is used primarily for the production of the NJR’s Annual Report, its associated outcomes analyses, and in some reporting services provided to stakeholders. The data is also used for specific research projects undertaken by the NJR. By linking to HES, PROMs and Civil Registration data, the NJR is able to enhance the quality and type of analyses that it undertakes. The priorities for NJR analyses and research are set by the NJR’s Research Committee and Editorial Board, both of which report to the NJR Steering Committee.

Research committee members (including lay members) will review all application statements on benefits to health and care. In all cases where the applicant and/or the study funding are commercial in nature, careful consideration will be given to weigh the expected benefits against any commercial gain to the commercial organisation. Commercial sublicensees must evidence potential benefit to health and care, proportionate to any commercial gain that may be derived from receiving the sublicensed data.

The data sets requested by the NJR consist of HES Patient Admitted Care, Civil Registrations of Death, and Patient Reported Outcomes Measures. Patient identifiers are required in order to link the data requested from NHS England to patient identifiable data held by the NJR. This linkage is necessary to undertake analysis at a record level when, for example, considering factors such as length of stay for different joint replacement types, PROMs associated with different devices/implants, or outcomes associated with deprivation.

Civil Registrations of Death Data is essential for any outcomes analyses: patients who have died with an implant intact, for example, have to be excluded from any cohort. Outcomes analyses also includes a consideration of time to death from operation, and the cause of death.

HES Patient Admitted Care data is required to look at potential factors affecting the outcomes of joint replacement surgery, such as length of stay, re-admission to theatre, associated diagnoses and co-morbidities, deprivation, geography and location.

The NJR currently uses two endpoints in outcomes analysis: death and revision. PROMs represents an invaluable, third endpoint, providing a patients view, as opposed to a clinical view, of the outcome of joint replacement surgery. The NHS England programme is the only source of PROMs data relating to hip and knee joint replacement surgery.

The NJR requires data from 2003 which is when it started its own data collection. The number of years being requested is justified on the need to undertake longitudinal analyses of the available data. Currently, recommended guidance on the expected survivorship of joint replacement (as published by NICE) is based on ten years of data. Long term outcomes analyses are required.

Patient identifiable data is required to link the record level HES/Civil Registration/PROMs data to the relevant patient procedure in the NJR. The primary data field used for linkage is the patient's NHS number, but other fields such as date of birth, local patient ID, postcode, and side of procedure are used to ensure a reliable linkage. Data sets used for analyses of the linked data set have patient identifiers removed.

In some circumstances, it is not possible for the NJR to determine whether a patient has consented or not (‘unknown’ cohort). This might be because their operation was the result of an emergency admission, or because the admitting hospital has been unable to ask the patient to consent for some other reason. For this cohort, the NJR has support under Section 251 (s251) to collect and process data. Research into the ‘unknown’ cohort comes under purpose 2 of the s251 support; and that there is sufficient oversight within the NJR internal assurance procedures to ensure that any research using the ‘unknown’ cohort’s data is within the parameters of purpose 2 of the s251 support.

The sub-cohort received from NHS England where the patient has declined consent do not get linked with NJR and are not shared with applicants under sublicensing, however, the sub-cohort received from NHS England where consent is unknown and there is s251 support for are linked with NJR and do form part of the dataset that are shared with applicants.

For clarity purposes, where a patient is not able to consent for themselves a representative of the individual may act as a consultee and determine that they would have liked to take part in this project. Consultees giving a favourable opinion will enable the patient to take part under Section 30-33 of the Mental Capacity Act 2005.

In order to achieve minimisation, the NJR is requesting data for those OPCS4 codes relating to hip, knee, ankle, elbow, and shoulder joint replacement only. This will also ensure that Civil Registration data relating to those procedures only will be provided. Requesting data related to specific OPCS4 codes will ensure that the data requested is appropriate to the NJR's needs. Minimisation will be achieved by focusing on specific procedure codes.

• Linkage to HES data enables the NJR to improve the type and quality of the analyses that it can undertake without having to collect the data itself. Such linkage may be length of stay linked to outcomes or co-morbidities linked to outcomes. The NJR has, for example, published a paper examining the increased risk of cancer to patients who have had hip replacement procedures involving the use of metal on metal bearing surfaces, i.e. the femoral head and the acetabular cup or liner are both made of metal.

• By linking to PROMs outcomes analyses will improve as poor performance will be detected even if the primary joint replacement procedure has not been revised. A patient’s view of the outcome of joint replacement is a key indicator to the reporting of outcomes and provides an additional endpoint in addition to revision and death. Linkage to PROMs, and analysis at the record level, enables the NJR to monitor the performance of hospitals, surgeons, and individual implants.

• The NJR is required to publish 90 day mortality rates for all those NHS England Trusts undertaking joint replacement surgery in its Annual Clinical Reports to Trusts, the NJR’s Annual Report and as part of the former NHS England Clinical Outcomes Programme (although COP does not exist in its original form, the NJR still publishes annual indicators. The data, originally shared with Choices, is now shared with the CQC). Date of death and any associated co-morbidities are essential for outcomes analyses and for risk-adjusting mortality analyses.

This will enable the NJR to continue to improve its analysis and data quality audits, activities which are expressly intended to improve patient outcomes and patient safety.

The NJR has developed a secure, online Data Access Portal which provides additional security to the data used by researchers and analysts outside of the NJR. The plan is to provide subsets of linked data to approved research projects such that it is processed on the Portal and cannot be removed from the NJR’s security domain. The data does not contain patient identifiable fields.

Data from Wales, Northern Ireland, the Isle of Man, and Guernsey are sourced separately.

SUB-LICENSING (introduced under v8 of this Data Sharing Agreement):

Researchers currently submit research applications to the NJR that require a linked cohort of HES, NJR, and Civil Registry data which requires the submission of two separate data applications to each controller: NHS England and HQIP. Both controllers must then undertake more processing to create the cohort required which includes the provision of patient identifiers for linkage and then adding the source data to those linked identifiers. The NJR would then supply patient identifiers to NHS England for the proposed cohort, the HES records and linked unique NJR ID would then be returned to the applicant, who must then supply the unique IDs to the NJR so that the NJR data can be supplied to the applicant and subsequently linked. This is time-consuming for controllers/processors and has deterred individuals and organisations from using the data to undertake much needed research and analyses. It also entails the transmission of patient identifiable data for each project.

The NJR already creates an annual linked dataset comprising of HES, Civil Registry, PROMs, and NJR data. Sub-licensing datasets extracted from this linked data set will save both NHS England and the NJR considerable effort, shorten timescales for applicants, make the data easily accessible for applicants and encourage greater use of the data for research. It will also reduce the processing and transmission of patient identifiable data. Such fully linked dataset is something NHS England do not routinely provide.

The NJR currently uses two endpoints for undertaking survivorship and outcomes analyses: revision and death. Whilst the NJR captures revision, the use of Civil Registry is essential to determine mortality status, data of death, age at death, and time to death following a joint replacement. Information about the cause of death is used to further refine the outcomes analysis.

The addition of PROMs data provides a third endpoint for the analysis of outcomes for knee and hip joint replacement and being able to measure health gain and improvements in quality of life can be used to support analyses into the outcomes associated with, for example, particular types of procedure or types of implant.

The NJR will not sub-licence HES/PROMs/Civil Registry data in its raw form to applicants. The data from NHS England will be linked to NJR data and used for approved research and analyses relating to joint replacement surgery and which support the NJR’s research priorities. The addition of HES/PROMS/Civil Registry data to data held by the NJR is necessary for all outcomes and survivorship analyses and provides a richer set of data: linking the data enables research that would otherwise be impossible. For example, NJR data provides detailed data on the medical devices used in the operation as well as specific information about surgical approach, throboprophylaxis and intra-operative complications. For shoulder surgery, NJR also collects a programme of Patient Recorded Outcome Measures (PROMs) which are not otherwise available from NHS England. The linked cohort is cleaned and is made ‘research ready’ prior to sub-licensing. This involves re-formatting the data in such a way that the meta-data is available to researchers and the data can easily be used in statistical analysis packages. It should be noted that only subsets of the cohort, relevant to the approved research, are made available to a researcher via a secure Data Access Portal. The cohort is agreed as part of the NJR’s approvals process.

It is anticipated that around 15 applications for sub-licensing would likely be approved each year. Of these, approximately 10 would require NJR data linked to Civil Registration data (date of death or time to death) and approximately 5 would require NJR data linked to HES and/or PROMS data. Sub-licensing, which will make the data more easily accessible, may lead to an increase in the number of applications for data.

The length of the licence is agreed as part of the approvals process and any extension to the agreement must be approved by the NJR’s Research Committee. The length of the sub-licence will depend upon the research being taken but would normally be for 12 months. The NJR has implemented a secure Data Access Portal (DAP) through which sub-licensees must access the data. Access to the data at the end of the licence period can be terminated via the DAP’s central administrative function. Cohorts of data used for research will be archived on the DAP, in line with current guidance, for a period of up to five years. This is necessary should the outcomes of, or the methods employed in, the research be challenged and the analysis needs to be repeated. Access to the archived data would be subject to approval by the Research Committee and DARG.

The approved organisations and researchers, who are granted an access to the linked data via the DAP, agree to terms and conditions of use, their rights and responsibilities as users of the linked data, as defined by the data controllers.

The NJR’s DAP will enable the implementation of some of NHS England’s requirements, especially with regard to the protection and security of the data. Researchers, for example, can only process the data on the DAP and all downloads are placed into a quarantine area prior to being approved for release by an appropriate system administrator. This ensures that the original data remains within the DAP’s security domain.

Sub license and approval groups:

• The NJR Research Committee is responsible to the NJR Steering Committee (NJRSC), and its objectives are:

- To protect NJR data (and any linked dataset) and strengthen its governance through safe effective, and efficient data management.

- To provide a single point of entry and management pathway for all research proposals and activity.

- Maximise access to the data for researchers.

- Align research activity to a framework of priority themes.

- To deliver the NJR’s research strategy through an expertise-based Research Committee.

• Specifically, the Research Committee is responsible for:

- The release of data for research using an impartial and objective protocol.

- Oversight of the use and reporting of data by research groups.

- Upholding the standard and consistency of work carried out using the data.

• All Expressions of Interest (EOI) are formally reviewed by the Research Committee and, following review, an applicant will either be invited to submit a formal application or informed that the NJR will not support the proposed project. Projects fall into two categories:

NJR Supported Project: this is an NJR supported project by a third part applicant. These will typically require aggregate or summary data, or pseudonymised or anonymised patient level data. These projects are external to, but supported by, the NJR. Requests requiring access to patient, surgeon, or unit personal identifiers, or implant batch number will not be granted. NJR will not provide personal identifiers or NJR linked datasets for external projects.

Recent examples include a study of the effect of cement type on the survivorship of cemented total hip replacement by University of Manchester; and examination of surgeon experience-related effects in patient outcomes in elective orthopaedic operations by Sussex University Hospitals NHS Trust.

NJR Partnership Project: this is a project delivered by a third party applicant with NJR involvement and oversight. These will be studies that require sensitive data items, flows of identifiable data, or data linked to external datasets, such as PROMS, HES, or Civil Registration data (requested through NHS England). These projects are undertaken in partnership with the NJR and require an identified named collaborator from the NJR Steering Committee or NJR Research Committee. NJR Partnership Projects require Principal Investigator attendance at a Research Committee meeting before an application can be approved.

Recent examples include an examination of the evaluation of complex hip arthroplasty in the United Kingdom by Northumbria Healthcare NHS Foundation Trust in collaboration with NJR; and the study of floor and ceiling effects of the Oxford Shoulder Score by University of Leicester in collaboration with NJR.

• Any projects requiring external approvals (e,g, CAG, NRES) must be channelled through the NJR application process in the first instance. Upon receipt of Research Committee and before data controller approval, the applicant will be required to provide all evidence of external approvals prior to data release.

• The Research Committee will consider the application in terms of its relevance to the NJR’s research priorities: whether the data requested is clinically appropriate to the proposed use; that the project is methodologically sound; whether aggregated or record level data is required; the applying organisation. In some instances, the Principal Investigator (PI) will be required to attend a Research Committee meeting in person. The NJR also stipulates that the PI must be a substantive employee of the requesting organisation and not be on a fixed-term or temporary contract.

• The use of the DAP will ensure the relevant security measures are adhered to for the sub-licensee’s processing and securing of the data. Data sets made for specific research projects can only be accessed and processed on the DAP by approved users. Physical and administrative processes prevent the download of the data and, if necessary, access to the data can be easily prevented by disabling the project area and its associated user accounts. All other due diligence for data requests (such as formal approval groups etc) will still be adhered to.

• Once a research project has been approved by the NJR Research Committee, the application is then submitted to HQIP’s Data Access Request Group (DARG) for review. Once DARG approval has been granted, the Principal Investigator will have an account set up on the DAP and the data will be loaded into a dedicated, private project area.

NJR Research Committee – Composition:

- NJRSC Public Health and Epidemiology Member (Chair).

- NJRSC Patient Representative Member.

- NJRSC Consultant Orthopaedic Surgeon Member (x4).

- Consultant Orthopaedic Surgeon (foot and ankle surgery expert).

- NJR Medical Director (Medical Advisory Committee representative).

- Expert in mechanical engineering.

- Health Economist.

- Professor of Clinical Epidemiology.

- Professor of Epidemiology.

- Consultant Senior Lecturer in Musculo-skeletal medicine

- NJR Director and Deputy Director of Operations.

- Data Quality Manager (data processing contractor representative)

- Principal Consultant (data processing contractor representative)

- HQIP Data Access Request Group - Composition

- Chief Executive HQIP

- Medical Director HQIP

- Data Protection Officer HQIP

- NHS England Representative

- Welsh Government Representative

- NJR Deputy Director of Operations

- NCAPOP Associate Director

The following Information Governance criteria are considered by the NJR Research Committee and the HQIP DARG:

• Could the project achieve its aims using already published data?

• Whether the application requires record-level data or whether it could achieve its aims using aggregated data.

• Are the data fields requested appropriate to the project or could they be reduced?

• The need, in certain circumstances, to suppress small numbers.

• Whether the project can achieve its aims using anonymised or pseudonymised data. Linked data provided by the DAP will not include data that could be used to identify a patient or surgeon.

• Ensuring that the requesting organisation adheres to appropriate processes, data protection, and security as evidenced by accreditation to standards such as, for example, ISO/IEC 27001, NHSP DSP Toolkit, Cyber Essentials. Whilst many IG requirements are met by the DAP implementation and processes, adherence to standards by requesting organisations demonstrates that staff will be aware of the need for IG and data protection and will have received appropriate training.

• The length of time that the data is required. This would normally be twelve months. Any extensions to that must be approved by the Research Committee.

• That any external approvals (CAG, NRES) have been granted. Evidence will be required before data is made available via the DAP and before the application is submitted to HQIP DARG for final approval.

The GDPR legal basis for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of data controller review ahead of approval. For most applications, no confidential data will be made available to applicants as datasets are pseudonymised prior to release (to the applicant or the DAP). In any application where data is not pseudonymised, the Common Law Duty of Confidentiality for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of the data controller’s review prior to approval. It is also required from the applicant to provide evidence of NHS ethics permissions or evidence that ethics is not required.

The primary consideration in the review of any application for data is that the aims of the project are commensurate with the research goals and priorities of the NJR. The current priority themes approved by the NJR Steering Committee are as follows:

- Device and technology assessment.

- Patient outcomes, safety, and risk management.

- Health economics and cost effectiveness.

- Surgeon education and training.

- Patient information and shared decision-making.

- Disease diagnosis and prognosis.

- Disease causation, prevention, and treatment.

- Injury or trauma in the setting of joint replacement.

- Healthy development and ageing in patients with joint disease and replacement.

All these themes are directly related to both the provision of healthcare and the promotion of health through the twin goals of improving patient safety and patient outcomes. An improvement in patient outcomes also leads to the provision of more cost effective services and reduces costs through a reduction in the need for revision surgery. If applications do not address these themes, they are likely to be rejected.

Any organisation with a legitimate use for the data can apply, the only stipulation being that any proposed project supports the goals of the NJR and that the organisation is capable of delivering a successful outcome within the terms and conditions set out be the NJR for the use of the data. The most likely sub-licensees will be:

- Universities in the UK.

- NHS England Trusts.

- Research Fellows appointed by professional societies and supervised by universities.

- Orthopaedic device manufactures (for technology assessment only).

Sub-licences will be granted on a cost recovery basis only.

All outputs are based on aggregated data: the NJR specifically forbids the use of record level data for publication, even if that data is anonymised. This means that the privacy of all service users is maintained. The NJR consent form and patient information leaflet explain that NJR data will be linked to other, specified data sets and made available to other organisations without any personal details included.

The territory of use for sublicenses will be limited to England and Wales.

A public register of NJR data releases is available here - http://www.njrcentre.org.uk/njrcentre/Research/Research-Portfolio

This public register includes information about where NHS England sourced datasets are shared.

The data controllers will take responsibility for the actions and omissions of all sub licensees and breach of a sub licence should automatically be regarded as breach of the Data Sharing Framework Contract with NHS England. In the event of termination or expiry of the Data Sharing Framework Contract between NHS England and the data controllers, all sub licences shall automatically terminate.

The outputs from the use of the data will vary each year, dependent upon the research and analysis priorities set by the NJR’s Research Committee and NJR’s Editorial Board and agreed by the NJR Steering Committee.

a) The main vehicle for publication is the NJR’s Annual Report which is published annually in September. In addition to the Annual Report, a Patient’s Guide to the Annual Report is also published and made available in both hard and electronic copy. The Annual Report and the patients guide can be accessed at http://www.njrreports.org.uk.

b) Additional analyses are undertaken throughout the year and these are published either on the NJR website or in relevant, professional medical journals. To date those journals have included The Lancet, the British Medical Journal, and the Journal of Bone and Joint Surgery. The NJR maintains a dedicated research section on its website and details of planned, approved, completed, and published analysis can be found at: http://www.njrcentre.org.uk/njrcentre/Research/ResearchPortfolio/tabid/313/Default.aspx

c) The outcomes of analyses are also disseminated at conferences and meetings of professional societies. These include the British Orthopaedic Association, the British Hip Society, the British Association for Surgery of the Knee, the British Elbow and Shoulder Society, and the British Orthopaedic Foot and Ankle Society. The NJR also holds two regional events per year which are aimed at those hospital staff responsible for collecting the data, providing them with an update on the work of the NJR, including its outputs.

d) Some outputs are also published through the NJR’s secure online reporting services. These services include NJR Clinician Feedback, NJR Management Feedback, and NJR Supplier Feedback which provide information for clinicians, Trust/hospital management, and the manufacturers/suppliers of orthopaedic devices respectively.

The NJR is constantly reviewing the ways in which it communicates with patients and the public more generally and are keen to improve its reach. Recent and planned activity include:

• working with NJR Patient Representatives to recruit new members to the NJR Patient Panel to broaden the scope of patient voices on the NJR

• work with relevant patient groups such as Versus Arthritis to ensure that messaging for people who have had or plan to have joint replacement surgery are well targeted

• produce a series of patient guides http://www.njrcentre.org.uk/njrcentre/Reports-Publications-and-Minutes/Public-and-Patient-Guide co-designed with patients

• to press release key developments to ensure appropriate media coverage. See example at https://www.dailymail.co.uk/health/article-7111723/Thinking-getting-new-hip-knee-Try-online-tool-tells-unique-risks.html and https://medicalxpress.com/news/2020-08-implant-choice-important-surgeon-skill.html

• to consult with the PPI group at the Royal National Orthopaedic Hospital, Stanmore about NJR’s data flows

• work with Understanding Patient Data to produce a video about the team's work https://understandingpatientdata.org.uk/case-study/monitoring-joint-replacement-surgery

• collaborating with HQIP and The Health Foundation as part of the Understanding Health Data Access (UHDA) programme which aims to improve information available publicly about secondary data uses

• make a newsletter available via the study website which the public are able to subscribe to. A blog posts from joint replacement patients will also be hosted on the NJR website.

• to remain active on social media platforms (Facebook, Twitter, Linkedin)

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

OUTPUTS FOR SUB-LICENSING:

The principal outputs relating to the sublicensing agreement will be publication in peer reviewed journals and presentation as speciality conferences. The nature of outputs is described by applicants in their application and reviewed by the NJR Research Committee and HQIP DARG. All outputs will be aggregated with small numbers suppressed in accordance with the HES analysis guide.

The exact nature of those outputs will vary depending on the project portfolio, but by way of example a list of the current applications in the pipeline are included below:.

a. The use of the Exeter Short Revision Stem in primary and revision total hip replacement (Royal Devon & Exeter NHS Foundation Trust; NJR:CivReg)

The Exeter Short Revision Stem which is a surgical implant used in hip replacement surgery, (44mm No00 L.125) was released in 2004 and has now been used in both primary and revision surgeries for more than 10 years. This stem is shorter and slimmer than other Exeter stems of the same offset.

Applicant Organisation is undertaking a local review of the survival of the Exeter Short Revision Stem (Exeter stem 44mm No00 L.125) in both primary and revision Total Hip Replacement (THR) in their unit. More importantly however, the performance of the stem in the National Joint Registry in both primary and revision hip arthroplasty needs to be assessed.

Researchers would like to investigate the national survival of prosthesis rates at 2 years, 5 years and 10 years compared with standard Exeter stems of the same offset.

b. Perioperative aspirin and prevention of prosthetic joint infection (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Some artificial joints become infected and need to be removed. This may occur within a short time such as a month after the surgical operation. Such infections may occur even though the skin is thoroughly cleansed, and antibiotics are given prior to the operation. In some cases, infections occur after a much longer period up to many years. When artificial joint infection does occur in the early post surgical period the bacteria that cause such infection typically come from the patient's skin. These bacteria are implanted into the deep part of the surgical wound during the operation to insert the artificial joint. The skin bacteria that most commonly cause these infections are staphylococci. When these and other bacteria come into contact with an artificial joint they form a slime that makes it very difficult for the antibiotics used to treat infection to penetrate and kill them. This slime is called a biofilm. It has been shown in the test tube that aspirin can reduce biofilm formation. For this reason, the applicant organisation believes that it may be possible that aspirin given to patients before their artificial joint surgery may help to prevent infections requiring removal of the joint. The applicant organisation is able to use the records kept in the National Joint registry to measure whether there is an effect of aspirin by comparing the numbers of infections in patients treated with this compared with other medicines to prevent development of clots in the leg veins that may complicate artificial joint surgery. If it seems that aspirin prevents these infections more studies including clinical trials could be done to work out if more patients having artificial joint surgery should be given this commonly used and very safe drug.

c. Distribution and determinants of American Society of Anesthesiologists ( ASA) grade amongst patients undergoing hip arthroplasty: Data from an International Consortium of Joint Registries (University of Oxford; NJR:CivReg:HES)

Total hip replacement is performed to relieve pain and restore function in thousands of patients with hip arthritis each year. Whether or not a total hip replacement has a successful outcome may be affected by a patient’s age, gender and body mass index at the time of surgery. The outcome of surgery may also be affected if a patient has other long-term health disorders, so called ‘comorbidities’, in addition to hip arthritis. For example, previous studies have shown that patients with more co-morbidities are more likely to have complications after total hip replacement, such as infections. The outcomes of total hip replacements in different countries are often compared to provide insights into healthcare in different counties. So that the comparisons are fair, it is necessary to account for differences in the patient's undergoing surgery.

To date, there have been a small number of studies that have investigated how patients undergoing surgery in different countries compare. One previous study found that the age and level of education of patients in Switzerland having total hip replacement was quite different to that of patients having surgery in the USA. However, to date no study has investigated patients differ in terms of co-morbidities. The study propose to investigate how patients undergoing total hip replacements in different countries worldwide differ in terms of how many co-morbidities they have.

Joint replacement registries will be used as the source of information: these are specialised databases that have been created in many countries worldwide and hold information on patients having total hip replacement.

d. The effect of “allowable” unmatched component size on revision rates and time to 1st revision following primary TKR (Nottingham University Hospitals NHS Trust; NJR:CivReg)

Total knee replacements (TKR) contain a femoral and tibial component. Accurate sizing of both components is necessary in order to improve the range of motion of the knee replacement, minimise any soft tissue irritation and avoid bony overhang of the components. It is common practice to choose a tibial component size that matches the femoral component size. However, this does not always guarantee an appropriate level of joint motion and therefore surgeons may choose a size of the tibial component that could be smaller or larger than that of the femoral component which is “allowable” by the manufacturers of knee replacements.

This is not a rare event; a recent Australian study reported that in almost half of the TKRs the two components were not equal in size. This same study found that patients with a femoral component larger than the tibial component were more likely to undergo revision than those with components of equal size or those where the femoral component was smaller than the tibial component. The authors postulated that this might be due to excess stress being placed on the tibial component thus increasing the risk that the TKR would wear out earlier and need revising.

e. Outcomes of revision total hip replacement following ceramic bearing fracture: an avoidable patient safety issue? (University Hospitals of Leicester NHS Trust; NJR:CivReg)

Ceramic is a commonly used material for hip replacement surfaces, but rarely this material may break. When a ceramic bearing fractures this requires further surgery (a revision) to replace the damaged components, of which the surgeon has several materials available to choose from. There have been numerous reported cases of early failures and even death following revision to metal containing bearings after fractures of ceramic components. This is thought to be due to the abrasive effect of hard ceramic particles eroding the softer metal surfaces and subsequently causing metal poisoning. The risk of developing such complications are however unknown, therefore this study aims to quantify the risk of re-revision or early death following such an event.

f. Health Economics Evaluation of Primary Total Hip Arthroplasty by Bearing Type and Fixation Modality (The Royal Orthopaedic Hospital NHS Foundation Trust; NJR:HES:CivReg:PROMs)

The objective of this study is to determine which hip replacement implants have performed best as per the data held by the National Joint Registry. Researcher will use information about each patient undergoing joint revision. The information will include need for revision surgery and Patient Reported Outcome Measures (PROMs).

In addition to this it is necessary to examine the PROMs data for those who have undergone revision surgery (revision is where the joint replacement has been removed & a new joint replacement implanted). This will allow researchers to examine patient characteristics which may have led to the revision surgery as well as seeing how well the joint replacement was performing for that individual patient prior to revision surgery.

Having this full set of data including the Hospital Episode Statistics (HES) data for each patient (which is why traceability is essential for this data) will allow researchers to examine whether patient factors such as comorbidities or previous surgery influence the satisfaction and performance after hip replacement surgery. By grouping data into specific hip replacements (of which there are many available on the UK market) researchers can then see whether there are any outliers either performing particularly well or poorly and whether the performance and satisfaction with these hip replacements may be related to type of implant, comorbidities and similar. This study is a comprehensive analysis of the data available for England, Wales, Northern Ireland and the Isle of Man.

g. Evaluation of complex hip arthroplasty using the NJR dataset (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Hip replacement is one of the most common operations performed in the NHS and one of the most successful procedures in terms of improving quality of life for patients with arthritis. Whilst a hip replacement will last many years for the majority patients, some may run in to problems and require ‘re-do’ (revision) surgery such as when an infection develops, the components become loose or when the bone around the implant breaks. This ‘re-do’ surgery is often more complex than the original surgery and has a higher chance of running into problems including requiring further ‘re-do’ surgery.

The aim of this project it to obtain a perspective of the current delivery of this complex surgery at a national, regional and local levels. The study aim to report on the kinds of operations which are taking place, where they are happening, who is doing them and how many they are doing. This study forms the basis of a wider project aiming to generate data on the best way to structure provision of such surgeries in the future.

h. What is the effect of the femoral head material on revision rate following primary total hip replacement (THR) using the Exeter cemented stem? (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

There are several reasons why primary joint replacements fail. Most orthopaedic studies have looked at the type of fixation (cemented or uncemented) or type of articulations (metal or metal, metal on polyethylene or ceramic). What has not been looked at is the difference in the material between the stem and modular head combinations. There is greater understanding that metal ions can be released from this so called “trunnion wear” head-stem articulation and contribute to failure of joint replacements. This is due to corrosion in this area caused by difference in electrical charge of the materials.

Researchers wish to take advantage of the data available from the National Joint Registry to compare revision rates of different combinations of stem and head in terms of material and explore if this has any effect on the long-term survival (prosthesis) of joint replacements.

i. A comparison of patient reported outcomes and adverse events following elective total shoulder arthroplasty and hemiarthroplasty – an exploratory study (Imperial College London; NJR:CivReg)

Shoulder pain is a common reason people visit their general practitioner and the number of shoulder replacements performed in England, Wales and Northern Ireland has doubled since 2012. A joint replacement is major surgery and it is associated with significant risks. It’s important that patients and surgeons know which type of replacement leads to the best results and the fewest complications. Young patients who are considering a shoulder replacement have a higher chance of further operations, this group need specific information to guide treatment. When people have a joint replacement pseudonymised information about the surgery is stored in the joint registry. In the years following surgery they are asked to complete questionnaires about their symptoms, function and quality of life. The aim is to use this large volume of data to compare the two main types of joint replacement performed for arthritis of the shoulder.

The applicant organisation is requesting data for all shoulder replacements performed since the national joint registry began collection in 2012. This will be linked to the questionnaires given to patients after their procedure and to the NHS database of patient admissions to hospital. The information remains pseudonymised; a patient number is used only to ensure all the information corresponds to the same patient. Researchers will compare the results of patients who have received a half shoulder replacement and total shoulder replacement. The work will influence the decision making of patients and medical staff. The proposed studies will provide information about the expected recovery and the risks and benefits of a shoulder replacement. The work aims to improve patients’ shoulder function after surgery and reduce the number of additional procedures, complications and hospital admissions.

j. Effect of COVID-19 on NJR Surgical Practice and Outcomes (University of Sheffield; NJR:HES:PROMs:CivReg)

COVID 19 has had a great impact on all of our lives and changed society in ways that are yet to be identified. In this work, researchers will look at the outcomes of joint replacement surgery before COVID and then after COVID. This will allow researchers to see how outcomes of joint replacement surgery differ before and after the start of the COVID pandemic by looking at pain and activity, death and repeat operation.

To do this, researchers will compare these measures over 2 years immediately before COVID and compare these to the 2 years after restart of routine operating. In making these analyses researchers will use a range of mathematical tools to best fit the information the study receive from the NJR. One of the approaches researchers will use is called “machine learning” in which the computer “learns” to predict the outcome based on the information put in.

The other methods researchers will use are more traditional statistical approaches. As part of the study researchers will look to see which method makes the best predictions. Researchers imagine that the extra waiting time might affect these outcomes, as might surgeon deskilling.

However, researchers cannot measure these things directly in the NJR, nor see if any patients had COVID around the time of their operation. Findings will be compared with 2 other registers that do record COVID positive tests. The work is exploratory, to see if there are differences in outcomes before and after COVID, and if these are found researchers will create “hypotheses” to explain the possible reason for the effects and test them in further patients

k. Periprosthetic fractures around primary total hip replacement (University of Leeds; NJR:CivReg)

Total Hip replacements (THR) are an effective treatment for managing end-stage hip arthritis. The implants used come in many different shapes and sizes, and work in different ways. The implants are designed to last as long as possible, but a small amount fail because of loosening, infection, instability or breakage of the thigh bone (PFF) amongst other causes.

In patients older than 70, PFF is the second most common cause of failure of THR.

PFF leads to significant problems for patients (more surgery, higher risk of medical complications including death) and their incidence is increasing. Unfortunately, PFF is difficult to prevent and one of the strongest predictors is the type of implants surgeons use. The NJR records when patients with PFF have their implants exchanged (revision).

Previously, researchers identified a number of implant designs which make PFF more likely. This is already making an impact to reduce the risk of PFF. Unfortunately, previous data does not fully explain what happens to patients because patients can also be treated with no surgery and fixation surgery, which are not recorded in the NJR. In order to get a full picture of how the implants affect the risk of PFF researchers need to get information from more sources, which includes implant data (from the NJR), hospital data (which tells us about PFF not recorded in the NJR) and death statistics. This will make sure researchers can work out the complete picture of how the implants used affect the risk of PFF and ultimately how practice can be changed to prevent PFF from harming patients.

This will enable researchers to look at the most complete picture of PFF; how and when it occurs, how it affects patients and identify things which w can be changed to make PFF less likely in future. This will be the largest study looking at PFF. Although this research does not seek to make a fail-safe conclusion about which implant is safest; it will help the surgeons, patients and other stakeholders make informed decision in choosing the right implant for the right patient.

The following benefits to date include, but are not limited to:

Providing outcomes data to the Medicines and Healthcare products Regulatory Agency (MHRA) that has led to the withdrawal of a number of orthopaedic devices from the market.

Quickly identifying to hospitals those patients affected by a Field Notice or Device Alert issued by the MHRA.

Direct influences on changes to clinical practice, including:

• Preventing the use of hip resurfacing procedures in women over the age of 50 (reported through the NJRs Annual Report).

• Getting It Right First Time (GIRFT) proscribing the use of cemented hip implants in patients over 70 years of age (reported in the NJR Annual Report, supplemented by further analysis, and added to Best Practice Tariff (BPT) targets).

• Undertaking analysis that significantly reduced the number of large head, metal-on-metal hip replacements being undertaken.

• Undertaking analysis that resulted in a significant reduction in the number of hip replacement using a metal-on-metal articulating bearing in hip replacement.

• Showing that patella resurfacing at the time of total knee replacement has a better long-term outcome than not resurfacing.

• Professional societies, including the British Orthopaedic Association (BOA), issuing guidance to members about the minimum number of procedure types of procedure to be undertaken in a year in order to remain ‘current’. This follows an analysis of outcomes based on the number of procedures carried out by surgeons.

• A realignment of services for shoulder, ankle, and elbow joint replacement to a ‘hub and spoke’ service delivery on advice from GIRFT, based on an analysis of outcomes and the effects of low volume surgeons.

The provision of enhanced data and reporting services to all stakeholders enabling them to make decisions about the care provided to patients. Stakeholders include clinicians, device manufacturers, service providers and commissioners, patients, regulators, and advisory bodies such as GIRFT and National Institute for Health and Care Excellence (NICE).

Reducing the cost of orthopaedic devices to many NHS Trusts by benchmarking outcomes data against the price paid for implants by those Trusts.

Using risk-adjusted outcomes analyses to identify suspected outlier performance in surgeons, hospitals, and Trusts. This has enabled the appropriate responsible organisations and individuals to take the action necessary to rectify individual or system issues affecting outcomes. Such resolution has included the re-training of surgeons in certain procedure types and surgeons agreeing not to undertake specific procedure types. The NJR, in conjunction with the BOA, provides support to hospitals requesting it.

By publishing and sharing data on outcomes, the NJR has informed changes in clinical practice that have reduced the rate of revision surgery. A reduction in the rate of revision surgery evidence improvements in patient outcomes and safety and also reduces the cost to the NHS of such surgery.

The National Joint Registry (NJR) is commissioned by the Healthcare Quality Improvement Partnership (HQIP) on behalf of NHS England (NHSE) as part of the Clinical Audit and Patient Outcomes Programme (NCAPOP). The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety.

This agreement has Joint Controllership between HQIP and NHSE.

The NCAPOP is a large programme of circa 35 projects consisting of National Clinical Audits. HQIP is commissioned by NHSE to commission and manage the NCAPOP. NHSE is a controller of the NCAPOP jointly with HQIP as together both organisations determine the purposes and means of processing. NHSE is responsible for determining which projects/topics are included as part of the NCAPOP. HQIP, as commissioner of the NCAPOP, is responsible for project specification development, procurement and extension activities, contract management and authorising data sharing requests. NHS England, as a funder of the NCAPOP, participates within specification development, procurement and project extension activities and authorises the publication of project outputs.

NHSE is involved with developing the scope and purpose of the NCAPOP projects through participation within specification development activities and may authorise (as chair of the specification development meetings) the final project specifications. These specifications set out the purpose of the project, the patient groups and clinical services to evaluate and the types of data to collect. NHSE are a representative upon the HQIP Data access request group which authorises data sharing applications from third parties.

LEGAL BASIS JUSTIFICATION:

HQIP and NHSE both rely on the Article 6 (1) (e) legal basis under GDPR - "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". This is justified through commissioning arrangements which link back to NHS England and other national bodies with statutory responsibilities to improve quality of health care services.

HQIP rely on Article 9 (2) (i) as the legal basis for processing under GDPR - "processing is necessary for reasons of public interest in the area of public health, such as protecting against serious cross-border threats to health or ensuring high standards of quality and safety of health care and of medicinal products or medical devices, on the basis of Union or Member State law which provides for suitable and specific measures to safeguard the rights and freedoms of the data subject, in particular professional secrecy". This is justified as all projects aim to drive improvements in the quality and safety of care and to improve outcomes for patients.

NHSE rely on Article 9(2)(h) of the GDPR as the legal basis for processing. "Processing is necessary for the purposes of preventive or occupational medicine, for the assessment of the working capacity of the employee, medical diagnosis, the provision of health or social care or treatment or the management of health or social care systems and services on the basis of Union or Member State law or pursuant to contract with a health professional and subject to the conditions and safeguards referred to in paragraph 3". NHS England are responsible for provision of health and social care, and management of systems and compliance.

The NJR was established in 2003 by the Department of Health following a National Audit Office (NAO) report into the higher than expected failure rate of the 3M hip replacement device. The NAO report concluded that, had a national register of hip replacements been in existence, the failure rate would have been detected earlier. Earlier identification would have meant less patients were affected and the costs of revision surgery for the NHS would have been considerably less. The NJR went live in April 2003. Since its establishment, the responsibility for delivering the NJR has passed to HQIP who deliver the National Joint Registry.

HQIP and NHSE are joint controllers for the NJR and contract to two other organisations for the delivery of the programme, both of which are processors:

• NEC Software Solutions (UK) Limited; (a software and outsourcing business) is responsible for the NJR’s data collection and processing activities, including data storage and the provision of stakeholder reporting services.

• The University of Bristol (UoB) is also a processor with responsibilities for statistical analysis and reporting.

HQIP and NHSE, have no access to the record level data. NEC Software Solutions (UK) Limited has access to record level data, including patient identifiers, in order to link the data provided to patient records held by the NJR. The linked dataset is pseudonymised by NEC Software Solutions (UK) Limited before it is made available to the statistical analysis team at the UoB. Although the UoB team has access to record level data, it does not have access to patient identifiers. The work undertaken by the NJR is to monitor the outcomes of hip, knee, shoulder, ankle, and elbow joint replacement surgery with regards to the performance of devices, surgical teams, and Trusts and hospitals. The monitoring is necessary to ensure patient safety and improve patient outcomes.

The overall purpose of the NJR is summarised in its mission statement:

‘The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety. In a continuous drive to improve the quality of outcomes and ensure the quality and cost effectiveness of joint replacement surgery, the NJR will monitor and report on outcomes, and support and enable related research.’

The strategic goals of the NJR are as follows:

• To monitor the outcomes achieved by brand of prosthesis, hospital and surgeon, and highlight where these fall below an expected performance in order to allow prompt investigation and to support follow-up action.

• To inform patients, clinicians, providers and commissioners of healthcare, regulators and implant suppliers of the outcomes achieved in joint replacement surgery.

• To evidence variations in outcome achieved across surgical practice in order to inform best practice.

• To enhance patient awareness of joint replacement outcomes to better inform patient choice and patients’ quality of experience through engagement with patients and patient organisations.

• To support evidence-based purchasing of joint replacement implants for healthcare providers to support quality and cost effectiveness.

• To support suppliers in the routine post-market surveillance of implants and provide information to clinicians, patients, hospital management and the regulatory authorities.

The NJR has been collecting data since 2003 and is an ongoing audit with no planned end date. The NJR’s Steering Committee reviews the NJR’s mission statement and strategic goals annually to ensure that they remain relevant to existing stakeholder priorities, clinical guidance, and legislation.

The NJR requires record level data from Hospital Episode Statistics (HES), Patient Reported Outcome Measures (PROMS) and Civil Registrations of Death data from NHS England (formerly NHS Digital) to help achieve some of NJR's strategic goals. The NJR creates annual, linked dataset comprising data from the NJR, HES, Patient Episode Database Wales (PEDW), PROMs (NHS England programme) and Civil Registration. This dataset is used primarily for the production of the NJR’s Annual Report, its associated outcomes analyses, and in some reporting services provided to stakeholders. The data is also used for specific research projects undertaken by the NJR. By linking to HES, PROMs and Civil Registration data, the NJR is able to enhance the quality and type of analyses that it undertakes. The priorities for NJR analyses and research are set by the NJR’s Research Committee and Editorial Board, both of which report to the NJR Steering Committee.

Research committee members (including lay members) will review all application statements on benefits to health and care. In all cases where the applicant and/or the study funding are commercial in nature, careful consideration will be given to weigh the expected benefits against any commercial gain to the commercial organisation. Commercial sublicensees must evidence potential benefit to health and care, proportionate to any commercial gain that may be derived from receiving the sublicensed data.

The data sets requested by the NJR consist of HES Patient Admitted Care, Civil Registrations of Death, and Patient Reported Outcomes Measures. Patient identifiers are required in order to link the data requested from NHS England to patient identifiable data held by the NJR. This linkage is necessary to undertake analysis at a record level when, for example, considering factors such as length of stay for different joint replacement types, PROMs associated with different devices/implants, or outcomes associated with deprivation.

Civil Registrations of Death Data is essential for any outcomes analyses: patients who have died with an implant intact, for example, have to be excluded from any cohort. Outcomes analyses also includes a consideration of time to death from operation, and the cause of death.

HES Patient Admitted Care data is required to look at potential factors affecting the outcomes of joint replacement surgery, such as length of stay, re-admission to theatre, associated diagnoses and co-morbidities, deprivation, geography and location.

The NJR currently uses two endpoints in outcomes analysis: death and revision. PROMs represents an invaluable, third endpoint, providing a patients view, as opposed to a clinical view, of the outcome of joint replacement surgery. The NHS England programme is the only source of PROMs data relating to hip and knee joint replacement surgery.

The NJR requires data from 2003 which is when it started its own data collection. The number of years being requested is justified on the need to undertake longitudinal analyses of the available data. Currently, recommended guidance on the expected survivorship of joint replacement (as published by NICE) is based on ten years of data. Long term outcomes analyses are required.

Patient identifiable data is required to link the record level HES/Civil Registration/PROMs data to the relevant patient procedure in the NJR. The primary data field used for linkage is the patient's NHS number, but other fields such as date of birth, local patient ID, postcode, and side of procedure are used to ensure a reliable linkage. Data sets used for analyses of the linked data set have patient identifiers removed.

In some circumstances, it is not possible for the NJR to determine whether a patient has consented or not (‘unknown’ cohort). This might be because their operation was the result of an emergency admission, or because the admitting hospital has been unable to ask the patient to consent for some other reason. For this cohort, the NJR has support under Section 251 (s251) to collect and process data. Research into the ‘unknown’ cohort comes under purpose 2 of the s251 support; and that there is sufficient oversight within the NJR internal assurance procedures to ensure that any research using the ‘unknown’ cohort’s data is within the parameters of purpose 2 of the s251 support.

The sub-cohort received from NHS England where the patient has declined consent do not get linked with NJR and are not shared with applicants under sublicensing, however, the sub-cohort received from NHS England where consent is unknown and there is s251 support for are linked with NJR and do form part of the dataset that are shared with applicants.

For clarity purposes, where a patient is not able to consent for themselves a representative of the individual may act as a consultee and determine that they would have liked to take part in this project. Consultees giving a favourable opinion will enable the patient to take part under Section 30-33 of the Mental Capacity Act 2005.

In order to achieve minimisation, the NJR is requesting data for those OPCS4 codes relating to hip, knee, ankle, elbow, and shoulder joint replacement only. This will also ensure that Civil Registration data relating to those procedures only will be provided. Requesting data related to specific OPCS4 codes will ensure that the data requested is appropriate to the NJR's needs. Minimisation will be achieved by focusing on specific procedure codes.

• Linkage to HES data enables the NJR to improve the type and quality of the analyses that it can undertake without having to collect the data itself. Such linkage may be length of stay linked to outcomes or co-morbidities linked to outcomes. The NJR has, for example, published a paper examining the increased risk of cancer to patients who have had hip replacement procedures involving the use of metal on metal bearing surfaces, i.e. the femoral head and the acetabular cup or liner are both made of metal.

• By linking to PROMs outcomes analyses will improve as poor performance will be detected even if the primary joint replacement procedure has not been revised. A patient’s view of the outcome of joint replacement is a key indicator to the reporting of outcomes and provides an additional endpoint in addition to revision and death. Linkage to PROMs, and analysis at the record level, enables the NJR to monitor the performance of hospitals, surgeons, and individual implants.

• The NJR is required to publish 90 day mortality rates for all those NHS England Trusts undertaking joint replacement surgery in its Annual Clinical Reports to Trusts, the NJR’s Annual Report and as part of the former NHS England Clinical Outcomes Programme (although COP does not exist in its original form, the NJR still publishes annual indicators. The data, originally shared with Choices, is now shared with the CQC). Date of death and any associated co-morbidities are essential for outcomes analyses and for risk-adjusting mortality analyses.

This will enable the NJR to continue to improve its analysis and data quality audits, activities which are expressly intended to improve patient outcomes and patient safety.

The NJR has developed a secure, online Data Access Portal which provides additional security to the data used by researchers and analysts outside of the NJR. The plan is to provide subsets of linked data to approved research projects such that it is processed on the Portal and cannot be removed from the NJR’s security domain. The data does not contain patient identifiable fields.

Data from Wales, Northern Ireland, the Isle of Man, and Guernsey are sourced separately.

SUB-LICENSING (introduced under v8 of this Data Sharing Agreement):

Researchers currently submit research applications to the NJR that require a linked cohort of HES, NJR, and Civil Registry data which requires the submission of two separate data applications to each controller: NHS England and HQIP. Both controllers must then undertake more processing to create the cohort required which includes the provision of patient identifiers for linkage and then adding the source data to those linked identifiers. The NJR would then supply patient identifiers to NHS England for the proposed cohort, the HES records and linked unique NJR ID would then be returned to the applicant, who must then supply the unique IDs to the NJR so that the NJR data can be supplied to the applicant and subsequently linked. This is time-consuming for controllers/processors and has deterred individuals and organisations from using the data to undertake much needed research and analyses. It also entails the transmission of patient identifiable data for each project.

The NJR already creates an annual linked dataset comprising of HES, Civil Registry, PROMs, and NJR data. Sub-licensing datasets extracted from this linked data set will save both NHS England and the NJR considerable effort, shorten timescales for applicants, make the data easily accessible for applicants and encourage greater use of the data for research. It will also reduce the processing and transmission of patient identifiable data. Such fully linked dataset is something NHS England do not routinely provide.

The NJR currently uses two endpoints for undertaking survivorship and outcomes analyses: revision and death. Whilst the NJR captures revision, the use of Civil Registry is essential to determine mortality status, data of death, age at death, and time to death following a joint replacement. Information about the cause of death is used to further refine the outcomes analysis.

The addition of PROMs data provides a third endpoint for the analysis of outcomes for knee and hip joint replacement and being able to measure health gain and improvements in quality of life can be used to support analyses into the outcomes associated with, for example, particular types of procedure or types of implant.

The NJR will not sub-licence HES/PROMs/Civil Registry data in its raw form to applicants. The data from NHS England will be linked to NJR data and used for approved research and analyses relating to joint replacement surgery and which support the NJR’s research priorities. The addition of HES/PROMS/Civil Registry data to data held by the NJR is necessary for all outcomes and survivorship analyses and provides a richer set of data: linking the data enables research that would otherwise be impossible. For example, NJR data provides detailed data on the medical devices used in the operation as well as specific information about surgical approach, throboprophylaxis and intra-operative complications. For shoulder surgery, NJR also collects a programme of Patient Recorded Outcome Measures (PROMs) which are not otherwise available from NHS England. The linked cohort is cleaned and is made ‘research ready’ prior to sub-licensing. This involves re-formatting the data in such a way that the meta-data is available to researchers and the data can easily be used in statistical analysis packages. It should be noted that only subsets of the cohort, relevant to the approved research, are made available to a researcher via a secure Data Access Portal. The cohort is agreed as part of the NJR’s approvals process.

It is anticipated that around 15 applications for sub-licensing would likely be approved each year. Of these, approximately 10 would require NJR data linked to Civil Registration data (date of death or time to death) and approximately 5 would require NJR data linked to HES and/or PROMS data. Sub-licensing, which will make the data more easily accessible, may lead to an increase in the number of applications for data.

The length of the licence is agreed as part of the approvals process and any extension to the agreement must be approved by the NJR’s Research Committee. The length of the sub-licence will depend upon the research being taken but would normally be for 12 months. The NJR has implemented a secure Data Access Portal (DAP) through which sub-licensees must access the data. Access to the data at the end of the licence period can be terminated via the DAP’s central administrative function. Cohorts of data used for research will be archived on the DAP, in line with current guidance, for a period of up to five years. This is necessary should the outcomes of, or the methods employed in, the research be challenged and the analysis needs to be repeated. Access to the archived data would be subject to approval by the Research Committee and DARG.

The approved organisations and researchers, who are granted an access to the linked data via the DAP, agree to terms and conditions of use, their rights and responsibilities as users of the linked data, as defined by the data controllers.

The NJR’s DAP will enable the implementation of some of NHS England’s requirements, especially with regard to the protection and security of the data. Researchers, for example, can only process the data on the DAP and all downloads are placed into a quarantine area prior to being approved for release by an appropriate system administrator. This ensures that the original data remains within the DAP’s security domain.

Sub license and approval groups:

• The NJR Research Committee is responsible to the NJR Steering Committee (NJRSC), and its objectives are:

- To protect NJR data (and any linked dataset) and strengthen its governance through safe effective, and efficient data management.

- To provide a single point of entry and management pathway for all research proposals and activity.

- Maximise access to the data for researchers.

- Align research activity to a framework of priority themes.

- To deliver the NJR’s research strategy through an expertise-based Research Committee.

• Specifically, the Research Committee is responsible for:

- The release of data for research using an impartial and objective protocol.

- Oversight of the use and reporting of data by research groups.

- Upholding the standard and consistency of work carried out using the data.

• All Expressions of Interest (EOI) are formally reviewed by the Research Committee and, following review, an applicant will either be invited to submit a formal application or informed that the NJR will not support the proposed project. Projects fall into two categories:

NJR Supported Project: this is an NJR supported project by a third part applicant. These will typically require aggregate or summary data, or pseudonymised or anonymised patient level data. These projects are external to, but supported by, the NJR. Requests requiring access to patient, surgeon, or unit personal identifiers, or implant batch number will not be granted. NJR will not provide personal identifiers or NJR linked datasets for external projects.

Recent examples include a study of the effect of cement type on the survivorship of cemented total hip replacement by University of Manchester; and examination of surgeon experience-related effects in patient outcomes in elective orthopaedic operations by Sussex University Hospitals NHS Trust.

NJR Partnership Project: this is a project delivered by a third party applicant with NJR involvement and oversight. These will be studies that require sensitive data items, flows of identifiable data, or data linked to external datasets, such as PROMS, HES, or Civil Registration data (requested through NHS England). These projects are undertaken in partnership with the NJR and require an identified named collaborator from the NJR Steering Committee or NJR Research Committee. NJR Partnership Projects require Principal Investigator attendance at a Research Committee meeting before an application can be approved.

Recent examples include an examination of the evaluation of complex hip arthroplasty in the United Kingdom by Northumbria Healthcare NHS Foundation Trust in collaboration with NJR; and the study of floor and ceiling effects of the Oxford Shoulder Score by University of Leicester in collaboration with NJR.

• Any projects requiring external approvals (e,g, CAG, NRES) must be channelled through the NJR application process in the first instance. Upon receipt of Research Committee and before data controller approval, the applicant will be required to provide all evidence of external approvals prior to data release.

• The Research Committee will consider the application in terms of its relevance to the NJR’s research priorities: whether the data requested is clinically appropriate to the proposed use; that the project is methodologically sound; whether aggregated or record level data is required; the applying organisation. In some instances, the Principal Investigator (PI) will be required to attend a Research Committee meeting in person. The NJR also stipulates that the PI must be a substantive employee of the requesting organisation and not be on a fixed-term or temporary contract.

• The use of the DAP will ensure the relevant security measures are adhered to for the sub-licensee’s processing and securing of the data. Data sets made for specific research projects can only be accessed and processed on the DAP by approved users. Physical and administrative processes prevent the download of the data and, if necessary, access to the data can be easily prevented by disabling the project area and its associated user accounts. All other due diligence for data requests (such as formal approval groups etc) will still be adhered to.

• Once a research project has been approved by the NJR Research Committee, the application is then submitted to HQIP’s Data Access Request Group (DARG) for review. Once DARG approval has been granted, the Principal Investigator will have an account set up on the DAP and the data will be loaded into a dedicated, private project area.

NJR Research Committee – Composition:

- NJRSC Public Health and Epidemiology Member (Chair).

- NJRSC Patient Representative Member.

- NJRSC Consultant Orthopaedic Surgeon Member (x4).

- Consultant Orthopaedic Surgeon (foot and ankle surgery expert).

- NJR Medical Director (Medical Advisory Committee representative).

- Expert in mechanical engineering.

- Health Economist.

- Professor of Clinical Epidemiology.

- Professor of Epidemiology.

- Consultant Senior Lecturer in Musculo-skeletal medicine

- NJR Director and Deputy Director of Operations.

- Data Quality Manager (data processing contractor representative)

- Principal Consultant (data processing contractor representative)

- HQIP Data Access Request Group - Composition

- Chief Executive HQIP

- Medical Director HQIP

- Data Protection Officer HQIP

- NHS England Representative

- Welsh Government Representative

- NJR Deputy Director of Operations

- NCAPOP Associate Director

The following Information Governance criteria are considered by the NJR Research Committee and the HQIP DARG:

• Could the project achieve its aims using already published data?

• Whether the application requires record-level data or whether it could achieve its aims using aggregated data.

• Are the data fields requested appropriate to the project or could they be reduced?

• The need, in certain circumstances, to suppress small numbers.

• Whether the project can achieve its aims using anonymised or pseudonymised data. Linked data provided by the DAP will not include data that could be used to identify a patient or surgeon.

• Ensuring that the requesting organisation adheres to appropriate processes, data protection, and security as evidenced by accreditation to standards such as, for example, ISO/IEC 27001, NHSP DSP Toolkit, Cyber Essentials. Whilst many IG requirements are met by the DAP implementation and processes, adherence to standards by requesting organisations demonstrates that staff will be aware of the need for IG and data protection and will have received appropriate training.

• The length of time that the data is required. This would normally be twelve months. Any extensions to that must be approved by the Research Committee.

• That any external approvals (CAG, NRES) have been granted. Evidence will be required before data is made available via the DAP and before the application is submitted to HQIP DARG for final approval.

The GDPR legal basis for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of data controller review ahead of approval. For most applications, no confidential data will be made available to applicants as datasets are pseudonymised prior to release (to the applicant or the DAP). In any application where data is not pseudonymised, the Common Law Duty of Confidentiality for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of the data controller’s review prior to approval. It is also required from the applicant to provide evidence of NHS ethics permissions or evidence that ethics is not required.

The primary consideration in the review of any application for data is that the aims of the project are commensurate with the research goals and priorities of the NJR. The current priority themes approved by the NJR Steering Committee are as follows:

- Device and technology assessment.

- Patient outcomes, safety, and risk management.

- Health economics and cost effectiveness.

- Surgeon education and training.

- Patient information and shared decision-making.

- Disease diagnosis and prognosis.

- Disease causation, prevention, and treatment.

- Injury or trauma in the setting of joint replacement.

- Healthy development and ageing in patients with joint disease and replacement.

All these themes are directly related to both the provision of healthcare and the promotion of health through the twin goals of improving patient safety and patient outcomes. An improvement in patient outcomes also leads to the provision of more cost effective services and reduces costs through a reduction in the need for revision surgery. If applications do not address these themes, they are likely to be rejected.

Any organisation with a legitimate use for the data can apply, the only stipulation being that any proposed project supports the goals of the NJR and that the organisation is capable of delivering a successful outcome within the terms and conditions set out be the NJR for the use of the data. The most likely sub-licensees will be:

- Universities in the UK.

- NHS England Trusts.

- Research Fellows appointed by professional societies and supervised by universities.

- Orthopaedic device manufactures (for technology assessment only).

Sub-licences will be granted on a cost recovery basis only.

All outputs are based on aggregated data: the NJR specifically forbids the use of record level data for publication, even if that data is anonymised. This means that the privacy of all service users is maintained. The NJR consent form and patient information leaflet explain that NJR data will be linked to other, specified data sets and made available to other organisations without any personal details included.

The territory of use for sublicenses will be limited to England and Wales.

A public register of NJR data releases is available here - http://www.njrcentre.org.uk/njrcentre/Research/Research-Portfolio

This public register includes information about where NHS England sourced datasets are shared.

The data controllers will take responsibility for the actions and omissions of all sub licensees and breach of a sub licence should automatically be regarded as breach of the Data Sharing Framework Contract with NHS England. In the event of termination or expiry of the Data Sharing Framework Contract between NHS England and the data controllers, all sub licences shall automatically terminate.

The outputs from the use of the data will vary each year, dependent upon the research and analysis priorities set by the NJR’s Research Committee and NJR’s Editorial Board and agreed by the NJR Steering Committee.

a) The main vehicle for publication is the NJR’s Annual Report which is published annually in September. In addition to the Annual Report, a Patient’s Guide to the Annual Report is also published and made available in both hard and electronic copy. The Annual Report and the patients guide can be accessed at http://www.njrreports.org.uk.

b) Additional analyses are undertaken throughout the year and these are published either on the NJR website or in relevant, professional medical journals. To date those journals have included The Lancet, the British Medical Journal, and the Journal of Bone and Joint Surgery. The NJR maintains a dedicated research section on its website and details of planned, approved, completed, and published analysis can be found at: http://www.njrcentre.org.uk/njrcentre/Research/ResearchPortfolio/tabid/313/Default.aspx

c) The outcomes of analyses are also disseminated at conferences and meetings of professional societies. These include the British Orthopaedic Association, the British Hip Society, the British Association for Surgery of the Knee, the British Elbow and Shoulder Society, and the British Orthopaedic Foot and Ankle Society. The NJR also holds two regional events per year which are aimed at those hospital staff responsible for collecting the data, providing them with an update on the work of the NJR, including its outputs.

d) Some outputs are also published through the NJR’s secure online reporting services. These services include NJR Clinician Feedback, NJR Management Feedback, and NJR Supplier Feedback which provide information for clinicians, Trust/hospital management, and the manufacturers/suppliers of orthopaedic devices respectively.

The NJR is constantly reviewing the ways in which it communicates with patients and the public more generally and are keen to improve its reach. Recent and planned activity include:

• working with NJR Patient Representatives to recruit new members to the NJR Patient Panel to broaden the scope of patient voices on the NJR

• work with relevant patient groups such as Versus Arthritis to ensure that messaging for people who have had or plan to have joint replacement surgery are well targeted

• produce a series of patient guides http://www.njrcentre.org.uk/njrcentre/Reports-Publications-and-Minutes/Public-and-Patient-Guide co-designed with patients

• to press release key developments to ensure appropriate media coverage. See example at https://www.dailymail.co.uk/health/article-7111723/Thinking-getting-new-hip-knee-Try-online-tool-tells-unique-risks.html and https://medicalxpress.com/news/2020-08-implant-choice-important-surgeon-skill.html

• to consult with the PPI group at the Royal National Orthopaedic Hospital, Stanmore about NJR’s data flows

• work with Understanding Patient Data to produce a video about the team's work https://understandingpatientdata.org.uk/case-study/monitoring-joint-replacement-surgery

• collaborating with HQIP and The Health Foundation as part of the Understanding Health Data Access (UHDA) programme which aims to improve information available publicly about secondary data uses

• make a newsletter available via the study website which the public are able to subscribe to. A blog posts from joint replacement patients will also be hosted on the NJR website.

• to remain active on social media platforms (Facebook, Twitter, Linkedin)

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

OUTPUTS FOR SUB-LICENSING:

The principal outputs relating to the sublicensing agreement will be publication in peer reviewed journals and presentation as speciality conferences. The nature of outputs is described by applicants in their application and reviewed by the NJR Research Committee and HQIP DARG. All outputs will be aggregated with small numbers suppressed in accordance with the HES analysis guide.

The exact nature of those outputs will vary depending on the project portfolio, but by way of example a list of the current applications in the pipeline are included below:.

a. The use of the Exeter Short Revision Stem in primary and revision total hip replacement (Royal Devon & Exeter NHS Foundation Trust; NJR:CivReg)

The Exeter Short Revision Stem which is a surgical implant used in hip replacement surgery, (44mm No00 L.125) was released in 2004 and has now been used in both primary and revision surgeries for more than 10 years. This stem is shorter and slimmer than other Exeter stems of the same offset.

Applicant Organisation is undertaking a local review of the survival of the Exeter Short Revision Stem (Exeter stem 44mm No00 L.125) in both primary and revision Total Hip Replacement (THR) in their unit. More importantly however, the performance of the stem in the National Joint Registry in both primary and revision hip arthroplasty needs to be assessed.

Researchers would like to investigate the national survival of prosthesis rates at 2 years, 5 years and 10 years compared with standard Exeter stems of the same offset.

b. Perioperative aspirin and prevention of prosthetic joint infection (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Some artificial joints become infected and need to be removed. This may occur within a short time such as a month after the surgical operation. Such infections may occur even though the skin is thoroughly cleansed, and antibiotics are given prior to the operation. In some cases, infections occur after a much longer period up to many years. When artificial joint infection does occur in the early post surgical period the bacteria that cause such infection typically come from the patient's skin. These bacteria are implanted into the deep part of the surgical wound during the operation to insert the artificial joint. The skin bacteria that most commonly cause these infections are staphylococci. When these and other bacteria come into contact with an artificial joint they form a slime that makes it very difficult for the antibiotics used to treat infection to penetrate and kill them. This slime is called a biofilm. It has been shown in the test tube that aspirin can reduce biofilm formation. For this reason, the applicant organisation believes that it may be possible that aspirin given to patients before their artificial joint surgery may help to prevent infections requiring removal of the joint. The applicant organisation is able to use the records kept in the National Joint registry to measure whether there is an effect of aspirin by comparing the numbers of infections in patients treated with this compared with other medicines to prevent development of clots in the leg veins that may complicate artificial joint surgery. If it seems that aspirin prevents these infections more studies including clinical trials could be done to work out if more patients having artificial joint surgery should be given this commonly used and very safe drug.

c. Distribution and determinants of American Society of Anesthesiologists ( ASA) grade amongst patients undergoing hip arthroplasty: Data from an International Consortium of Joint Registries (University of Oxford; NJR:CivReg:HES)

Total hip replacement is performed to relieve pain and restore function in thousands of patients with hip arthritis each year. Whether or not a total hip replacement has a successful outcome may be affected by a patient’s age, gender and body mass index at the time of surgery. The outcome of surgery may also be affected if a patient has other long-term health disorders, so called ‘comorbidities’, in addition to hip arthritis. For example, previous studies have shown that patients with more co-morbidities are more likely to have complications after total hip replacement, such as infections. The outcomes of total hip replacements in different countries are often compared to provide insights into healthcare in different counties. So that the comparisons are fair, it is necessary to account for differences in the patient's undergoing surgery.

To date, there have been a small number of studies that have investigated how patients undergoing surgery in different countries compare. One previous study found that the age and level of education of patients in Switzerland having total hip replacement was quite different to that of patients having surgery in the USA. However, to date no study has investigated patients differ in terms of co-morbidities. The study propose to investigate how patients undergoing total hip replacements in different countries worldwide differ in terms of how many co-morbidities they have.

Joint replacement registries will be used as the source of information: these are specialised databases that have been created in many countries worldwide and hold information on patients having total hip replacement.

d. The effect of “allowable” unmatched component size on revision rates and time to 1st revision following primary TKR (Nottingham University Hospitals NHS Trust; NJR:CivReg)

Total knee replacements (TKR) contain a femoral and tibial component. Accurate sizing of both components is necessary in order to improve the range of motion of the knee replacement, minimise any soft tissue irritation and avoid bony overhang of the components. It is common practice to choose a tibial component size that matches the femoral component size. However, this does not always guarantee an appropriate level of joint motion and therefore surgeons may choose a size of the tibial component that could be smaller or larger than that of the femoral component which is “allowable” by the manufacturers of knee replacements.

This is not a rare event; a recent Australian study reported that in almost half of the TKRs the two components were not equal in size. This same study found that patients with a femoral component larger than the tibial component were more likely to undergo revision than those with components of equal size or those where the femoral component was smaller than the tibial component. The authors postulated that this might be due to excess stress being placed on the tibial component thus increasing the risk that the TKR would wear out earlier and need revising.

e. Outcomes of revision total hip replacement following ceramic bearing fracture: an avoidable patient safety issue? (University Hospitals of Leicester NHS Trust; NJR:CivReg)

Ceramic is a commonly used material for hip replacement surfaces, but rarely this material may break. When a ceramic bearing fractures this requires further surgery (a revision) to replace the damaged components, of which the surgeon has several materials available to choose from. There have been numerous reported cases of early failures and even death following revision to metal containing bearings after fractures of ceramic components. This is thought to be due to the abrasive effect of hard ceramic particles eroding the softer metal surfaces and subsequently causing metal poisoning. The risk of developing such complications are however unknown, therefore this study aims to quantify the risk of re-revision or early death following such an event.

f. Health Economics Evaluation of Primary Total Hip Arthroplasty by Bearing Type and Fixation Modality (The Royal Orthopaedic Hospital NHS Foundation Trust; NJR:HES:CivReg:PROMs)

The objective of this study is to determine which hip replacement implants have performed best as per the data held by the National Joint Registry. Researcher will use information about each patient undergoing joint revision. The information will include need for revision surgery and Patient Reported Outcome Measures (PROMs).

In addition to this it is necessary to examine the PROMs data for those who have undergone revision surgery (revision is where the joint replacement has been removed & a new joint replacement implanted). This will allow researchers to examine patient characteristics which may have led to the revision surgery as well as seeing how well the joint replacement was performing for that individual patient prior to revision surgery.

Having this full set of data including the Hospital Episode Statistics (HES) data for each patient (which is why traceability is essential for this data) will allow researchers to examine whether patient factors such as comorbidities or previous surgery influence the satisfaction and performance after hip replacement surgery. By grouping data into specific hip replacements (of which there are many available on the UK market) researchers can then see whether there are any outliers either performing particularly well or poorly and whether the performance and satisfaction with these hip replacements may be related to type of implant, comorbidities and similar. This study is a comprehensive analysis of the data available for England, Wales, Northern Ireland and the Isle of Man.

g. Evaluation of complex hip arthroplasty using the NJR dataset (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Hip replacement is one of the most common operations performed in the NHS and one of the most successful procedures in terms of improving quality of life for patients with arthritis. Whilst a hip replacement will last many years for the majority patients, some may run in to problems and require ‘re-do’ (revision) surgery such as when an infection develops, the components become loose or when the bone around the implant breaks. This ‘re-do’ surgery is often more complex than the original surgery and has a higher chance of running into problems including requiring further ‘re-do’ surgery.

The aim of this project it to obtain a perspective of the current delivery of this complex surgery at a national, regional and local levels. The study aim to report on the kinds of operations which are taking place, where they are happening, who is doing them and how many they are doing. This study forms the basis of a wider project aiming to generate data on the best way to structure provision of such surgeries in the future.

h. What is the effect of the femoral head material on revision rate following primary total hip replacement (THR) using the Exeter cemented stem? (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

There are several reasons why primary joint replacements fail. Most orthopaedic studies have looked at the type of fixation (cemented or uncemented) or type of articulations (metal or metal, metal on polyethylene or ceramic). What has not been looked at is the difference in the material between the stem and modular head combinations. There is greater understanding that metal ions can be released from this so called “trunnion wear” head-stem articulation and contribute to failure of joint replacements. This is due to corrosion in this area caused by difference in electrical charge of the materials.

Researchers wish to take advantage of the data available from the National Joint Registry to compare revision rates of different combinations of stem and head in terms of material and explore if this has any effect on the long-term survival (prosthesis) of joint replacements.

i. A comparison of patient reported outcomes and adverse events following elective total shoulder arthroplasty and hemiarthroplasty – an exploratory study (Imperial College London; NJR:CivReg)

Shoulder pain is a common reason people visit their general practitioner and the number of shoulder replacements performed in England, Wales and Northern Ireland has doubled since 2012. A joint replacement is major surgery and it is associated with significant risks. It’s important that patients and surgeons know which type of replacement leads to the best results and the fewest complications. Young patients who are considering a shoulder replacement have a higher chance of further operations, this group need specific information to guide treatment. When people have a joint replacement pseudonymised information about the surgery is stored in the joint registry. In the years following surgery they are asked to complete questionnaires about their symptoms, function and quality of life. The aim is to use this large volume of data to compare the two main types of joint replacement performed for arthritis of the shoulder.

The applicant organisation is requesting data for all shoulder replacements performed since the national joint registry began collection in 2012. This will be linked to the questionnaires given to patients after their procedure and to the NHS database of patient admissions to hospital. The information remains pseudonymised; a patient number is used only to ensure all the information corresponds to the same patient. Researchers will compare the results of patients who have received a half shoulder replacement and total shoulder replacement. The work will influence the decision making of patients and medical staff. The proposed studies will provide information about the expected recovery and the risks and benefits of a shoulder replacement. The work aims to improve patients’ shoulder function after surgery and reduce the number of additional procedures, complications and hospital admissions.

j. Effect of COVID-19 on NJR Surgical Practice and Outcomes (University of Sheffield; NJR:HES:PROMs:CivReg)

COVID 19 has had a great impact on all of our lives and changed society in ways that are yet to be identified. In this work, researchers will look at the outcomes of joint replacement surgery before COVID and then after COVID. This will allow researchers to see how outcomes of joint replacement surgery differ before and after the start of the COVID pandemic by looking at pain and activity, death and repeat operation.

To do this, researchers will compare these measures over 2 years immediately before COVID and compare these to the 2 years after restart of routine operating. In making these analyses researchers will use a range of mathematical tools to best fit the information the study receive from the NJR. One of the approaches researchers will use is called “machine learning” in which the computer “learns” to predict the outcome based on the information put in.

The other methods researchers will use are more traditional statistical approaches. As part of the study researchers will look to see which method makes the best predictions. Researchers imagine that the extra waiting time might affect these outcomes, as might surgeon deskilling.

However, researchers cannot measure these things directly in the NJR, nor see if any patients had COVID around the time of their operation. Findings will be compared with 2 other registers that do record COVID positive tests. The work is exploratory, to see if there are differences in outcomes before and after COVID, and if these are found researchers will create “hypotheses” to explain the possible reason for the effects and test them in further patients

k. Periprosthetic fractures around primary total hip replacement (University of Leeds; NJR:CivReg)

Total Hip replacements (THR) are an effective treatment for managing end-stage hip arthritis. The implants used come in many different shapes and sizes, and work in different ways. The implants are designed to last as long as possible, but a small amount fail because of loosening, infection, instability or breakage of the thigh bone (PFF) amongst other causes.

In patients older than 70, PFF is the second most common cause of failure of THR.

PFF leads to significant problems for patients (more surgery, higher risk of medical complications including death) and their incidence is increasing. Unfortunately, PFF is difficult to prevent and one of the strongest predictors is the type of implants surgeons use. The NJR records when patients with PFF have their implants exchanged (revision).

Previously, researchers identified a number of implant designs which make PFF more likely. This is already making an impact to reduce the risk of PFF. Unfortunately, previous data does not fully explain what happens to patients because patients can also be treated with no surgery and fixation surgery, which are not recorded in the NJR. In order to get a full picture of how the implants affect the risk of PFF researchers need to get information from more sources, which includes implant data (from the NJR), hospital data (which tells us about PFF not recorded in the NJR) and death statistics. This will make sure researchers can work out the complete picture of how the implants used affect the risk of PFF and ultimately how practice can be changed to prevent PFF from harming patients.

This will enable researchers to look at the most complete picture of PFF; how and when it occurs, how it affects patients and identify things which w can be changed to make PFF less likely in future. This will be the largest study looking at PFF. Although this research does not seek to make a fail-safe conclusion about which implant is safest; it will help the surgeons, patients and other stakeholders make informed decision in choosing the right implant for the right patient.

The following benefits to date include, but are not limited to:

Providing outcomes data to the Medicines and Healthcare products Regulatory Agency (MHRA) that has led to the withdrawal of a number of orthopaedic devices from the market.

Quickly identifying to hospitals those patients affected by a Field Notice or Device Alert issued by the MHRA.

Direct influences on changes to clinical practice, including:

• Preventing the use of hip resurfacing procedures in women over the age of 50 (reported through the NJRs Annual Report).

• Getting It Right First Time (GIRFT) proscribing the use of cemented hip implants in patients over 70 years of age (reported in the NJR Annual Report, supplemented by further analysis, and added to Best Practice Tariff (BPT) targets).

• Undertaking analysis that significantly reduced the number of large head, metal-on-metal hip replacements being undertaken.

• Undertaking analysis that resulted in a significant reduction in the number of hip replacement using a metal-on-metal articulating bearing in hip replacement.

• Showing that patella resurfacing at the time of total knee replacement has a better long-term outcome than not resurfacing.

• Professional societies, including the British Orthopaedic Association (BOA), issuing guidance to members about the minimum number of procedure types of procedure to be undertaken in a year in order to remain ‘current’. This follows an analysis of outcomes based on the number of procedures carried out by surgeons.

• A realignment of services for shoulder, ankle, and elbow joint replacement to a ‘hub and spoke’ service delivery on advice from GIRFT, based on an analysis of outcomes and the effects of low volume surgeons.

The provision of enhanced data and reporting services to all stakeholders enabling them to make decisions about the care provided to patients. Stakeholders include clinicians, device manufacturers, service providers and commissioners, patients, regulators, and advisory bodies such as GIRFT and National Institute for Health and Care Excellence (NICE).

Reducing the cost of orthopaedic devices to many NHS Trusts by benchmarking outcomes data against the price paid for implants by those Trusts.

Using risk-adjusted outcomes analyses to identify suspected outlier performance in surgeons, hospitals, and Trusts. This has enabled the appropriate responsible organisations and individuals to take the action necessary to rectify individual or system issues affecting outcomes. Such resolution has included the re-training of surgeons in certain procedure types and surgeons agreeing not to undertake specific procedure types. The NJR, in conjunction with the BOA, provides support to hospitals requesting it.

By publishing and sharing data on outcomes, the NJR has informed changes in clinical practice that have reduced the rate of revision surgery. A reduction in the rate of revision surgery evidence improvements in patient outcomes and safety and also reduces the cost to the NHS of such surgery.

The National Joint Registry (NJR) is commissioned by the Healthcare Quality Improvement Partnership (HQIP) on behalf of NHS England (NHSE) as part of the Clinical Audit and Patient Outcomes Programme (NCAPOP). The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety.

This agreement has Joint Controllership between HQIP and NHSE.

The NCAPOP is a large programme of circa 35 projects consisting of National Clinical Audits. HQIP is commissioned by NHSE to commission and manage the NCAPOP. NHSE is a controller of the NCAPOP jointly with HQIP as together both organisations determine the purposes and means of processing. NHSE is responsible for determining which projects/topics are included as part of the NCAPOP. HQIP, as commissioner of the NCAPOP, is responsible for project specification development, procurement and extension activities, contract management and authorising data sharing requests. NHS England, as a funder of the NCAPOP, participates within specification development, procurement and project extension activities and authorises the publication of project outputs.

NHSE is involved with developing the scope and purpose of the NCAPOP projects through participation within specification development activities and may authorise (as chair of the specification development meetings) the final project specifications. These specifications set out the purpose of the project, the patient groups and clinical services to evaluate and the types of data to collect. NHSE are a representative upon the HQIP Data access request group which authorises data sharing applications from third parties.

LEGAL BASIS JUSTIFICATION:

HQIP and NHSE both rely on the Article 6 (1) (e) legal basis under GDPR - "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". This is justified through commissioning arrangements which link back to NHS England and other national bodies with statutory responsibilities to improve quality of health care services.

HQIP rely on Article 9 (2) (i) as the legal basis for processing under GDPR - "processing is necessary for reasons of public interest in the area of public health, such as protecting against serious cross-border threats to health or ensuring high standards of quality and safety of health care and of medicinal products or medical devices, on the basis of Union or Member State law which provides for suitable and specific measures to safeguard the rights and freedoms of the data subject, in particular professional secrecy". This is justified as all projects aim to drive improvements in the quality and safety of care and to improve outcomes for patients.

NHSE rely on Article 9(2)(h) of the GDPR as the legal basis for processing. "Processing is necessary for the purposes of preventive or occupational medicine, for the assessment of the working capacity of the employee, medical diagnosis, the provision of health or social care or treatment or the management of health or social care systems and services on the basis of Union or Member State law or pursuant to contract with a health professional and subject to the conditions and safeguards referred to in paragraph 3". NHS England are responsible for provision of health and social care, and management of systems and compliance.

The NJR was established in 2003 by the Department of Health following a National Audit Office (NAO) report into the higher than expected failure rate of the 3M hip replacement device. The NAO report concluded that, had a national register of hip replacements been in existence, the failure rate would have been detected earlier. Earlier identification would have meant less patients were affected and the costs of revision surgery for the NHS would have been considerably less. The NJR went live in April 2003. Since its establishment, the responsibility for delivering the NJR has passed to HQIP who deliver the National Joint Registry.

HQIP and NHSE are joint controllers for the NJR and contract to two other organisations for the delivery of the programme, both of which are processors:

• NEC Software Solutions (UK) Limited; (a software and outsourcing business) is responsible for the NJR’s data collection and processing activities, including data storage and the provision of stakeholder reporting services.

• The University of Bristol (UoB) is also a processor with responsibilities for statistical analysis and reporting.

HQIP and NHSE, have no access to the record level data. NEC Software Solutions (UK) Limited has access to record level data, including patient identifiers, in order to link the data provided to patient records held by the NJR. The linked dataset is pseudonymised by NEC Software Solutions (UK) Limited before it is made available to the statistical analysis team at the UoB. Although the UoB team has access to record level data, it does not have access to patient identifiers. The work undertaken by the NJR is to monitor the outcomes of hip, knee, shoulder, ankle, and elbow joint replacement surgery with regards to the performance of devices, surgical teams, and Trusts and hospitals. The monitoring is necessary to ensure patient safety and improve patient outcomes.

The overall purpose of the NJR is summarised in its mission statement:

‘The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety. In a continuous drive to improve the quality of outcomes and ensure the quality and cost effectiveness of joint replacement surgery, the NJR will monitor and report on outcomes, and support and enable related research.’

The strategic goals of the NJR are as follows:

• To monitor the outcomes achieved by brand of prosthesis, hospital and surgeon, and highlight where these fall below an expected performance in order to allow prompt investigation and to support follow-up action.

• To inform patients, clinicians, providers and commissioners of healthcare, regulators and implant suppliers of the outcomes achieved in joint replacement surgery.

• To evidence variations in outcome achieved across surgical practice in order to inform best practice.

• To enhance patient awareness of joint replacement outcomes to better inform patient choice and patients’ quality of experience through engagement with patients and patient organisations.

• To support evidence-based purchasing of joint replacement implants for healthcare providers to support quality and cost effectiveness.

• To support suppliers in the routine post-market surveillance of implants and provide information to clinicians, patients, hospital management and the regulatory authorities.

The NJR has been collecting data since 2003 and is an ongoing audit with no planned end date. The NJR’s Steering Committee reviews the NJR’s mission statement and strategic goals annually to ensure that they remain relevant to existing stakeholder priorities, clinical guidance, and legislation.

The NJR requires record level data from Hospital Episode Statistics (HES), Patient Reported Outcome Measures (PROMS) and Civil Registrations of Death data from NHS England (formerly NHS Digital) to help achieve some of NJR's strategic goals. The NJR creates annual, linked dataset comprising data from the NJR, HES, Patient Episode Database Wales (PEDW), PROMs (NHS England programme) and Civil Registration. This dataset is used primarily for the production of the NJR’s Annual Report, its associated outcomes analyses, and in some reporting services provided to stakeholders. The data is also used for specific research projects undertaken by the NJR. By linking to HES, PROMs and Civil Registration data, the NJR is able to enhance the quality and type of analyses that it undertakes. The priorities for NJR analyses and research are set by the NJR’s Research Committee and Editorial Board, both of which report to the NJR Steering Committee.

Research committee members (including lay members) will review all application statements on benefits to health and care. In all cases where the applicant and/or the study funding are commercial in nature, careful consideration will be given to weigh the expected benefits against any commercial gain to the commercial organisation. Commercial sublicensees must evidence potential benefit to health and care, proportionate to any commercial gain that may be derived from receiving the sublicensed data.

The data sets requested by the NJR consist of HES Patient Admitted Care, Civil Registrations of Death, and Patient Reported Outcomes Measures. Patient identifiers are required in order to link the data requested from NHS England to patient identifiable data held by the NJR. This linkage is necessary to undertake analysis at a record level when, for example, considering factors such as length of stay for different joint replacement types, PROMs associated with different devices/implants, or outcomes associated with deprivation.

Civil Registrations of Death Data is essential for any outcomes analyses: patients who have died with an implant intact, for example, have to be excluded from any cohort. Outcomes analyses also includes a consideration of time to death from operation, and the cause of death.

HES Patient Admitted Care data is required to look at potential factors affecting the outcomes of joint replacement surgery, such as length of stay, re-admission to theatre, associated diagnoses and co-morbidities, deprivation, geography and location.

The NJR currently uses two endpoints in outcomes analysis: death and revision. PROMs represents an invaluable, third endpoint, providing a patients view, as opposed to a clinical view, of the outcome of joint replacement surgery. The NHS England programme is the only source of PROMs data relating to hip and knee joint replacement surgery.

The NJR requires data from 2003 which is when it started its own data collection. The number of years being requested is justified on the need to undertake longitudinal analyses of the available data. Currently, recommended guidance on the expected survivorship of joint replacement (as published by NICE) is based on ten years of data. Long term outcomes analyses are required.

Patient identifiable data is required to link the record level HES/Civil Registration/PROMs data to the relevant patient procedure in the NJR. The primary data field used for linkage is the patient's NHS number, but other fields such as date of birth, local patient ID, postcode, and side of procedure are used to ensure a reliable linkage. Data sets used for analyses of the linked data set have patient identifiers removed.

In some circumstances, it is not possible for the NJR to determine whether a patient has consented or not (‘unknown’ cohort). This might be because their operation was the result of an emergency admission, or because the admitting hospital has been unable to ask the patient to consent for some other reason. For this cohort, the NJR has support under Section 251 (s251) to collect and process data. Research into the ‘unknown’ cohort comes under purpose 2 of the s251 support; and that there is sufficient oversight within the NJR internal assurance procedures to ensure that any research using the ‘unknown’ cohort’s data is within the parameters of purpose 2 of the s251 support.

The sub-cohort received from NHS England where the patient has declined consent do not get linked with NJR and are not shared with applicants under sublicensing, however, the sub-cohort received from NHS England where consent is unknown and there is s251 support for are linked with NJR and do form part of the dataset that are shared with applicants.

In order to achieve minimisation, the NJR is requesting data for those OPCS4 codes relating to hip, knee, ankle, elbow, and shoulder joint replacement only. This will also ensure that Civil Registration data relating to those procedures only will be provided. Requesting data related to specific OPCS4 codes will ensure that the data requested is appropriate to the NJR's needs. Minimisation will be achieved by focusing on specific procedure codes.

• Linkage to HES data enables the NJR to improve the type and quality of the analyses that it can undertake without having to collect the data itself. Such linkage may be length of stay linked to outcomes or co-morbidities linked to outcomes. The NJR has, for example, published a paper examining the increased risk of cancer to patients who have had hip replacement procedures involving the use of metal on metal bearing surfaces, i.e. the femoral head and the acetabular cup or liner are both made of metal.

• By linking to PROMs outcomes analyses will improve as poor performance will be detected even if the primary joint replacement procedure has not been revised. A patient’s view of the outcome of joint replacement is a key indicator to the reporting of outcomes and provides an additional endpoint in addition to revision and death. Linkage to PROMs, and analysis at the record level, enables the NJR to monitor the performance of hospitals, surgeons, and individual implants.

• The NJR is required to publish 90 day mortality rates for all those NHS England Trusts undertaking joint replacement surgery in its Annual Clinical Reports to Trusts, the NJR’s Annual Report and as part of the former NHS England Clinical Outcomes Programme (although COP does not exist in its original form, the NJR still publishes annual indicators. The data, originally shared with Choices, is now shared with the CQC). Date of death and any associated co-morbidities are essential for outcomes analyses and for risk-adjusting mortality analyses.

This will enable the NJR to continue to improve its analysis and data quality audits, activities which are expressly intended to improve patient outcomes and patient safety.

The NJR has developed a secure, online Data Access Portal which provides additional security to the data used by researchers and analysts outside of the NJR. The plan is to provide subsets of linked data to approved research projects such that it is processed on the Portal and cannot be removed from the NJR’s security domain. The data does not contain patient identifiable fields.

Data from Wales, Northern Ireland, the Isle of Man, and Guernsey are sourced separately.

SUB-LICENSING (introduced under v8 of this agreement):

Researchers currently submit research applications to the NJR that require a linked cohort of HES, NJR, and Civil Registry data which requires the submission of two separate data applications to each controller: NHS England and HQIP. Both controllers must then undertake more processing to create the cohort required which includes the provision of patient identifiers for linkage and then adding the source data to those linked identifiers. The NJR would then supply patient identifiers to NHS England for the proposed cohort, the HES records and linked unique NJR ID would then be returned to the applicant, who must then supply the unique IDs to the NJR so that the NJR data can be supplied to the applicant and subsequently linked. This is time-consuming for controllers/processors and has deterred individuals and organisations from using the data to undertake much needed research and analyses. It also entails the transmission of patient identifiable data for each project.

The NJR already creates an annual linked dataset comprising of HES, Civil Registry, PROMs, and NJR data. Sub-licensing datasets extracted from this linked data set will save both NHS England and the NJR considerable effort, shorten timescales for applicants, make the data easily accessible for applicants and encourage greater use of the data for research. It will also reduce the processing and transmission of patient identifiable data. Such fully linked dataset is something NHS England do not routinely provide.

The NJR currently uses two endpoints for undertaking survivorship and outcomes analyses: revision and death. Whilst the NJR captures revision, the use of Civil Registry is essential to determine mortality status, data of death, age at death, and time to death following a joint replacement. Information about the cause of death is used to further refine the outcomes analysis.

The addition of PROMs data provides a third endpoint for the analysis of outcomes for knee and hip joint replacement and being able to measure health gain and improvements in quality of life can be used to support analyses into the outcomes associated with, for example, particular types of procedure or types of implant.

The NJR will not sub-licence HES/PROMs/Civil Registry data in its raw form to applicants. The data from NHS England will be linked to NJR data and used for approved research and analyses relating to joint replacement surgery and which support the NJR’s research priorities. The addition of HES/PROMS/Civil Registry data to data held by the NJR is necessary for all outcomes and survivorship analyses and provides a richer set of data: linking the data enables research that would otherwise be impossible. For example, NJR data provides detailed data on the medical devices used in the operation as well as specific information about surgical approach, throboprophylaxis and intra-operative complications. For shoulder surgery, NJR also collects a programme of Patient Recorded Outcome Measures (PROMs) which are not otherwise available from NHS England. The linked cohort is cleaned and is made ‘research ready’ prior to sub-licensing. This involves re-formatting the data in such a way that the meta-data is available to researchers and the data can easily be used in statistical analysis packages. It should be noted that only subsets of the cohort, relevant to the approved research, are made available to a researcher via a secure Data Access Portal. The cohort is agreed as part of the NJR’s approvals process.

It is anticipated that around 15 applications for sub-licensing would likely be approved each year. Of these, approximately 10 would require NJR data linked to Civil Registration data (date of death or time to death) and approximately 5 would require NJR data linked to HES and/or PROMS data. Sub-licensing, which will make the data more easily accessible, may lead to an increase in the number of applications for data.

The length of the licence is agreed as part of the approvals process and any extension to the agreement must be approved by the NJR’s Research Committee. The length of the sub-licence will depend upon the research being taken but would normally be for 12 months. The NJR has implemented a secure Data Access Portal (DAP) through which sub-licensees must access the data. Access to the data at the end of the licence period can be terminated via the DAP’s central administrative function. Cohorts of data used for research will be archived on the DAP, in line with current guidance, for a period of up to five years. This is necessary should the outcomes of, or the methods employed in, the research be challenged and the analysis needs to be repeated. Access to the archived data would be subject to approval by the Research Committee and DARG.

The approved organisations and researchers, who are granted an access to the linked data via the DAP, agree to terms and conditions of use, their rights and responsibilities as users of the linked data, as defined by the data controllers.

The NJR’s DAP will enable the implementation of some of NHS England’s requirements, especially with regard to the protection and security of the data. Researchers, for example, can only process the data on the DAP and all downloads are placed into a quarantine area prior to being approved for release by an appropriate system administrator. This ensures that the original data remains within the DAP’s security domain.

Sub license and approval groups:

• The NJR Research Committee is responsible to the NJR Steering Committee (NJRSC), and its objectives are:

- To protect NJR data (and any linked dataset) and strengthen its governance through safe effective, and efficient data management.

- To provide a single point of entry and management pathway for all research proposals and activity.

- Maximise access to the data for researchers.

- Align research activity to a framework of priority themes.

- To deliver the NJR’s research strategy through an expertise-based Research Committee.

• Specifically, the Research Committee is responsible for:

- The release of data for research using an impartial and objective protocol.

- Oversight of the use and reporting of data by research groups.

- Upholding the standard and consistency of work carried out using the data.

• All Expressions of Interest (EOI) are formally reviewed by the Research Committee and, following review, an applicant will either be invited to submit a formal application or informed that the NJR will not support the proposed project. Projects fall into two categories:

NJR Supported Project: this is an NJR supported project by a third part applicant. These will typically require aggregate or summary data, or pseudonymised or anonymised patient level data. These projects are external to, but supported by, the NJR. Requests requiring access to patient, surgeon, or unit personal identifiers, or implant batch number will not be granted. NJR will not provide personal identifiers or NJR linked datasets for external projects.

Recent examples include a study of the effect of cement type on the survivorship of cemented total hip replacement by University of Manchester; and examination of surgeon experience-related effects in patient outcomes in elective orthopaedic operations by Sussex University Hospitals NHS Trust.

NJR Partnership Project: this is a project delivered by a third party applicant with NJR involvement and oversight. These will be studies that require sensitive data items, flows of identifiable data, or data linked to external datasets, such as PROMS, HES, or Civil Registration data (requested through NHS England). These projects are undertaken in partnership with the NJR and require an identified named collaborator from the NJR Steering Committee or NJR Research Committee. NJR Partnership Projects require Principal Investigator attendance at a Research Committee meeting before an application can be approved.

Recent examples include an examination of the evaluation of complex hip arthroplasty in the United Kingdom by Northumbria Healthcare NHS Foundation Trust in collaboration with NJR; and the study of floor and ceiling effects of the Oxford Shoulder Score by University of Leicester in collaboration with NJR.

• Any projects requiring external approvals (e,g, CAG, NRES) must be channelled through the NJR application process in the first instance. Upon receipt of Research Committee and before data controller approval, the applicant will be required to provide all evidence of external approvals prior to data release.

• The Research Committee will consider the application in terms of its relevance to the NJR’s research priorities: whether the data requested is clinically appropriate to the proposed use; that the project is methodologically sound; whether aggregated or record level data is required; the applying organisation. In some instances, the Principal Investigator (PI) will be required to attend a Research Committee meeting in person. The NJR also stipulates that the PI must be a substantive employee of the requesting organisation and not be on a fixed-term or temporary contract.

• The use of the DAP will ensure the relevant security measures are adhered to for the sub-licensee’s processing and securing of the data. Data sets made for specific research projects can only be accessed and processed on the DAP by approved users. Physical and administrative processes prevent the download of the data and, if necessary, access to the data can be easily prevented by disabling the project area and its associated user accounts. All other due diligence for data requests (such as formal approval groups etc) will still be adhered to.

• Once a research project has been approved by the NJR Research Committee, the application is then submitted to HQIP’s Data Access Request Group (DARG) for review. Once DARG approval has been granted, the Principal Investigator will have an account set up on the DAP and the data will be loaded into a dedicated, private project area.

NJR Research Committee – Composition:

- NJRSC Public Health and Epidemiology Member (Chair).

- NJRSC Patient Representative Member.

- NJRSC Consultant Orthopaedic Surgeon Member (x4).

- Consultant Orthopaedic Surgeon (foot and ankle surgery expert).

- NJR Medical Director (Medical Advisory Committee representative).

- Expert in mechanical engineering.

- Health Economist.

- Professor of Clinical Epidemiology.

- Professor of Epidemiology.

- Consultant Senior Lecturer in Musculo-skeletal medicine

- NJR Director and Deputy Director of Operations.

- Data Quality Manager (data processing contractor representative)

- Principal Consultant (data processing contractor representative)

- HQIP Data Access Request Group - Composition

- Chief Executive HQIP

- Medical Director HQIP

- Data Protection Officer HQIP

- NHS England Representative

- Welsh Government Representative

- NJR Deputy Director of Operations

- NCAPOP Associate Director

The following Information Governance criteria are considered by the NJR Research Committee and the HQIP DARG:

• Could the project achieve its aims using already published data?

• Whether the application requires record-level data or whether it could achieve its aims using aggregated data.

• Are the data fields requested appropriate to the project or could they be reduced?

• The need, in certain circumstances, to suppress small numbers.

• Whether the project can achieve its aims using anonymised or pseudonymised data. Linked data provided by the DAP will not include data that could be used to identify a patient or surgeon.

• Ensuring that the requesting organisation adheres to appropriate processes, data protection, and security as evidenced by accreditation to standards such as, for example, ISO/IEC 27001, NHSP DSP Toolkit, Cyber Essentials. Whilst many IG requirements are met by the DAP implementation and processes, adherence to standards by requesting organisations demonstrates that staff will be aware of the need for IG and data protection and will have received appropriate training.

• The length of time that the data is required. This would normally be twelve months. Any extensions to that must be approved by the Research Committee.

• That any external approvals (CAG, NRES) have been granted. Evidence will be required before data is made available via the DAP and before the application is submitted to HQIP DARG for final approval.

The GDPR legal basis for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of data controller review ahead of approval. For most applications, no confidential data will be made available to applicants as datasets are pseudonymised prior to release (to the applicant or the DAP). In any application where data is not pseudonymised, the Common Law Duty of Confidentiality for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of the data controller’s review prior to approval. It is also required from the applicant to provide evidence of NHS ethics permissions or evidence that ethics is not required.

The primary consideration in the review of any application for data is that the aims of the project are commensurate with the research goals and priorities of the NJR. The current priority themes approved by the NJR Steering Committee are as follows:

- Device and technology assessment.

- Patient outcomes, safety, and risk management.

- Health economics and cost effectiveness.

- Surgeon education and training.

- Patient information and shared decision-making.

- Disease diagnosis and prognosis.

- Disease causation, prevention, and treatment.

- Injury or trauma in the setting of joint replacement.

- Healthy development and ageing in patients with joint disease and replacement.

All these themes are directly related to both the provision of healthcare and the promotion of health through the twin goals of improving patient safety and patient outcomes. An improvement in patient outcomes also leads to the provision of more cost effective services and reduces costs through a reduction in the need for revision surgery. If applications do not address these themes, they are likely to be rejected.

Any organisation with a legitimate use for the data can apply, the only stipulation being that any proposed project supports the goals of the NJR and that the organisation is capable of delivering a successful outcome within the terms and conditions set out be the NJR for the use of the data. The most likely sub-licensees will be:

- Universities in the UK.

- NHS England Trusts.

- Research Fellows appointed by professional societies and supervised by universities.

- Orthopaedic device manufactures (for technology assessment only).

Sub-licences will be granted on a cost recovery basis only.

All outputs are based on aggregated data: the NJR specifically forbids the use of record level data for publication, even if that data is anonymised. This means that the privacy of all service users is maintained. The NJR consent form and patient information leaflet explain that NJR data will be linked to other, specified data sets and made available to other organisations without any personal details included.

The territory of use for sublicenses will be limited to England and Wales.

A public register of NJR data releases is available here - http://www.njrcentre.org.uk/njrcentre/Research/Research-Portfolio

This public register includes information about where NHS England sourced datasets are shared.

The data controllers will take responsibility for the actions and omissions of all sub licensees and breach of a sub licence should automatically be regarded as breach of the Data Sharing Framework Contract with NHS England. In the event of termination or expiry of the Data Sharing Framework Contract between NHS England and the data controllers, all sub licences shall automatically terminate.

The outputs from the use of the data will vary each year, dependent upon the research and analysis priorities set by the NJR’s Research Committee and NJR’s Editorial Board and agreed by the NJR Steering Committee.

a) The main vehicle for publication is the NJR’s Annual Report which is published annually in September. In addition to the Annual Report, a Patient’s Guide to the Annual Report is also published and made available in both hard and electronic copy. The Annual Report and the patients guide can be accessed at http://www.njrreports.org.uk.

b) Additional analyses are undertaken throughout the year and these are published either on the NJR website or in relevant, professional medical journals. To date those journals have included The Lancet, the British Medical Journal, and the Journal of Bone and Joint Surgery. The NJR maintains a dedicated research section on its website and details of planned, approved, completed, and published analysis can be found at: http://www.njrcentre.org.uk/njrcentre/Research/ResearchPortfolio/tabid/313/Default.aspx

c) The outcomes of analyses are also disseminated at conferences and meetings of professional societies. These include the British Orthopaedic Association, the British Hip Society, the British Association for Surgery of the Knee, the British Elbow and Shoulder Society, and the British Orthopaedic Foot and Ankle Society. The NJR also holds two regional events per year which are aimed at those hospital staff responsible for collecting the data, providing them with an update on the work of the NJR, including its outputs.

d) Some outputs are also published through the NJR’s secure online reporting services. These services include NJR Clinician Feedback, NJR Management Feedback, and NJR Supplier Feedback which provide information for clinicians, Trust/hospital management, and the manufacturers/suppliers of orthopaedic devices respectively.

The NJR is constantly reviewing the ways in which it communicates with patients and the public more generally and are keen to improve its reach. Recent and planned activity include:

• working with NJR Patient Representatives to recruit new members to the NJR Patient Panel to broaden the scope of patient voices on the NJR

• work with relevant patient groups such as Versus Arthritis to ensure that messaging for people who have had or plan to have joint replacement surgery are well targeted

• produce a series of patient guides http://www.njrcentre.org.uk/njrcentre/Reports-Publications-and-Minutes/Public-and-Patient-Guide co-designed with patients

• to press release key developments to ensure appropriate media coverage. See example at https://www.dailymail.co.uk/health/article-7111723/Thinking-getting-new-hip-knee-Try-online-tool-tells-unique-risks.html and https://medicalxpress.com/news/2020-08-implant-choice-important-surgeon-skill.html

• to consult with the PPI group at the Royal National Orthopaedic Hospital, Stanmore about NJR’s data flows

• work with Understanding Patient Data to produce a video about the team's work https://understandingpatientdata.org.uk/case-study/monitoring-joint-replacement-surgery

• collaborating with HQIP and The Health Foundation as part of the Understanding Health Data Access (UHDA) programme which aims to improve information available publicly about secondary data uses

• make a newsletter available via the study website which the public are able to subscribe to. A blog posts from joint replacement patients will also be hosted on the NJR website.

• to remain active on social media platforms (Facebook, Twitter, Linkedin)

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

OUTPUTS FOR SUB-LICENSING:

The principal outputs relating to the sublicensing agreement will be publication in peer reviewed journals and presentation as speciality conferences. The nature of outputs is described by applicants in their application and reviewed by the NJR Research Committee and HQIP DARG. All outputs will be aggregated with small numbers suppressed in accordance with the HES analysis guide.

The exact nature of those outputs will vary depending on the project portfolio, but by way of example a list of the current applications in the pipeline are included below:.

a. The use of the Exeter Short Revision Stem in primary and revision total hip replacement (Royal Devon & Exeter NHS Foundation Trust; NJR:CivReg)

The Exeter Short Revision Stem which is a surgical implant used in hip replacement surgery, (44mm No00 L.125) was released in 2004 and has now been used in both primary and revision surgeries for more than 10 years. This stem is shorter and slimmer than other Exeter stems of the same offset.

Applicant Organisation is undertaking a local review of the survival of the Exeter Short Revision Stem (Exeter stem 44mm No00 L.125) in both primary and revision Total Hip Replacement (THR) in their unit. More importantly however, the performance of the stem in the National Joint Registry in both primary and revision hip arthroplasty needs to be assessed.

Researchers would like to investigate the national survival of prosthesis rates at 2 years, 5 years and 10 years compared with standard Exeter stems of the same offset.

b. Perioperative aspirin and prevention of prosthetic joint infection (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Some artificial joints become infected and need to be removed. This may occur within a short time such as a month after the surgical operation. Such infections may occur even though the skin is thoroughly cleansed, and antibiotics are given prior to the operation. In some cases, infections occur after a much longer period up to many years. When artificial joint infection does occur in the early post surgical period the bacteria that cause such infection typically come from the patient's skin. These bacteria are implanted into the deep part of the surgical wound during the operation to insert the artificial joint. The skin bacteria that most commonly cause these infections are staphylococci. When these and other bacteria come into contact with an artificial joint they form a slime that makes it very difficult for the antibiotics used to treat infection to penetrate and kill them. This slime is called a biofilm. It has been shown in the test tube that aspirin can reduce biofilm formation. For this reason, the applicant organisation believes that it may be possible that aspirin given to patients before their artificial joint surgery may help to prevent infections requiring removal of the joint. The applicant organisation is able to use the records kept in the National Joint registry to measure whether there is an effect of aspirin by comparing the numbers of infections in patients treated with this compared with other medicines to prevent development of clots in the leg veins that may complicate artificial joint surgery. If it seems that aspirin prevents these infections more studies including clinical trials could be done to work out if more patients having artificial joint surgery should be given this commonly used and very safe drug.

c. Distribution and determinants of American Society of Anesthesiologists ( ASA) grade amongst patients undergoing hip arthroplasty: Data from an International Consortium of Joint Registries (University of Oxford; NJR:CivReg:HES)

Total hip replacement is performed to relieve pain and restore function in thousands of patients with hip arthritis each year. Whether or not a total hip replacement has a successful outcome may be affected by a patient’s age, gender and body mass index at the time of surgery. The outcome of surgery may also be affected if a patient has other long-term health disorders, so called ‘comorbidities’, in addition to hip arthritis. For example, previous studies have shown that patients with more co-morbidities are more likely to have complications after total hip replacement, such as infections. The outcomes of total hip replacements in different countries are often compared to provide insights into healthcare in different counties. So that the comparisons are fair, it is necessary to account for differences in the patient's undergoing surgery.

To date, there have been a small number of studies that have investigated how patients undergoing surgery in different countries compare. One previous study found that the age and level of education of patients in Switzerland having total hip replacement was quite different to that of patients having surgery in the USA. However, to date no study has investigated patients differ in terms of co-morbidities. The study propose to investigate how patients undergoing total hip replacements in different countries worldwide differ in terms of how many co-morbidities they have.

Joint replacement registries will be used as the source of information: these are specialised databases that have been created in many countries worldwide and hold information on patients having total hip replacement.

d. The effect of “allowable” unmatched component size on revision rates and time to 1st revision following primary TKR (Nottingham University Hospitals NHS Trust; NJR:CivReg)

Total knee replacements (TKR) contain a femoral and tibial component. Accurate sizing of both components is necessary in order to improve the range of motion of the knee replacement, minimise any soft tissue irritation and avoid bony overhang of the components. It is common practice to choose a tibial component size that matches the femoral component size. However, this does not always guarantee an appropriate level of joint motion and therefore surgeons may choose a size of the tibial component that could be smaller or larger than that of the femoral component which is “allowable” by the manufacturers of knee replacements.

This is not a rare event; a recent Australian study reported that in almost half of the TKRs the two components were not equal in size. This same study found that patients with a femoral component larger than the tibial component were more likely to undergo revision than those with components of equal size or those where the femoral component was smaller than the tibial component. The authors postulated that this might be due to excess stress being placed on the tibial component thus increasing the risk that the TKR would wear out earlier and need revising.

e. Outcomes of revision total hip replacement following ceramic bearing fracture: an avoidable patient safety issue? (University Hospitals of Leicester NHS Trust; NJR:CivReg)

Ceramic is a commonly used material for hip replacement surfaces, but rarely this material may break. When a ceramic bearing fractures this requires further surgery (a revision) to replace the damaged components, of which the surgeon has several materials available to choose from. There have been numerous reported cases of early failures and even death following revision to metal containing bearings after fractures of ceramic components. This is thought to be due to the abrasive effect of hard ceramic particles eroding the softer metal surfaces and subsequently causing metal poisoning. The risk of developing such complications are however unknown, therefore this study aims to quantify the risk of re-revision or early death following such an event.

f. Health Economics Evaluation of Primary Total Hip Arthroplasty by Bearing Type and Fixation Modality (The Royal Orthopaedic Hospital NHS Foundation Trust; NJR:HES:CivReg:PROMs)

The objective of this study is to determine which hip replacement implants have performed best as per the data held by the National Joint Registry. Researcher will use information about each patient undergoing joint revision. The information will include need for revision surgery and Patient Reported Outcome Measures (PROMs).

In addition to this it is necessary to examine the PROMs data for those who have undergone revision surgery (revision is where the joint replacement has been removed & a new joint replacement implanted). This will allow researchers to examine patient characteristics which may have led to the revision surgery as well as seeing how well the joint replacement was performing for that individual patient prior to revision surgery.

Having this full set of data including the Hospital Episode Statistics (HES) data for each patient (which is why traceability is essential for this data) will allow researchers to examine whether patient factors such as comorbidities or previous surgery influence the satisfaction and performance after hip replacement surgery. By grouping data into specific hip replacements (of which there are many available on the UK market) researchers can then see whether there are any outliers either performing particularly well or poorly and whether the performance and satisfaction with these hip replacements may be related to type of implant, comorbidities and similar. This study is a comprehensive analysis of the data available for England, Wales, Northern Ireland and the Isle of Man.

g. Evaluation of complex hip arthroplasty using the NJR dataset (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Hip replacement is one of the most common operations performed in the NHS and one of the most successful procedures in terms of improving quality of life for patients with arthritis. Whilst a hip replacement will last many years for the majority patients, some may run in to problems and require ‘re-do’ (revision) surgery such as when an infection develops, the components become loose or when the bone around the implant breaks. This ‘re-do’ surgery is often more complex than the original surgery and has a higher chance of running into problems including requiring further ‘re-do’ surgery.

The aim of this project it to obtain a perspective of the current delivery of this complex surgery at a national, regional and local levels. The study aim to report on the kinds of operations which are taking place, where they are happening, who is doing them and how many they are doing. This study forms the basis of a wider project aiming to generate data on the best way to structure provision of such surgeries in the future.

h. What is the effect of the femoral head material on revision rate following primary total hip replacement (THR) using the Exeter cemented stem? (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

There are several reasons why primary joint replacements fail. Most orthopaedic studies have looked at the type of fixation (cemented or uncemented) or type of articulations (metal or metal, metal on polyethylene or ceramic). What has not been looked at is the difference in the material between the stem and modular head combinations. There is greater understanding that metal ions can be released from this so called “trunnion wear” head-stem articulation and contribute to failure of joint replacements. This is due to corrosion in this area caused by difference in electrical charge of the materials.

Researchers wish to take advantage of the data available from the National Joint Registry to compare revision rates of different combinations of stem and head in terms of material and explore if this has any effect on the long-term survival (prosthesis) of joint replacements.

i. A comparison of patient reported outcomes and adverse events following elective total shoulder arthroplasty and hemiarthroplasty – an exploratory study (Imperial College London; NJR:CivReg)

Shoulder pain is a common reason people visit their general practitioner and the number of shoulder replacements performed in England, Wales and Northern Ireland has doubled since 2012. A joint replacement is major surgery and it is associated with significant risks. It’s important that patients and surgeons know which type of replacement leads to the best results and the fewest complications. Young patients who are considering a shoulder replacement have a higher chance of further operations, this group need specific information to guide treatment. When people have a joint replacement pseudonymised information about the surgery is stored in the joint registry. In the years following surgery they are asked to complete questionnaires about their symptoms, function and quality of life. The aim is to use this large volume of data to compare the two main types of joint replacement performed for arthritis of the shoulder.

The applicant organisation is requesting data for all shoulder replacements performed since the national joint registry began collection in 2012. This will be linked to the questionnaires given to patients after their procedure and to the NHS database of patient admissions to hospital. The information remains pseudonymised; a patient number is used only to ensure all the information corresponds to the same patient. Researchers will compare the results of patients who have received a half shoulder replacement and total shoulder replacement. The work will influence the decision making of patients and medical staff. The proposed studies will provide information about the expected recovery and the risks and benefits of a shoulder replacement. The work aims to improve patients’ shoulder function after surgery and reduce the number of additional procedures, complications and hospital admissions.

j. Effect of COVID-19 on NJR Surgical Practice and Outcomes (University of Sheffield; NJR:HES:PROMs:CivReg)

COVID 19 has had a great impact on all of our lives and changed society in ways that are yet to be identified. In this work, researchers will look at the outcomes of joint replacement surgery before COVID and then after COVID. This will allow researchers to see how outcomes of joint replacement surgery differ before and after the start of the COVID pandemic by looking at pain and activity, death and repeat operation.

To do this, researchers will compare these measures over 2 years immediately before COVID and compare these to the 2 years after restart of routine operating. In making these analyses researchers will use a range of mathematical tools to best fit the information the study receive from the NJR. One of the approaches researchers will use is called “machine learning” in which the computer “learns” to predict the outcome based on the information put in.

The other methods researchers will use are more traditional statistical approaches. As part of the study researchers will look to see which method makes the best predictions. Researchers imagine that the extra waiting time might affect these outcomes, as might surgeon deskilling.

However, researchers cannot measure these things directly in the NJR, nor see if any patients had COVID around the time of their operation. Findings will be compared with 2 other registers that do record COVID positive tests. The work is exploratory, to see if there are differences in outcomes before and after COVID, and if these are found researchers will create “hypotheses” to explain the possible reason for the effects and test them in further patients

k. Periprosthetic fractures around primary total hip replacement (University of Leeds; NJR:CivReg)

Total Hip replacements (THR) are an effective treatment for managing end-stage hip arthritis. The implants used come in many different shapes and sizes, and work in different ways. The implants are designed to last as long as possible, but a small amount fail because of loosening, infection, instability or breakage of the thigh bone (PFF) amongst other causes.

In patients older than 70, PFF is the second most common cause of failure of THR.

PFF leads to significant problems for patients (more surgery, higher risk of medical complications including death) and their incidence is increasing. Unfortunately, PFF is difficult to prevent and one of the strongest predictors is the type of implants surgeons use. The NJR records when patients with PFF have their implants exchanged (revision).

Previously, researchers identified a number of implant designs which make PFF more likely. This is already making an impact to reduce the risk of PFF. Unfortunately, previous data does not fully explain what happens to patients because patients can also be treated with no surgery and fixation surgery, which are not recorded in the NJR. In order to get a full picture of how the implants affect the risk of PFF researchers need to get information from more sources, which includes implant data (from the NJR), hospital data (which tells us about PFF not recorded in the NJR) and death statistics. This will make sure researchers can work out the complete picture of how the implants used affect the risk of PFF and ultimately how practice can be changed to prevent PFF from harming patients.

This will enable researchers to look at the most complete picture of PFF; how and when it occurs, how it affects patients and identify things which w can be changed to make PFF less likely in future. This will be the largest study looking at PFF. Although this research does not seek to make a fail-safe conclusion about which implant is safest; it will help the surgeons, patients and other stakeholders make informed decision in choosing the right implant for the right patient.

The following benefits to date include, but are not limited to:

Providing outcomes data to the Medicines and Healthcare products Regulatory Agency (MHRA) that has led to the withdrawal of a number of orthopaedic devices from the market.

Quickly identifying to hospitals those patients affected by a Field Notice or Device Alert issued by the MHRA.

Direct influences on changes to clinical practice, including:

• Preventing the use of hip resurfacing procedures in women over the age of 50 (reported through the NJRs Annual Report).

• Getting It Right First Time (GIRFT) proscribing the use of cemented hip implants in patients over 70 years of age (reported in the NJR Annual Report, supplemented by further analysis, and added to Best Practice Tariff (BPT) targets).

• Undertaking analysis that significantly reduced the number of large head, metal-on-metal hip replacements being undertaken.

• Undertaking analysis that resulted in a significant reduction in the number of hip replacement using a metal-on-metal articulating bearing in hip replacement.

• Showing that patella resurfacing at the time of total knee replacement has a better long-term outcome than not resurfacing.

• Professional societies, including the British Orthopaedic Association (BOA), issuing guidance to members about the minimum number of procedure types of procedure to be undertaken in a year in order to remain ‘current’. This follows an analysis of outcomes based on the number of procedures carried out by surgeons.

• A realignment of services for shoulder, ankle, and elbow joint replacement to a ‘hub and spoke’ service delivery on advice from GIRFT, based on an analysis of outcomes and the effects of low volume surgeons.

The provision of enhanced data and reporting services to all stakeholders enabling them to make decisions about the care provided to patients. Stakeholders include clinicians, device manufacturers, service providers and commissioners, patients, regulators, and advisory bodies such as GIRFT and National Institute for Health and Care Excellence (NICE).

Reducing the cost of orthopaedic devices to many NHS Trusts by benchmarking outcomes data against the price paid for implants by those Trusts.

Using risk-adjusted outcomes analyses to identify suspected outlier performance in surgeons, hospitals, and Trusts. This has enabled the appropriate responsible organisations and individuals to take the action necessary to rectify individual or system issues affecting outcomes. Such resolution has included the re-training of surgeons in certain procedure types and surgeons agreeing not to undertake specific procedure types. The NJR, in conjunction with the BOA, provides support to hospitals requesting it.

By publishing and sharing data on outcomes, the NJR has informed changes in clinical practice that have reduced the rate of revision surgery. A reduction in the rate of revision surgery evidence improvements in patient outcomes and safety and also reduces the cost to the NHS of such surgery.

The National Joint Registry (NJR) is commissioned by the Healthcare Quality Improvement Partnership (HQIP) on behalf of NHS England (NHSE) as part of the Clinical Audit and Patient Outcomes Programme (NCAPOP). The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety.

This agreement has Joint Controllership between HQIP and NHSE.

The NCAPOP is a large programme of circa 35 projects consisting of National Clinical Audits. HQIP is commissioned by NHSE to commission and manage the NCAPOP. NHSE is a controller of the NCAPOP jointly with HQIP as together both organisations determine the purposes and means of processing. NHSE is responsible for determining which projects/topics are included as part of the NCAPOP. HQIP, as commissioner of the NCAPOP, is responsible for project specification development, procurement and extension activities, contract management and authorising data sharing requests. NHS England, as a funder of the NCAPOP, participates within specification development, procurement and project extension activities and authorises the publication of project outputs.

NHSE is involved with developing the scope and purpose of the NCAPOP projects through participation within specification development activities and may authorise (as chair of the specification development meetings) the final project specifications. These specifications set out the purpose of the project, the patient groups and clinical services to evaluate and the types of data to collect. NHSE are a representative upon the HQIP Data access request group which authorises data sharing applications from third parties.

LEGAL BASIS JUSTIFICATION:

HQIP and NHSE both rely on the Article 6 (1) (e) legal basis under GDPR - "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". This is justified through commissioning arrangements which link back to NHS England and other national bodies with statutory responsibilities to improve quality of health care services.

HQIP rely on Article 9 (2) (i) as the legal basis for processing under GDPR - "processing is necessary for reasons of public interest in the area of public health, such as protecting against serious cross-border threats to health or ensuring high standards of quality and safety of health care and of medicinal products or medical devices, on the basis of Union or Member State law which provides for suitable and specific measures to safeguard the rights and freedoms of the data subject, in particular professional secrecy". This is justified as all projects aim to drive improvements in the quality and safety of care and to improve outcomes for patients.

NHSE rely on Article 9(2)(h) of the GDPR as the legal basis for processing. "Processing is necessary for the purposes of preventive or occupational medicine, for the assessment of the working capacity of the employee, medical diagnosis, the provision of health or social care or treatment or the management of health or social care systems and services on the basis of Union or Member State law or pursuant to contract with a health professional and subject to the conditions and safeguards referred to in paragraph 3". NHS England are responsible for provision of health and social care, and management of systems and compliance.

The NJR was established in 2003 by the Department of Health following a National Audit Office (NAO) report into the higher than expected failure rate of the 3M hip replacement device. The NAO report concluded that, had a national register of hip replacements been in existence, the failure rate would have been detected earlier. Earlier identification would have meant less patients were affected and the costs of revision surgery for the NHS would have been considerably less. The NJR went live in April 2003. Since its establishment, the responsibility for delivering the NJR has passed to HQIP who deliver the National Joint Registry.

HQIP and NHSE are joint controllers for the NJR and contract to two other organisations for the delivery of the programme, both of which are processors:

• NEC Software Solutions (UK) Limited; (a software and outsourcing business) is responsible for the NJR’s data collection and processing activities, including data storage and the provision of stakeholder reporting services.

• The University of Bristol (UoB) is also a processor with responsibilities for statistical analysis and reporting.

HQIP and NHSE, have no access to the record level data. NEC Software Solutions (UK) Limited has access to record level data, including patient identifiers, in order to link the data provided to patient records held by the NJR. The linked dataset is pseudonymised by NEC Software Solutions (UK) Limited before it is made available to the statistical analysis team at the UoB. Although the UoB team has access to record level data, it does not have access to patient identifiers. The work undertaken by the NJR is to monitor the outcomes of hip, knee, shoulder, ankle, and elbow joint replacement surgery with regards to the performance of devices, surgical teams, and Trusts and hospitals. The monitoring is necessary to ensure patient safety and improve patient outcomes.

The overall purpose of the NJR is summarised in its mission statement:

‘The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety. In a continuous drive to improve the quality of outcomes and ensure the quality and cost effectiveness of joint replacement surgery, the NJR will monitor and report on outcomes, and support and enable related research.’

The strategic goals of the NJR are as follows:

• To monitor the outcomes achieved by brand of prosthesis, hospital and surgeon, and highlight where these fall below an expected performance in order to allow prompt investigation and to support follow-up action.

• To inform patients, clinicians, providers and commissioners of healthcare, regulators and implant suppliers of the outcomes achieved in joint replacement surgery.

• To evidence variations in outcome achieved across surgical practice in order to inform best practice.

• To enhance patient awareness of joint replacement outcomes to better inform patient choice and patients’ quality of experience through engagement with patients and patient organisations.

• To support evidence-based purchasing of joint replacement implants for healthcare providers to support quality and cost effectiveness.

• To support suppliers in the routine post-market surveillance of implants and provide information to clinicians, patients, hospital management and the regulatory authorities.

The NJR has been collecting data since 2003 and is an ongoing audit with no planned end date. The NJR’s Steering Committee reviews the NJR’s mission statement and strategic goals annually to ensure that they remain relevant to existing stakeholder priorities, clinical guidance, and legislation.

The NJR requires record level data from Hospital Episode Statistics (HES), Patient Reported Outcome Measures (PROMS) and Civil Registrations of Death data from NHS England (formerly NHS Digital) to help achieve some of NJR's strategic goals. The NJR creates annual, linked dataset comprising data from the NJR, HES, Patient Episode Database Wales (PEDW), PROMs (NHS England programme) and Civil Registration. This dataset is used primarily for the production of the NJR’s Annual Report, its associated outcomes analyses, and in some reporting services provided to stakeholders. The data is also used for specific research projects undertaken by the NJR. By linking to HES, PROMs and Civil Registration data, the NJR is able to enhance the quality and type of analyses that it undertakes. The priorities for NJR analyses and research are set by the NJR’s Research Committee and Editorial Board, both of which report to the NJR Steering Committee.

Research committee members (including lay members) will review all application statements on benefits to health and care. In all cases where the applicant and/or the study funding are commercial in nature, careful consideration will be given to weigh the expected benefits against any commercial gain to the commercial organisation. Commercial sublicensees must evidence potential benefit to health and care, proportionate to any commercial gain that may be derived from receiving the sublicensed data.

The data sets requested by the NJR consist of HES Patient Admitted Care, Civil Registrations of Death, and Patient Reported Outcomes Measures. Patient identifiers are required in order to link the data requested from NHS England to patient identifiable data held by the NJR. This linkage is necessary to undertake analysis at a record level when, for example, considering factors such as length of stay for different joint replacement types, PROMs associated with different devices/implants, or outcomes associated with deprivation.

Civil Registrations of Death Data is essential for any outcomes analyses: patients who have died with an implant intact, for example, have to be excluded from any cohort. Outcomes analyses also includes a consideration of time to death from operation, and the cause of death.

HES Patient Admitted Care data is required to look at potential factors affecting the outcomes of joint replacement surgery, such as length of stay, re-admission to theatre, associated diagnoses and co-morbidities, deprivation, geography and location.

The NJR currently uses two endpoints in outcomes analysis: death and revision. PROMs represents an invaluable, third endpoint, providing a patients view, as opposed to a clinical view, of the outcome of joint replacement surgery. The NHS England programme is the only source of PROMs data relating to hip and knee joint replacement surgery.

The NJR requires data from 2003 which is when it started its own data collection. The number of years being requested is justified on the need to undertake longitudinal analyses of the available data. Currently, recommended guidance on the expected survivorship of joint replacement (as published by NICE) is based on ten years of data. Long term outcomes analyses are required.

Patient identifiable data is required to link the record level HES/Civil Registration/PROMs data to the relevant patient procedure in the NJR. The primary data field used for linkage is the patient's NHS number, but other fields such as date of birth, local patient ID, postcode, and side of procedure are used to ensure a reliable linkage. Data sets used for analyses of the linked data set have patient identifiers removed.

In some circumstances, it is not possible for the NJR to determine whether a patient has consented or not (‘unknown’ cohort). This might be because their operation was the result of an emergency admission, or because the admitting hospital has been unable to ask the patient to consent for some other reason. For this cohort, the NJR has support under Section 251 (s251) to collect and process data. Research into the ‘unknown’ cohort comes under purpose 2 of the s251 support; and that there is sufficient oversight within the NJR internal assurance procedures to ensure that any research using the ‘unknown’ cohort’s data is within the parameters of purpose 2 of the s251 support.

The sub-cohort received from NHS England where the patient has declined consent do not get linked with NJR and are not shared with applicants under sublicensing, however, the sub-cohort received from NHS England where consent is unknown and there is s251 support for are linked with NJR and do form part of the dataset that are shared with applicants.

In order to achieve minimisation, the NJR is requesting data for those OPCS4 codes relating to hip, knee, ankle, elbow, and shoulder joint replacement only. This will also ensure that Civil Registration data relating to those procedures only will be provided. Requesting data related to specific OPCS4 codes will ensure that the data requested is appropriate to the NJR's needs. Minimisation will be achieved by focusing on specific procedure codes.

• Linkage to HES data enables the NJR to improve the type and quality of the analyses that it can undertake without having to collect the data itself. Such linkage may be length of stay linked to outcomes or co-morbidities linked to outcomes. The NJR has, for example, published a paper examining the increased risk of cancer to patients who have had hip replacement procedures involving the use of metal on metal bearing surfaces, i.e. the femoral head and the acetabular cup or liner are both made of metal.

• By linking to PROMs outcomes analyses will improve as poor performance will be detected even if the primary joint replacement procedure has not been revised. A patient’s view of the outcome of joint replacement is a key indicator to the reporting of outcomes and provides an additional endpoint in addition to revision and death. Linkage to PROMs, and analysis at the record level, enables the NJR to monitor the performance of hospitals, surgeons, and individual implants.

• The NJR is required to publish 90 day mortality rates for all those NHS England Trusts undertaking joint replacement surgery in its Annual Clinical Reports to Trusts, the NJR’s Annual Report and as part of the former NHS England Clinical Outcomes Programme (although COP does not exist in its original form, the NJR still publishes annual indicators. The data, originally shared with Choices, is now shared with the CQC). Date of death and any associated co-morbidities are essential for outcomes analyses and for risk-adjusting mortality analyses.

This will enable the NJR to continue to improve its analysis and data quality audits, activities which are expressly intended to improve patient outcomes and patient safety.

The NJR has developed a secure, online Data Access Portal which provides additional security to the data used by researchers and analysts outside of the NJR. The plan is to provide subsets of linked data to approved research projects such that it is processed on the Portal and cannot be removed from the NJR’s security domain. The data does not contain patient identifiable fields.

Data from Wales, Northern Ireland, the Isle of Man, and Guernsey are sourced separately.

SUB-LICENSING (introduced under v8 of this agreement):

Researchers currently submit research applications to the NJR that require a linked cohort of HES, NJR, and Civil Registry data which requires the submission of two separate data applications to each controller: NHS England and HQIP. Both controllers must then undertake more processing to create the cohort required which includes the provision of patient identifiers for linkage and then adding the source data to those linked identifiers. The NJR would then supply patient identifiers to NHS England for the proposed cohort, the HES records and linked unique NJR ID would then be returned to the applicant, who must then supply the unique IDs to the NJR so that the NJR data can be supplied to the applicant and subsequently linked. This is time-consuming for controllers/processors and has deterred individuals and organisations from using the data to undertake much needed research and analyses. It also entails the transmission of patient identifiable data for each project.

The NJR already creates an annual linked dataset comprising of HES, Civil Registry, PROMs, and NJR data. Sub-licensing datasets extracted from this linked data set will save both NHS England and the NJR considerable effort, shorten timescales for applicants, make the data easily accessible for applicants and encourage greater use of the data for research. It will also reduce the processing and transmission of patient identifiable data. Such fully linked dataset is something NHS England do not routinely provide.

The NJR currently uses two endpoints for undertaking survivorship and outcomes analyses: revision and death. Whilst the NJR captures revision, the use of Civil Registry is essential to determine mortality status, data of death, age at death, and time to death following a joint replacement. Information about the cause of death is used to further refine the outcomes analysis.

The addition of PROMs data provides a third endpoint for the analysis of outcomes for knee and hip joint replacement and being able to measure health gain and improvements in quality of life can be used to support analyses into the outcomes associated with, for example, particular types of procedure or types of implant.

The NJR will not sub-licence HES/PROMs/Civil Registry data in its raw form to applicants. The data from NHS England will be linked to NJR data and used for approved research and analyses relating to joint replacement surgery and which support the NJR’s research priorities. The addition of HES/PROMS/Civil Registry data to data held by the NJR is necessary for all outcomes and survivorship analyses and provides a richer set of data: linking the data enables research that would otherwise be impossible. For example, NJR data provides detailed data on the medical devices used in the operation as well as specific information about surgical approach, throboprophylaxis and intra-operative complications. For shoulder surgery, NJR also collects a programme of Patient Recorded Outcome Measures (PROMs) which are not otherwise available from NHS England. The linked cohort is cleaned and is made ‘research ready’ prior to sub-licensing. This involves re-formatting the data in such a way that the meta-data is available to researchers and the data can easily be used in statistical analysis packages. It should be noted that only subsets of the cohort, relevant to the approved research, are made available to a researcher via a secure Data Access Portal. The cohort is agreed as part of the NJR’s approvals process.

It is anticipated that around 15 applications for sub-licensing would likely be approved each year. Of these, approximately 10 would require NJR data linked to Civil Registration data (date of death or time to death) and approximately 5 would require NJR data linked to HES and/or PROMS data. Sub-licensing, which will make the data more easily accessible, may lead to an increase in the number of applications for data.

The length of the licence is agreed as part of the approvals process and any extension to the agreement must be approved by the NJR’s Research Committee. The length of the sub-licence will depend upon the research being taken but would normally be for 12 months. The NJR has implemented a secure Data Access Portal (DAP) through which sub-licensees must access the data. Access to the data at the end of the licence period can be terminated via the DAP’s central administrative function. Cohorts of data used for research will be archived on the DAP, in line with current guidance, for a period of up to five years. This is necessary should the outcomes of, or the methods employed in, the research be challenged and the analysis needs to be repeated. Access to the archived data would be subject to approval by the Research Committee and DARG.

The approved organisations and researchers, who are granted an access to the linked data via the DAP, agree to terms and conditions of use, their rights and responsibilities as users of the linked data, as defined by the data controllers.

The NJR’s DAP will enable the implementation of some of NHS England’s requirements, especially with regard to the protection and security of the data. Researchers, for example, can only process the data on the DAP and all downloads are placed into a quarantine area prior to being approved for release by an appropriate system administrator. This ensures that the original data remains within the DAP’s security domain.

Sub license and approval groups:

• The NJR Research Committee is responsible to the NJR Steering Committee (NJRSC), and its objectives are:

- To protect NJR data (and any linked dataset) and strengthen its governance through safe effective, and efficient data management.

- To provide a single point of entry and management pathway for all research proposals and activity.

- Maximise access to the data for researchers.

- Align research activity to a framework of priority themes.

- To deliver the NJR’s research strategy through an expertise-based Research Committee.

• Specifically, the Research Committee is responsible for:

- The release of data for research using an impartial and objective protocol.

- Oversight of the use and reporting of data by research groups.

- Upholding the standard and consistency of work carried out using the data.

• All Expressions of Interest (EOI) are formally reviewed by the Research Committee and, following review, an applicant will either be invited to submit a formal application or informed that the NJR will not support the proposed project. Projects fall into two categories:

NJR Supported Project: this is an NJR supported project by a third part applicant. These will typically require aggregate or summary data, or pseudonymised or anonymised patient level data. These projects are external to, but supported by, the NJR. Requests requiring access to patient, surgeon, or unit personal identifiers, or implant batch number will not be granted. NJR will not provide personal identifiers or NJR linked datasets for external projects.

Recent examples include a study of the effect of cement type on the survivorship of cemented total hip replacement by University of Manchester; and examination of surgeon experience-related effects in patient outcomes in elective orthopaedic operations by Sussex University Hospitals NHS Trust.

NJR Partnership Project: this is a project delivered by a third party applicant with NJR involvement and oversight. These will be studies that require sensitive data items, flows of identifiable data, or data linked to external datasets, such as PROMS, HES, or Civil Registration data (requested through NHS England). These projects are undertaken in partnership with the NJR and require an identified named collaborator from the NJR Steering Committee or NJR Research Committee. NJR Partnership Projects require Principal Investigator attendance at a Research Committee meeting before an application can be approved.

Recent examples include an examination of the evaluation of complex hip arthroplasty in the United Kingdom by Northumbria Healthcare NHS Foundation Trust in collaboration with NJR; and the study of floor and ceiling effects of the Oxford Shoulder Score by University of Leicester in collaboration with NJR.

• Any projects requiring external approvals (e,g, CAG, NRES) must be channelled through the NJR application process in the first instance. Upon receipt of Research Committee and before data controller approval, the applicant will be required to provide all evidence of external approvals prior to data release.

• The Research Committee will consider the application in terms of its relevance to the NJR’s research priorities: whether the data requested is clinically appropriate to the proposed use; that the project is methodologically sound; whether aggregated or record level data is required; the applying organisation. In some instances, the Principal Investigator (PI) will be required to attend a Research Committee meeting in person. The NJR also stipulates that the PI must be a substantive employee of the requesting organisation and not be on a fixed-term or temporary contract.

• The use of the DAP will ensure the relevant security measures are adhered to for the sub-licensee’s processing and securing of the data. Data sets made for specific research projects can only be accessed and processed on the DAP by approved users. Physical and administrative processes prevent the download of the data and, if necessary, access to the data can be easily prevented by disabling the project area and its associated user accounts. All other due diligence for data requests (such as formal approval groups etc) will still be adhered to.

• Once a research project has been approved by the NJR Research Committee, the application is then submitted to HQIP’s Data Access Request Group (DARG) for review. Once DARG approval has been granted, the Principal Investigator will have an account set up on the DAP and the data will be loaded into a dedicated, private project area.

NJR Research Committee – Composition:

- NJRSC Public Health and Epidemiology Member (Chair).

- NJRSC Patient Representative Member.

- NJRSC Consultant Orthopaedic Surgeon Member (x4).

- Consultant Orthopaedic Surgeon (foot and ankle surgery expert).

- NJR Medical Director (Medical Advisory Committee representative).

- Expert in mechanical engineering.

- Health Economist.

- Professor of Clinical Epidemiology.

- Professor of Epidemiology.

- Consultant Senior Lecturer in Musculo-skeletal medicine

- NJR Director and Deputy Director of Operations.

- Data Quality Manager (data processing contractor representative)

- Principal Consultant (data processing contractor representative)

- HQIP Data Access Request Group - Composition

- Chief Executive HQIP

- Medical Director HQIP

- Data Protection Officer HQIP

- NHS England Representative

- Welsh Government Representative

- NJR Deputy Director of Operations

- NCAPOP Associate Director

The following Information Governance criteria are considered by the NJR Research Committee and the HQIP DARG:

• Could the project achieve its aims using already published data?

• Whether the application requires record-level data or whether it could achieve its aims using aggregated data.

• Are the data fields requested appropriate to the project or could they be reduced?

• The need, in certain circumstances, to suppress small numbers.

• Whether the project can achieve its aims using anonymised or pseudonymised data. Linked data provided by the DAP will not include data that could be used to identify a patient or surgeon.

• Ensuring that the requesting organisation adheres to appropriate processes, data protection, and security as evidenced by accreditation to standards such as, for example, ISO/IEC 27001, NHSP DSP Toolkit, Cyber Essentials. Whilst many IG requirements are met by the DAP implementation and processes, adherence to standards by requesting organisations demonstrates that staff will be aware of the need for IG and data protection and will have received appropriate training.

• The length of time that the data is required. This would normally be twelve months. Any extensions to that must be approved by the Research Committee.

• That any external approvals (CAG, NRES) have been granted. Evidence will be required before data is made available via the DAP and before the application is submitted to HQIP DARG for final approval.

The GDPR legal basis for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of data controller review ahead of approval. For most applications, no confidential data will be made available to applicants as datasets are pseudonymised prior to release (to the applicant or the DAP). In any application where data is not pseudonymised, the Common Law Duty of Confidentiality for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of the data controller’s review prior to approval. It is also required from the applicant to provide evidence of NHS ethics permissions or evidence that ethics is not required.

The primary consideration in the review of any application for data is that the aims of the project are commensurate with the research goals and priorities of the NJR. The current priority themes approved by the NJR Steering Committee are as follows:

- Device and technology assessment.

- Patient outcomes, safety, and risk management.

- Health economics and cost effectiveness.

- Surgeon education and training.

- Patient information and shared decision-making.

- Disease diagnosis and prognosis.

- Disease causation, prevention, and treatment.

- Injury or trauma in the setting of joint replacement.

- Healthy development and ageing in patients with joint disease and replacement.

All these themes are directly related to both the provision of healthcare and the promotion of health through the twin goals of improving patient safety and patient outcomes. An improvement in patient outcomes also leads to the provision of more cost effective services and reduces costs through a reduction in the need for revision surgery. If applications do not address these themes, they are likely to be rejected.

Any organisation with a legitimate use for the data can apply, the only stipulation being that any proposed project supports the goals of the NJR and that the organisation is capable of delivering a successful outcome within the terms and conditions set out be the NJR for the use of the data. The most likely sub-licensees will be:

- Universities in the UK.

- NHS England Trusts.

- Research Fellows appointed by professional societies and supervised by universities.

- Orthopaedic device manufactures (for technology assessment only).

Sub-licences will be granted on a cost recovery basis only.

All outputs are based on aggregated data: the NJR specifically forbids the use of record level data for publication, even if that data is anonymised. This means that the privacy of all service users is maintained. The NJR consent form and patient information leaflet explain that NJR data will be linked to other, specified data sets and made available to other organisations without any personal details included.

The territory of use for sublicenses will be limited to England and Wales.

A public register of NJR data releases is available here - http://www.njrcentre.org.uk/njrcentre/Research/Research-Portfolio

This public register includes information about where NHS England sourced datasets are shared.

The data controllers will take responsibility for the actions and omissions of all sub licensees and breach of a sub licence should automatically be regarded as breach of the Data Sharing Framework Contract with NHS England. In the event of termination or expiry of the Data Sharing Framework Contract between NHS England and the data controllers, all sub licences shall automatically terminate.

The outputs from the use of the data will vary each year, dependent upon the research and analysis priorities set by the NJR’s Research Committee and NJR’s Editorial Board and agreed by the NJR Steering Committee.

a) The main vehicle for publication is the NJR’s Annual Report which is published annually in September. In addition to the Annual Report, a Patient’s Guide to the Annual Report is also published and made available in both hard and electronic copy. The Annual Report and the patients guide can be accessed at http://www.njrreports.org.uk.

b) Additional analyses are undertaken throughout the year and these are published either on the NJR website or in relevant, professional medical journals. To date those journals have included The Lancet, the British Medical Journal, and the Journal of Bone and Joint Surgery. The NJR maintains a dedicated research section on its website and details of planned, approved, completed, and published analysis can be found at: http://www.njrcentre.org.uk/njrcentre/Research/ResearchPortfolio/tabid/313/Default.aspx

c) The outcomes of analyses are also disseminated at conferences and meetings of professional societies. These include the British Orthopaedic Association, the British Hip Society, the British Association for Surgery of the Knee, the British Elbow and Shoulder Society, and the British Orthopaedic Foot and Ankle Society. The NJR also holds two regional events per year which are aimed at those hospital staff responsible for collecting the data, providing them with an update on the work of the NJR, including its outputs.

d) Some outputs are also published through the NJR’s secure online reporting services. These services include NJR Clinician Feedback, NJR Management Feedback, and NJR Supplier Feedback which provide information for clinicians, Trust/hospital management, and the manufacturers/suppliers of orthopaedic devices respectively.

The NJR is constantly reviewing the ways in which it communicates with patients and the public more generally and are keen to improve its reach. Recent and planned activity include:

• working with NJR Patient Representatives to recruit new members to the NJR Patient Panel to broaden the scope of patient voices on the NJR

• work with relevant patient groups such as Versus Arthritis to ensure that messaging for people who have had or plan to have joint replacement surgery are well targeted

• produce a series of patient guides http://www.njrcentre.org.uk/njrcentre/Reports-Publications-and-Minutes/Public-and-Patient-Guide co-designed with patients

• to press release key developments to ensure appropriate media coverage. See example at https://www.dailymail.co.uk/health/article-7111723/Thinking-getting-new-hip-knee-Try-online-tool-tells-unique-risks.html and https://medicalxpress.com/news/2020-08-implant-choice-important-surgeon-skill.html

• to consult with the PPI group at the Royal National Orthopaedic Hospital, Stanmore about NJR’s data flows

• work with Understanding Patient Data to produce a video about the team's work https://understandingpatientdata.org.uk/case-study/monitoring-joint-replacement-surgery

• collaborating with HQIP and The Health Foundation as part of the Understanding Health Data Access (UHDA) programme which aims to improve information available publicly about secondary data uses

• make a newsletter available via the study website which the public are able to subscribe to. A blog posts from joint replacement patients will also be hosted on the NJR website.

• to remain active on social media platforms (Facebook, Twitter, Linkedin)

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

OUTPUTS FOR SUB-LICENSING:

The principal outputs relating to the sublicensing agreement will be publication in peer reviewed journals and presentation as speciality conferences. The nature of outputs is described by applicants in their application and reviewed by the NJR Research Committee and HQIP DARG. All outputs will be aggregated with small numbers suppressed in accordance with the HES analysis guide.

The exact nature of those outputs will vary depending on the project portfolio, but by way of example a list of the current applications in the pipeline are included below:.

a. The use of the Exeter Short Revision Stem in primary and revision total hip replacement (Royal Devon & Exeter NHS Foundation Trust; NJR:CivReg)

The Exeter Short Revision Stem which is a surgical implant used in hip replacement surgery, (44mm No00 L.125) was released in 2004 and has now been used in both primary and revision surgeries for more than 10 years. This stem is shorter and slimmer than other Exeter stems of the same offset.

Applicant Organisation is undertaking a local review of the survival of the Exeter Short Revision Stem (Exeter stem 44mm No00 L.125) in both primary and revision Total Hip Replacement (THR) in their unit. More importantly however, the performance of the stem in the National Joint Registry in both primary and revision hip arthroplasty needs to be assessed.

Researchers would like to investigate the national survival of prosthesis rates at 2 years, 5 years and 10 years compared with standard Exeter stems of the same offset.

b. Perioperative aspirin and prevention of prosthetic joint infection (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Some artificial joints become infected and need to be removed. This may occur within a short time such as a month after the surgical operation. Such infections may occur even though the skin is thoroughly cleansed, and antibiotics are given prior to the operation. In some cases, infections occur after a much longer period up to many years. When artificial joint infection does occur in the early post surgical period the bacteria that cause such infection typically come from the patient's skin. These bacteria are implanted into the deep part of the surgical wound during the operation to insert the artificial joint. The skin bacteria that most commonly cause these infections are staphylococci. When these and other bacteria come into contact with an artificial joint they form a slime that makes it very difficult for the antibiotics used to treat infection to penetrate and kill them. This slime is called a biofilm. It has been shown in the test tube that aspirin can reduce biofilm formation. For this reason, the applicant organisation believes that it may be possible that aspirin given to patients before their artificial joint surgery may help to prevent infections requiring removal of the joint. The applicant organisation is able to use the records kept in the National Joint registry to measure whether there is an effect of aspirin by comparing the numbers of infections in patients treated with this compared with other medicines to prevent development of clots in the leg veins that may complicate artificial joint surgery. If it seems that aspirin prevents these infections more studies including clinical trials could be done to work out if more patients having artificial joint surgery should be given this commonly used and very safe drug.

c. Distribution and determinants of American Society of Anesthesiologists ( ASA) grade amongst patients undergoing hip arthroplasty: Data from an International Consortium of Joint Registries (University of Oxford; NJR:CivReg:HES)

Total hip replacement is performed to relieve pain and restore function in thousands of patients with hip arthritis each year. Whether or not a total hip replacement has a successful outcome may be affected by a patient’s age, gender and body mass index at the time of surgery. The outcome of surgery may also be affected if a patient has other long-term health disorders, so called ‘comorbidities’, in addition to hip arthritis. For example, previous studies have shown that patients with more co-morbidities are more likely to have complications after total hip replacement, such as infections. The outcomes of total hip replacements in different countries are often compared to provide insights into healthcare in different counties. So that the comparisons are fair, it is necessary to account for differences in the patient's undergoing surgery.

To date, there have been a small number of studies that have investigated how patients undergoing surgery in different countries compare. One previous study found that the age and level of education of patients in Switzerland having total hip replacement was quite different to that of patients having surgery in the USA. However, to date no study has investigated patients differ in terms of co-morbidities. The study propose to investigate how patients undergoing total hip replacements in different countries worldwide differ in terms of how many co-morbidities they have.

Joint replacement registries will be used as the source of information: these are specialised databases that have been created in many countries worldwide and hold information on patients having total hip replacement.

d. The effect of “allowable” unmatched component size on revision rates and time to 1st revision following primary TKR (Nottingham University Hospitals NHS Trust; NJR:CivReg)

Total knee replacements (TKR) contain a femoral and tibial component. Accurate sizing of both components is necessary in order to improve the range of motion of the knee replacement, minimise any soft tissue irritation and avoid bony overhang of the components. It is common practice to choose a tibial component size that matches the femoral component size. However, this does not always guarantee an appropriate level of joint motion and therefore surgeons may choose a size of the tibial component that could be smaller or larger than that of the femoral component which is “allowable” by the manufacturers of knee replacements.

This is not a rare event; a recent Australian study reported that in almost half of the TKRs the two components were not equal in size. This same study found that patients with a femoral component larger than the tibial component were more likely to undergo revision than those with components of equal size or those where the femoral component was smaller than the tibial component. The authors postulated that this might be due to excess stress being placed on the tibial component thus increasing the risk that the TKR would wear out earlier and need revising.

e. Outcomes of revision total hip replacement following ceramic bearing fracture: an avoidable patient safety issue? (University Hospitals of Leicester NHS Trust; NJR:CivReg)

Ceramic is a commonly used material for hip replacement surfaces, but rarely this material may break. When a ceramic bearing fractures this requires further surgery (a revision) to replace the damaged components, of which the surgeon has several materials available to choose from. There have been numerous reported cases of early failures and even death following revision to metal containing bearings after fractures of ceramic components. This is thought to be due to the abrasive effect of hard ceramic particles eroding the softer metal surfaces and subsequently causing metal poisoning. The risk of developing such complications are however unknown, therefore this study aims to quantify the risk of re-revision or early death following such an event.

f. Health Economics Evaluation of Primary Total Hip Arthroplasty by Bearing Type and Fixation Modality (The Royal Orthopaedic Hospital NHS Foundation Trust; NJR:HES:CivReg:PROMs)

The objective of this study is to determine which hip replacement implants have performed best as per the data held by the National Joint Registry. Researcher will use information about each patient undergoing joint revision. The information will include need for revision surgery and Patient Reported Outcome Measures (PROMs).

In addition to this it is necessary to examine the PROMs data for those who have undergone revision surgery (revision is where the joint replacement has been removed & a new joint replacement implanted). This will allow researchers to examine patient characteristics which may have led to the revision surgery as well as seeing how well the joint replacement was performing for that individual patient prior to revision surgery.

Having this full set of data including the Hospital Episode Statistics (HES) data for each patient (which is why traceability is essential for this data) will allow researchers to examine whether patient factors such as comorbidities or previous surgery influence the satisfaction and performance after hip replacement surgery. By grouping data into specific hip replacements (of which there are many available on the UK market) researchers can then see whether there are any outliers either performing particularly well or poorly and whether the performance and satisfaction with these hip replacements may be related to type of implant, comorbidities and similar. This study is a comprehensive analysis of the data available for England, Wales, Northern Ireland and the Isle of Man.

g. Evaluation of complex hip arthroplasty using the NJR dataset (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Hip replacement is one of the most common operations performed in the NHS and one of the most successful procedures in terms of improving quality of life for patients with arthritis. Whilst a hip replacement will last many years for the majority patients, some may run in to problems and require ‘re-do’ (revision) surgery such as when an infection develops, the components become loose or when the bone around the implant breaks. This ‘re-do’ surgery is often more complex than the original surgery and has a higher chance of running into problems including requiring further ‘re-do’ surgery.

The aim of this project it to obtain a perspective of the current delivery of this complex surgery at a national, regional and local levels. The study aim to report on the kinds of operations which are taking place, where they are happening, who is doing them and how many they are doing. This study forms the basis of a wider project aiming to generate data on the best way to structure provision of such surgeries in the future.

h. What is the effect of the femoral head material on revision rate following primary total hip replacement (THR) using the Exeter cemented stem? (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

There are several reasons why primary joint replacements fail. Most orthopaedic studies have looked at the type of fixation (cemented or uncemented) or type of articulations (metal or metal, metal on polyethylene or ceramic). What has not been looked at is the difference in the material between the stem and modular head combinations. There is greater understanding that metal ions can be released from this so called “trunnion wear” head-stem articulation and contribute to failure of joint replacements. This is due to corrosion in this area caused by difference in electrical charge of the materials.

Researchers wish to take advantage of the data available from the National Joint Registry to compare revision rates of different combinations of stem and head in terms of material and explore if this has any effect on the long-term survival (prosthesis) of joint replacements.

i. A comparison of patient reported outcomes and adverse events following elective total shoulder arthroplasty and hemiarthroplasty – an exploratory study (Imperial College London; NJR:CivReg)

Shoulder pain is a common reason people visit their general practitioner and the number of shoulder replacements performed in England, Wales and Northern Ireland has doubled since 2012. A joint replacement is major surgery and it is associated with significant risks. It’s important that patients and surgeons know which type of replacement leads to the best results and the fewest complications. Young patients who are considering a shoulder replacement have a higher chance of further operations, this group need specific information to guide treatment. When people have a joint replacement pseudonymised information about the surgery is stored in the joint registry. In the years following surgery they are asked to complete questionnaires about their symptoms, function and quality of life. The aim is to use this large volume of data to compare the two main types of joint replacement performed for arthritis of the shoulder.

The applicant organisation is requesting data for all shoulder replacements performed since the national joint registry began collection in 2012. This will be linked to the questionnaires given to patients after their procedure and to the NHS database of patient admissions to hospital. The information remains pseudonymised; a patient number is used only to ensure all the information corresponds to the same patient. Researchers will compare the results of patients who have received a half shoulder replacement and total shoulder replacement. The work will influence the decision making of patients and medical staff. The proposed studies will provide information about the expected recovery and the risks and benefits of a shoulder replacement. The work aims to improve patients’ shoulder function after surgery and reduce the number of additional procedures, complications and hospital admissions.

j. Effect of COVID-19 on NJR Surgical Practice and Outcomes (University of Sheffield; NJR:HES:PROMs:CivReg)

COVID 19 has had a great impact on all of our lives and changed society in ways that are yet to be identified. In this work, researchers will look at the outcomes of joint replacement surgery before COVID and then after COVID. This will allow researchers to see how outcomes of joint replacement surgery differ before and after the start of the COVID pandemic by looking at pain and activity, death and repeat operation.

To do this, researchers will compare these measures over 2 years immediately before COVID and compare these to the 2 years after restart of routine operating. In making these analyses researchers will use a range of mathematical tools to best fit the information the study receive from the NJR. One of the approaches researchers will use is called “machine learning” in which the computer “learns” to predict the outcome based on the information put in.

The other methods researchers will use are more traditional statistical approaches. As part of the study researchers will look to see which method makes the best predictions. Researchers imagine that the extra waiting time might affect these outcomes, as might surgeon deskilling.

However, researchers cannot measure these things directly in the NJR, nor see if any patients had COVID around the time of their operation. Findings will be compared with 2 other registers that do record COVID positive tests. The work is exploratory, to see if there are differences in outcomes before and after COVID, and if these are found researchers will create “hypotheses” to explain the possible reason for the effects and test them in further patients

k. Periprosthetic fractures around primary total hip replacement (University of Leeds; NJR:CivReg)

Total Hip replacements (THR) are an effective treatment for managing end-stage hip arthritis. The implants used come in many different shapes and sizes, and work in different ways. The implants are designed to last as long as possible, but a small amount fail because of loosening, infection, instability or breakage of the thigh bone (PFF) amongst other causes.

In patients older than 70, PFF is the second most common cause of failure of THR.

PFF leads to significant problems for patients (more surgery, higher risk of medical complications including death) and their incidence is increasing. Unfortunately, PFF is difficult to prevent and one of the strongest predictors is the type of implants surgeons use. The NJR records when patients with PFF have their implants exchanged (revision).

Previously, researchers identified a number of implant designs which make PFF more likely. This is already making an impact to reduce the risk of PFF. Unfortunately, previous data does not fully explain what happens to patients because patients can also be treated with no surgery and fixation surgery, which are not recorded in the NJR. In order to get a full picture of how the implants affect the risk of PFF researchers need to get information from more sources, which includes implant data (from the NJR), hospital data (which tells us about PFF not recorded in the NJR) and death statistics. This will make sure researchers can work out the complete picture of how the implants used affect the risk of PFF and ultimately how practice can be changed to prevent PFF from harming patients.

This will enable researchers to look at the most complete picture of PFF; how and when it occurs, how it affects patients and identify things which w can be changed to make PFF less likely in future. This will be the largest study looking at PFF. Although this research does not seek to make a fail-safe conclusion about which implant is safest; it will help the surgeons, patients and other stakeholders make informed decision in choosing the right implant for the right patient.

The following benefits to date include, but are not limited to:

Providing outcomes data to the Medicines and Healthcare products Regulatory Agency (MHRA) that has led to the withdrawal of a number of orthopaedic devices from the market.

Quickly identifying to hospitals those patients affected by a Field Notice or Device Alert issued by the MHRA.

Direct influences on changes to clinical practice, including:

• Preventing the use of hip resurfacing procedures in women over the age of 50 (reported through the NJRs Annual Report).

• Getting It Right First Time (GIRFT) proscribing the use of cemented hip implants in patients over 70 years of age (reported in the NJR Annual Report, supplemented by further analysis, and added to Best Practice Tariff (BPT) targets).

• Undertaking analysis that significantly reduced the number of large head, metal-on-metal hip replacements being undertaken.

• Undertaking analysis that resulted in a significant reduction in the number of hip replacement using a metal-on-metal articulating bearing in hip replacement.

• Showing that patella resurfacing at the time of total knee replacement has a better long-term outcome than not resurfacing.

• Professional societies, including the British Orthopaedic Association (BOA), issuing guidance to members about the minimum number of procedure types of procedure to be undertaken in a year in order to remain ‘current’. This follows an analysis of outcomes based on the number of procedures carried out by surgeons.

• A realignment of services for shoulder, ankle, and elbow joint replacement to a ‘hub and spoke’ service delivery on advice from GIRFT, based on an analysis of outcomes and the effects of low volume surgeons.

The provision of enhanced data and reporting services to all stakeholders enabling them to make decisions about the care provided to patients. Stakeholders include clinicians, device manufacturers, service providers and commissioners, patients, regulators, and advisory bodies such as GIRFT and National Institute for Health and Care Excellence (NICE).

Reducing the cost of orthopaedic devices to many NHS Trusts by benchmarking outcomes data against the price paid for implants by those Trusts.

Using risk-adjusted outcomes analyses to identify suspected outlier performance in surgeons, hospitals, and Trusts. This has enabled the appropriate responsible organisations and individuals to take the action necessary to rectify individual or system issues affecting outcomes. Such resolution has included the re-training of surgeons in certain procedure types and surgeons agreeing not to undertake specific procedure types. The NJR, in conjunction with the BOA, provides support to hospitals requesting it.

By publishing and sharing data on outcomes, the NJR has informed changes in clinical practice that have reduced the rate of revision surgery. A reduction in the rate of revision surgery evidence improvements in patient outcomes and safety and also reduces the cost to the NHS of such surgery.

The National Joint Registry (NJR) is commissioned by the Healthcare Quality Improvement Partnership (HQIP) on behalf of NHS England (NHSE) as part of the Clinical Audit and Patient Outcomes Programme (NCAPOP). The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety. A longstanding agreement has been in place between NHS Digital, HQIP and NHSE, under version 8 of this agreement, approval is being sought on HQIP and NHSE being able to sub-license NHS Digital data.

This agreement has Joint Data Controllership between HQIP and NHSE.

The NCAPOP is a large programme of circa 35 projects consisting of National Clinical Audits. HQIP is commissioned by NHSE to commission and manage the NCAPOP. NHSE is a controller of the NCAPOP jointly with HQIP as together both organisations determine the purposes and means of processing. NHSE is responsible for determining which projects/topics are included as part of the NCAPOP. HQIP, as commissioner of the NCAPOP, is responsible for project specification development, procurement and extension activities, contract management and authorising data sharing requests. NHS England, as a funder of the NCAPOP, participates within specification development, procurement and project extension activities and authorises the publication of project outputs.

NHSE is involved with developing the scope and purpose of the NCAPOP projects through participation within specification development activities and may authorise (as chair of the specification development meetings) the final project specifications. These specifications set out the purpose of the project, the patient groups and clinical services to evaluate and the types of data to collect. NHSE are a representative upon the HQIP Data access request group which authorises data sharing applications from third parties.

LEGAL BASIS JUSTIFICATION:

HQIP and NHSE both rely on the Article 6 (1) (e) legal basis under GDPR - "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". This is justified through commissioning arrangements which link back to NHS England and other national bodies with statutory responsibilities to improve quality of health care services.

HQIP rely on Article 9 (2) (i) as the legal basis for processing under GDPR - "processing is necessary for reasons of public interest in the area of public health, such as protecting against serious cross-border threats to health or ensuring high standards of quality and safety of health care and of medicinal products or medical devices, on the basis of Union or Member State law which provides for suitable and specific measures to safeguard the rights and freedoms of the data subject, in particular professional secrecy". This is justified as all projects aim to drive improvements in the quality and safety of care and to improve outcomes for patients.

NHSE rely on Article 9(2)(h) of the GDPR as the legal basis for processing. "Processing is necessary for the purposes of preventive or occupational medicine, for the assessment of the working capacity of the employee, medical diagnosis, the provision of health or social care or treatment or the management of health or social care systems and services on the basis of Union or Member State law or pursuant to contract with a health professional and subject to the conditions and safeguards referred to in paragraph 3". NHS England are responsible for provision of health and social care, and management of systems and compliance.

The NJR was established in 2003 by the Department of Health following a National Audit Office (NAO) report into the higher than expected failure rate of the 3M hip replacement device. The NAO report concluded that, had a national register of hip replacements been in existence, the failure rate would have been detected earlier. Earlier identification would have meant less patients were affected and the costs of revision surgery for the NHS would have been considerably less. The NJR went live in April 2003. Since its establishment, the responsibility for delivering the NJR has passed to HQIP who deliver the National Joint Registry.

HQIP and NHSE are joint data controllers for the NJR and contract to two other organisations for the delivery of the programme, both of which are data processors:

• NEC Software Solutions (UK) Limited;. a software and outsourcing business) is responsible for the NJR’s data collection and processing activities, including data storage and the provision of stakeholder reporting services.

• The University of Bristol (UoB)is also a data processor with responsibilities for statistical analysis and reporting.

HQIP and NHSE, have no access to the record level data. NEC Software Solutions (UK) Limited. has access to record level data, including patient identifiers, in order to link the data provided to patient records held by the NJR. The linked dataset is pseudonymised by NEC Software Solutions (UK) Limited. before it is made available to the statistical analysis team at the UoB. Although the UoB team has access to record level data, it does not have access to patient identifiers. The work undertaken by the NJR is to monitor the outcomes of hip, knee, shoulder, ankle, and elbow joint replacement surgery with regards to the performance of devices, surgical teams, and Trusts and hospitals. The monitoring is necessary to ensure patient safety and improve patient outcomes.

The overall purpose of the NJR is summarised in its mission statement:

‘The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety. In a continuous drive to improve the quality of outcomes and ensure the quality and cost effectiveness of joint replacement surgery, the NJR will monitor and report on outcomes, and support and enable related research.’

The strategic goals of the NJR are as follows:

• To monitor the outcomes achieved by brand of prosthesis, hospital and surgeon, and highlight where these fall below an expected performance in order to allow prompt investigation and to support follow-up action.

• To inform patients, clinicians, providers and commissioners of healthcare, regulators and implant suppliers of the outcomes achieved in joint replacement surgery.

• To evidence variations in outcome achieved across surgical practice in order to inform best practice.

• To enhance patient awareness of joint replacement outcomes to better inform patient choice and patients’ quality of experience through engagement with patients and patient organisations.

• To support evidence-based purchasing of joint replacement implants for healthcare providers to support quality and cost effectiveness.

• To support suppliers in the routine post-market surveillance of implants and provide information to clinicians, patients, hospital management and the regulatory authorities.

The NJR has been collecting data since 2003 and is an ongoing audit with no planned end date. The NJR’s Steering Committee reviews the NJR’s mission statement and strategic goals annually to ensure that they remain relevant to existing stakeholder priorities, clinical guidance, and legislation.

The NJR requires record level data from Hospital Episode Statistics (HES), Patient Reported Outcome Measures (PROMS) and Civil Registration (Deaths) data from NHS Digital to help achieve some of NJRs strategic goals. The NJR creates annual, linked dataset comprising data from the NJR, HES, Patient Episode Database Wales (PEDW), PROMs (NHS England programme) and Civil Registration. This dataset is used primarily for the production of the NJR’s Annual Report, its associated outcomes analyses, and in some reporting services provided to stakeholders. The data is also used for specific research projects undertaken by the NJR. By linking to HES, PROMs and Civil Registration data, the NJR is able to enhance the quality and type of analyses that it undertakes. The priorities for NJR analyses and research are set by the NJR’s Research Committee and Editorial Board, both of which report to the NJR Steering Committee.

Research committee members (including lay members) will review all application statements on benefits to health and care. In all cases where the applicant and/or the study funding are commercial in nature, careful consideration will be given to weigh the expected benefits against any commercial gain to the commercial organisation. Commercial sublicensees must evidence potential benefit to health and care, proportionate to any commercial gain that may be derived from receiving the sublicensed data.

The data sets requested by the NJR consist of HES Patient Admitted Care, Civil Registration (Deaths), and Patient Reported Outcomes Measures. Patient identifiers are required in order to link the data requested from NHSD to patient identifiable data held by the NJR. This linkage is necessary to undertake analysis at a record level when, for example, considering factors such as length of stay for different joint replacement types, PROMs associated with different devices/implants, or outcomes associated with deprivation.

Civil Registration Data (Deaths) is essential for any outcomes analyses: patients who have died with an implant intact, for example, have to be excluded from any cohort. Outcomes analyses also includes a consideration of time to death from operation, and the cause of death.

HES Patient Admitted Care data is required to look at potential factors affecting the outcomes of joint replacement surgery, such as length of stay, re-admission to theatre, associated diagnoses and co-morbidities, deprivation, geography and location.

The NJR currently uses two endpoints in outcomes analysis: death and revision. PROMs represents an invaluable, third endpoint, providing a patients view, as opposed to a clinical view, of the outcome of joint replacement surgery. The NHS England programme is the only source of PROMs data relating to hip and knee joint replacement surgery.

The NJR requires data from 2003 which is when it started its own data collection. The number of years being requested is justified on the need to undertake longitudinal analyses of the available data. Currently, recommended guidance on the expected survivorship of joint replacement (as published by NICE) is based on ten years of data. Long term outcomes analyses are required.

Patient identifiable data is required to link the record level HES/Civil Registration/PROMs data to the relevant patient procedure in the NJR. The primary data field used for linkage is the patient's NHS number, but other fields such as date of birth, local patient ID, postcode, and side of procedure are used to ensure a reliable linkage. Data sets used for analyses of the linked data set have patient identifiers removed.

In some circumstances, it is not possible for the NJR to determine whether a patient has consented or not (‘unknown’ cohort). This might be because their operation was the result of an emergency admission, or because the admitting hospital has been unable to ask the patient to consent for some other reason. For this cohort, the NJR has support under S251 to collect and process data. Research into the ‘unknown’ cohort comes under purpose 2 of the s251 support; and that there is sufficient oversight within the NJR internal assurance procedures to ensure that any research using the ‘unknown’ cohort’s data is within the parameters of purpose 2 of the s251 support.

The sub-cohort received from NHS Digital where the patient has declined consent do not get linked with NJR and are not shared with applicants under sublicensing, however, the sub-cohort received from NHSD where consent is unknown and there is S251 support for are linked with NJR and do form part of the dataset that are shared with applicants.

In order to achieve minimisation, the NJR is requesting data for those OPCS4 codes relating to hip, knee, ankle, elbow, and shoulder joint replacement only. This will also ensure that Civil Registration data relating to those procedures only will be provided. By requesting data related to specific OPCS4 codes will ensure that the data requested is appropriate to the NJR's needs. Minimisation will be achieved by focusing on specific procedure codes.

• Linkage to HES data enables the NJR to improve the type and quality of the analyses that it can undertake without having to collect the data itself. Such linkage may be length of stay linked to outcomes or co-morbidities linked to outcomes. The NJR has, for example, published a paper examining the increased risk of cancer to patients who have had hip replacement procedures involving the use of metal on metal bearing surfaces, i.e. the femoral head and the acetabular cup or liner are both made of metal.

• By linking to PROMs outcomes analyses will improve as poor performance will be detected even if the primary joint replacement procedure has not been revised. A patient’s view of the outcome of joint replacement is a key indicator to the reporting of outcomes and provides an additional endpoint in addition to revision and death. Linkage to PROMs, and analysis at the record level, enables the NJR to monitor the performance of hospitals, surgeons, and individual implants.

• The NJR is required to publish 90 day mortality rates for all those NHS England Trusts undertaking joint replacement surgery in its Annual Clinical Reports to Trusts, the NJR’s Annual Report and as part of the former NHS England Clinical Outcomes Programme (although COP does not exist in its original form, the NJR still publishes annual indicators. The data, originally shared with Choices, is now shared with the CQC). Date of death and any associated co-morbidities are essential for outcomes analyses and for risk-adjusting mortality analyses.

This will enable the NJR to continue to improve its analysis and data quality audits, activities which are expressly intended to improve patient outcomes and patient safety.

The NJR has developed a secure, online Data Access Portal which provides additional security to the data used by researchers and analysts outside of the NJR. The plan is to provide subsets of linked data to approved research projects such that it is processed on the Portal and cannot be removed from the NJR’s security domain. The data does not contain patient identifiable fields.

Data from Wales, Northern Ireland, the Isle of Man, and Guernsey are sourced separately.

SUB-LICENSING (introduced under v8 of this agreement):

Researchers currently submit research applications to the NJR that require a linked cohort of HES, NJR, and Civil Registry data which requires the submission of two separate data applications to each data controller: NHS Digital and HQIP. Both data controllers must then undertake more processing to create the cohort required which includes the provision of patient identifiers for linkage and then adding the source data to those linked identifiers. The NJR would then supply patient identifiers to NHS Digital for the proposed cohort, the HES records and linked unique NJR ID would then be returned to the applicant, who must then supply the unique IDs to the NJR so that the NJR data can be supplied to the applicant and subsequently linked. This is time-consuming for controllers/processors and has deterred individuals and organisations from using the data to undertake much needed research and analyses. It also entails the transmission of patient identifiable data for each project.

The NJR already creates an annual linked dataset comprising of HES, Civil Registry, PROMs, and NJR data. Sub-licensing datasets extracted from this linked data set will save both NHS Digital and the NJR considerable effort, shorten timescales for applicants, make the data easily accessible for applicants and encourage greater use of the data for research. It will also reduce the processing and transmission of patient identifiable data. Such fully linked dataset is something NHS Digital do not routinely provide.

The NJR currently uses two endpoints for undertaking survivorship and outcomes analyses: revision and death. Whilst the NJR captures revision, the use of Civil Registry is essential to determine mortality status, data of death, age at death, and time to death following a joint replacement. Information about the cause of death is used to further refine the outcomes analysis.

The addition of PROMs data provides a third endpoint for the analysis of outcomes for knee and hip joint replacement and being able to measure health gain and improvements in quality of life can be used to support analyses into the outcomes associated with, for example, particular types of procedure or types of implant.

The NJR will not sub-licence HES/PROMs/Civil Registry data in its raw form to applicants. The data from NHS Digital will be linked to NJR data and used for approved research and analyses relating to joint replacement surgery and which support the NJR’s research priorities. The addition of HES/PROMS/Civil Registry data to data held by the NJR is necessary for all outcomes and survivorship analyses and provides a richer set of data: linking the data enables research that would otherwise be impossible. For example, NJR data provides detailed data on the medical devices used in the operation as well as specific information about surgical approach, throboprophylaxis and intra-operative complications. For shoulder surgery, NJR also collects a programme of Patient Recorded Outcome Measures (PROMs) which are not otherwise available from NHS Digital. The linked cohort is cleaned and is made ‘research ready’ prior to sub-licensing. This involves re-formatting the data in such a way that the meta-data is available to researchers and the data can easily be used in statistical analysis packages. It should be noted that only subsets of the cohort, relevant to the approved research, are made available to a researcher via a secure Data Access Portal. The cohort is agreed as part of the NJR’s approvals process.

It is anticipated that around 15 applications for sub-licensing would likely be approved each year. Of these, approximately 10 would require NJR data linked to Civil Registration data (date of death or time to death) and approximately 5 would require NJR data linked to HES and/or PROMS data. Sub-licensing, which will make the data more easily accessible, may lead to an increase in the number of applications for data.

The length of the licence is agreed as part of the approvals process and any extension to the agreement must be approved by the NJR’s Research Committee. The length of the sub-licence will depend upon the research being taken but would normally be for 12 months. The NJR has implemented a secure Data Access Portal (DAP) through which sub-licensees must access the data. Access to the data at the end of the licence period can be terminated via the DAP’s central administrative function. Cohorts of data used for research will be archived on the DAP, in line with current guidance, for a period of up to five years. This is necessary should the outcomes of, or the methods employed in, the research be challenged and the analysis needs to be repeated. Access to the archived data would be subject to approval by the Research Committee and DARG.

The approved organisations and researchers, who are granted an access to the linked data via the DAP, agree to terms and conditions of use, their rights and responsibilities as users of the linked data, as defined by the data controllers.

The NJR’s DAP will enable the implementation of some of NHSD’s requirements, especially with regard to the protection and security of the data. Researchers, for example, can only process the data on the DAP and all downloads are placed into a quarantine area prior to being approved for release by an appropriate system administrator. This ensures that the original data remains within the DAP’s security domain.

Sub license and approval groups:

• The NJR Research Committee is responsible to the NJR Steering Committee (NJRSC), and its objectives are:

- To protect NJR data (and any linked dataset) and strengthen its governance through safe effective, and efficient data management.

- To provide a single point of entry and management pathway for all research proposals and activity.

- Maximise access to the data for researchers.

- Align research activity to a framework of priority themes.

- To deliver the NJR’s research strategy through an expertise-based Research Committee.

• Specifically, the Research Committee is responsible for:

- The release of data for research using an impartial and objective protocol.

- Oversight of the use and reporting of data by research groups.

- Upholding the standard and consistency of work carried out using the data.

• All Expressions of Interest (EOI) are formally reviewed by the Research Committee and, following review, an applicant will either be invited to submit a formal application or informed that the NJR will not support the proposed project. Projects fall into two categories:

NJR Supported Project: this is an NJR supported project by a third part applicant. These will typically require aggregate or summary data, or pseudonymised or anonymised patient level data. These projects are external to, but supported by, the NJR. Requests requiring access to patient, surgeon, or unit personal identifiers, or implant batch number will not be granted. NJR will not provide personal identifiers or NJR linked datasets for external projects.

Recent examples include a study of the effect of cement type on the survivorship of cemented total hip replacement by University of Manchester; and examination of surgeon experience-related effects in patient outcomes in elective orthopaedic operations by Sussex University Hospitals NHS Trust.

NJR Partnership Project: this is a project delivered by a third party applicant with NJR involvement and oversight. These will be studies that require sensitive data items, flows of identifiable data, or data linked to external datasets, such as PROMS, HES, or Civil Registration data (requested through NHS Digital). These projects are undertaken in partnership with the NJR and require an identified named collaborator from the NJR Steering Committee or NJR Research Committee. NJR Partnership Projects require Principal Investigator attendance at a Research Committee meeting before an application can be approved.

Recent examples include an examination of the evaluation of complex hip arthroplasty in the United Kingdom by Northumbria Healthcare NHS Foundation Trust in collaboration with NJR; and the study of floor and ceiling effects of the Oxford Shoulder Score by University of Leicester in collaboration with NJR.

• Any projects requiring external approvals (e,g, CAG, NRES) must be channelled through the NJR application process in the first instance. Upon receipt of Research Committee and before data controller approval, the applicant will be required to provide all evidence of external approvals prior to data release.

• The Research Committee will consider the application in terms of its relevance to the NJR’s research priorities: whether the data requested is clinically appropriate to the proposed use; that the project is methodologically sound; whether aggregated or record level data is required; the applying organisation. In some instances, the Principal Investigator (PI) will be required to attend a Research Committee meeting in person. The NJR also stipulates that the PI must be a substantive employee of the requesting organisation and not be on a fixed-term or temporary contract.

• The use of the DAP will ensure the relevant security measures are adhered to for the sub-licensee’s processing and securing of the data. Data sets made for specific research projects can only be accessed and processed on the DAP by approved users. Physical and administrative processes prevent the download of the data and, if necessary, access to the data can be easily prevented by disabling the project area and its associated user accounts. All other due diligence for data requests (such as formal approval groups etc) will still be adhered to.

• Once a research project has been approved by the NJR Research Committee, the application is then submitted to HQIP’s Data Access Request Group (DARG) for review. Once DARG approval has been granted, the Principal Investigator will have an account set up on the DAP and the data will be loaded into a dedicated, private project area.

NJR Research Committee – Composition:

- NJRSC Public Health and Epidemiology Member (Chair).

- NJRSC Patient Representative Member.

- NJRSC Consultant Orthopaedic Surgeon Member (x4).

- Consultant Orthopaedic Surgeon (foot and ankle surgery expert).

- NJR Medical Director (Medical Advisory Committee representative).

- Expert in mechanical engineering.

- Health Economist.

- Professor of Clinical Epidemiology.

- Professor of Epidemiology.

- Consultant Senior Lecturer in Musculo-skeletal medicine

- NJR Director and Deputy Director of Operations.

- Data Quality Manager (data processing contractor representative)

- Principal Consultant (data processing contractor representative)

- HQIP Data Access Request Group - Composition

- Chief Executive HQIP

- Medical Director HQIP

- Data Protection Officer HQIP

- NHS England Representative

- Welsh Government Representative

- NJR Deputy Director of Operations

- NCAPOP Associate Director

The following Information Governance criteria are considered by the NJR Research Committee and the HQIP DARG:

• Could the project achieve its aims using already published data?

• Whether the application requires record-level data or whether it could achieve its aims using aggregated data.

• Are the data fields requested appropriate to the project or could they be reduced?

• The need, in certain circumstances, to suppress small numbers.

• Whether the project can achieve its aims using anonymised or pseudonymised data. Linked data provided by the DAP will not include data that could be used to identify a patient or surgeon.

• Ensuring that the requesting organisation adheres to appropriate processes, data protection, and security as evidenced by accreditation to standards such as, for example, ISO/IEC 27001, NHSP DSP Toolkit, Cyber Essentials. Whilst many IG requirements are met by the DAP implementation and processes, adherence to standards by requesting organisations demonstrates that staff will be aware of the need for IG and data protection and will have received appropriate training.

• The length of time that the data is required. This would normally be twelve months. Any extensions to that must be approved by the Research Committee.

• That any external approvals (CAG, NRES) have been granted. Evidence will be required before data is made available via the DAP and before the application is submitted to HQIP DARG for final approval.

The GDPR legal basis for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of data controller review ahead of approval. For most applications, no confidential data will be made available to applicants as datasets are pseudonymised prior to release (to the applicant or the DAP). In any application where data is not pseudonymised, the Common Law Duty of Confidentiality for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of the data controller’s review prior to approval. It is also required from the applicant to provide evidence of NHS ethics permissions or evidence that ethics is not required.

The primary consideration in the review of any application for data is that the aims of the project are commensurate with the research goals and priorities of the NJR. The current priority themes approved by the NJR Steering Committee are as follows:

- Device and technology assessment.

- Patient outcomes, safety, and risk management.

- Health economics and cost effectiveness.

- Surgeon education and training.

- Patient information and shared decision-making.

- Disease diagnosis and prognosis.

- Disease causation, prevention, and treatment.

- Injury or trauma in the setting of joint replacement.

- Healthy development and ageing in patients with joint disease and replacement.

All these themes are directly related to both the provision of healthcare and the promotion of health through the twin goals of improving patient safety and patient outcomes. An improvement in patient outcomes also leads to the provision of more cost effective services and reduces costs through a reduction in the need for revision surgery. If applications do not address these themes, they are likely to be rejected.

Any organisation with a legitimate use for the data can apply, the only stipulation being that any proposed project supports the goals of the NJR and that the organisation is capable of delivering a successful outcome within the terms and conditions set out be the NJR for the use of the data. The most likely sub-licensees will be:

- Universities in the UK.

- NHS England trusts.

- Research Fellows appointed by professional societies and supervised by universities.

- Orthopaedic device manufactures (for technology assessment only).

Sub-licences will be granted on a cost recovery basis only.

All outputs are based on aggregated data: the NJR specifically forbids the use of record level data for publication, even if that data is anonymised. This means that the privacy of all service users is maintained. The NJR consent form and patient information leaflet explain that NJR data will be linked to other, specified data sets and made available to other organisations without any personal details included.

The territory of use for sublicenses will be limited to England and Wales.

A public register of NJR data releases is available here - http://www.njrcentre.org.uk/njrcentre/Research/Research-Portfolio

This public register includes information about where NHS Digital sourced datasets are shared.

The data controllers will take responsibility for the actions and omissions of all sub licensees and breach of a sub licence should automatically be regarded as breach of the Data Sharing Framework Contract with NHS Digital. In the event of termination or expiry of the Data Sharing Framework Contract between NHS Digital and the data controllers, all sub licences shall automatically terminate.

The outputs from the use of the data will vary each year, dependent upon the research and analysis priorities set by the NJR’s Research Committee and NJR’s Editorial Board and agreed by the NJR Steering Committee.

a) The main vehicle for publication is the NJR’s Annual Report which is published annually in September. In addition to the Annual Report, a Patient’s Guide to the Annual Report is also published and made available in both hard and electronic copy. The Annual Report and the patients guide can be accessed at http://www.njrreports.org.uk.

b) Additional analyses are undertaken throughout the year and these are published either on the NJR website or in relevant, professional medical journals. To date those journals have included The Lancet, the British Medical Journal, and the Journal of Bone and Joint Surgery. The NJR maintains a dedicated research section on its website and details of planned, approved, completed, and published analysis can be found at: http://www.njrcentre.org.uk/njrcentre/Research/ResearchPortfolio/tabid/313/Default.aspx

c) The outcomes of analyses are also disseminated at conferences and meetings of professional societies. These include the British Orthopaedic Association, the British Hip Society, the British Association for Surgery of the Knee, the British Elbow and Shoulder Society, and the British Orthopaedic Foot and Ankle Society. The NJR also holds two regional events per year which are aimed at those hospital staff responsible for collecting the data, providing them with an update on the work of the NJR, including its outputs.

d) Some outputs are also published through the NJR’s secure online reporting services. These services include NJR Clinician Feedback, NJR Management Feedback, and NJR Supplier Feedback which provide information for clinicians, trust/hospital management, and the manufacturers/suppliers of orthopaedic devices respectively.

The NJR is constantly reviewing the ways in which it communicates with patients and the public more generally and are keen to improve its reach. Recent and planned activity include:

• working with NJR Patient Representatives to recruit new members to the NJR Patient Panel to broaden the scope of patient voices on the NJR

• work with relevant patient groups such as Versus Arthritis to ensure that messaging for people who have had or plan to have joint replacement surgery are well targeted

• produce a series of patient guides http://www.njrcentre.org.uk/njrcentre/Reports-Publications-and-Minutes/Public-and-Patient-Guide co-designed with patients

• to press release key developments to ensure appropriate media coverage. See example at https://www.dailymail.co.uk/health/article-7111723/Thinking-getting-new-hip-knee-Try-online-tool-tells-unique-risks.html and https://medicalxpress.com/news/2020-08-implant-choice-important-surgeon-skill.html

• to consult with the PPI group at the Royal National Orthopaedic Hospital, Stanmore about NJR’s data flows

• work with Understanding Patient Data to produce a video about our work https://understandingpatientdata.org.uk/case-study/monitoring-joint-replacement-surgery

• collaborating with HQIP and The Health Foundation as part of the Understanding Health Data Access (UHDA) programme which aims to improve information available publicly about secondary data uses

• make a newsletter available via our website which the public are able to subscribe to. A blog posts from joint replacement patients will also be hosted on the NJR website.

• to remain active on social media platforms (Facebook, Twitter, Linkedin)

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

OUTPUTS FOR SUB-LICENSING:

The principal outputs relating to the sublicensing agreement will be publication in peer reviewed journals and presentation as speciality conferences. The nature of outputs is described by applicants in their application and reviewed by the NJR Research Committee and HQIP DARG. All outputs will be aggregated with small numbers suppressed in accordance with the HES analysis guide.

The exact nature of those outputs will vary depending on the project portfolio, but by way of example a list of the current applications in the pipeline are included below:.

a. The use of the Exeter Short Revision Stem in primary and revision total hip replacement (Royal Devon & Exeter NHS Foundation Trust; NJR:CivReg)

The Exeter Short Revision Stem which is a surgical implant used in hip replacement surgery, (44mm No00 L.125) was released in 2004 and has now been used in both primary and revision surgeries for more than 10 years. This stem is shorter and slimmer than other Exeter stems of the same offset.

Applicant Organisation is undertaking a local review of the survival of the Exeter Short Revision Stem (Exeter stem 44mm No00 L.125) in both primary and revision Total Hip Replacement (THR) in their unit. More importantly however, the performance of the stem in the National Joint Registry in both primary and revision hip arthroplasty needs to be assessed.

Researchers would like to investigate the national survival of prosthesis rates at 2 years, 5 years and 10 years compared with standard Exeter stems of the same offset.

b. Perioperative aspirin and prevention of prosthetic joint infection (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Some artificial joints become infected and need to be removed. This may occur within a short time such as a month after the surgical operation. Such infections may occur even though the skin is thoroughly cleansed, and antibiotics are given prior to the operation. In some cases, infections occur after a much longer period up to many years. When artificial joint infection does occur in the early post surgical period the bacteria that cause such infection typically come from the patient's skin. These bacteria are implanted into the deep part of the surgical wound during the operation to insert the artificial joint. The skin bacteria that most commonly cause these infections are staphylococci. When these and other bacteria come into contact with an artificial joint they form a slime that makes it very difficult for the antibiotics used to treat infection to penetrate and kill them. This slime is called a biofilm. It has been shown in the test tube that aspirin can reduce biofilm formation. For this reason, the applicant organisation believes that it may be possible that aspirin given to patients before their artificial joint surgery may help to prevent infections requiring removal of the joint. The applicant organisation is able to use the records kept in the National Joint registry to measure whether there is an effect of aspirin by comparing the numbers of infections in patients treated with this compared with other medicines to prevent development of clots in the leg veins that may complicate artificial joint surgery. If it seems that aspirin prevents these infections more studies including clinical trials could be done to work out if more patients having artificial joint surgery should be given this commonly used and very safe drug.

c. Distribution and determinants of American Society of Anesthesiologists ( ASA) grade amongst patients undergoing hip arthroplasty: Data from an International Consortium of Joint Registries (University of Oxford; NJR:CivReg:HES)

Total hip replacement is performed to relieve pain and restore function in thousands of patients with hip arthritis each year. Whether or not a total hip replacement has a successful outcome may be affected by a patient’s age, gender and body mass index at the time of surgery. The outcome of surgery may also be affected if a patient has other long-term health disorders, so called ‘comorbidities’, in addition to hip arthritis. For example, previous studies have shown that patients with more co-morbidities are more likely to have complications after total hip replacement, such as infections. The outcomes of total hip replacements in different countries are often compared to provide insights into healthcare in different counties. So that the comparisons are fair, it is necessary to account for differences in the patient's undergoing surgery.

To date, there have been a small number of studies that have investigated how patients undergoing surgery in different countries compare. One previous study found that the age and level of education of patients in Switzerland having total hip replacement was quite different to that of patients having surgery in the USA. However, to date no study has investigated patients differ in terms of co-morbidities. The study propose to investigate how patients undergoing total hip replacements in different countries worldwide differ in terms of how many co-morbidities they have.

Joint replacement registries will be used as the source of information: these are specialised databases that have been created in many countries worldwide and hold information on patients having total hip replacement.

d. The effect of “allowable” unmatched component size on revision rates and time to 1st revision following primary TKR (Nottingham University Hospitals NHS Trust; NJR:CivReg)

Total knee replacements (TKR) contain a femoral and tibial component. Accurate sizing of both components is necessary in order to improve the range of motion of the knee replacement, minimise any soft tissue irritation and avoid bony overhang of the components. It is common practice to choose a tibial component size that matches the femoral component size. However, this does not always guarantee an appropriate level of joint motion and therefore surgeons may choose a size of the tibial component that could be smaller or larger than that of the femoral component which is “allowable” by the manufacturers of knee replacements.

This is not a rare event; a recent Australian study reported that in almost half of the TKRs the two components were not equal in size. This same study found that patients with a femoral component larger than the tibial component were more likely to undergo revision than those with components of equal size or those where the femoral component was smaller than the tibial component. The authors postulated that this might be due to excess stress being placed on the tibial component thus increasing the risk that the TKR would wear out earlier and need revising.

e. Outcomes of revision total hip replacement following ceramic bearing fracture: an avoidable patient safety issue? (University Hospitals of Leicester NHS Trust; NJR:CivReg)

Ceramic is a commonly used material for hip replacement surfaces, but rarely this material may break. When a ceramic bearing fractures this requires further surgery (a revision) to replace the damaged components, of which the surgeon has several materials available to choose from. There have been numerous reported cases of early failures and even death following revision to metal containing bearings after fractures of ceramic components. This is thought to be due to the abrasive effect of hard ceramic particles eroding the softer metal surfaces and subsequently causing metal poisoning. The risk of developing such complications are however unknown, therefore this study aims to quantify the risk of re-revision or early death following such an event.

f. Health Economics Evaluation of Primary Total Hip Arthroplasty by Bearing Type and Fixation Modality (The Royal Orthopaedic Hospital NHS Foundation Trust; NJR:HES:CivReg:PROMs)

The objective of this study is to determine which hip replacement implants have performed best as per the data held by the National Joint Registry. Researcher will use information about each patient undergoing joint revision. The information will include need for revision surgery and Patient Reported Outcome Measures (PROMs).

In addition to this it is necessary to examine the PROMs data for those who have undergone revision surgery (revision is where the joint replacement has been removed & a new joint replacement implanted). This will allow researchers to examine patient characteristics which may have led to the revision surgery as well as seeing how well the joint replacement was performing for that individual patient prior to revision surgery.

Having this full set of data including the Hospital Episode Statistics (HES) data for each patient (which is why traceability is essential for this data) will allow researchers to examine whether patient factors such as comorbidities or previous surgery influence the satisfaction and performance after hip replacement surgery. By grouping data into specific hip replacements (of which there are many available on the UK market) researchers can then see whether there are any outliers either performing particularly well or poorly and whether the performance and satisfaction with these hip replacements may be related to type of implant, comorbidities and similar. This study is a comprehensive analysis of the data available for England, Wales, Northern Ireland and the Isle of Man.

g. Evaluation of complex hip arthroplasty using the NJR dataset (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Hip replacement is one of the most common operations performed in the NHS and one of the most successful procedures in terms of improving quality of life for patients with arthritis. Whilst a hip replacement will last many years for the majority patients, some may run in to problems and require ‘re-do’ (revision) surgery such as when an infection develops, the components become loose or when the bone around the implant breaks. This ‘re-do’ surgery is often more complex than the original surgery and has a higher chance of running into problems including requiring further ‘re-do’ surgery.

The aim of this project it to obtain a perspective of the current delivery of this complex surgery at a national, regional and local levels. The study aim to report on the kinds of operations which are taking place, where they are happening, who is doing them and how many they are doing. This study forms the basis of a wider project aiming to generate data on the best way to structure provision of such surgeries in the future.

h. What is the effect of the femoral head material on revision rate following primary total hip replacement (THR) using the Exeter cemented stem? (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

There are several reasons why primary joint replacements fail. Most orthopaedic studies have looked at the type of fixation (cemented or uncemented) or type of articulations (metal or metal, metal on polyethylene or ceramic). What has not been looked at is the difference in the material between the stem and modular head combinations. There is greater understanding that metal ions can be released from this so called “trunnion wear” head-stem articulation and contribute to failure of joint replacements. This is due to corrosion in this area caused by difference in electrical charge of the materials.

Researchers wish to take advantage of the data available from the National Joint Registry to compare revision rates of different combinations of stem and head in terms of material and explore if this has any effect on the long-term survival (prosthesis) of joint replacements.

i. A comparison of patient reported outcomes and adverse events following elective total shoulder arthroplasty and hemiarthroplasty – an exploratory study (Imperial College London; NJR:CivReg)

Shoulder pain is a common reason people visit their general practitioner and the number of shoulder replacements performed in England, Wales and Northern Ireland has doubled since 2012. A joint replacement is major surgery and it is associated with significant risks. It’s important that patients and surgeons know which type of replacement leads to the best results and the fewest complications. Young patients who are considering a shoulder replacement have a higher chance of further operations, this group need specific information to guide treatment. When people have a joint replacement pseudonymised information about the surgery is stored in the joint registry. In the years following surgery they are asked to complete questionnaires about their symptoms, function and quality of life. The aim is to use this large volume of data to compare the two main types of joint replacement performed for arthritis of the shoulder.

The applicant organisation is requesting data for all shoulder replacements performed since the national joint registry began collection in 2012. This will be linked to the questionnaires given to patients after their procedure and to the NHS database of patient admissions to hospital. The information remains pseudonymised; a patient number is used only to ensure all the information corresponds to the same patient. Researchers will compare the results of patients who have received a half shoulder replacement and total shoulder replacement. The work will influence the decision making of patients and medical staff. The proposed studies will provide information about the expected recovery and the risks and benefits of a shoulder replacement. The work aims to improve patients’ shoulder function after surgery and reduce the number of additional procedures, complications and hospital admissions.

j. Effect of COVID-19 on NJR Surgical Practice and Outcomes (University of Sheffield; NJR:HES:PROMs:CivReg)

COVID 19 has had a great impact on all of our lives and changed society in ways that are yet to be identified. In this work, researchers will look at the outcomes of joint replacement surgery before COVID and then after COVID. This will allow researchers to see how outcomes of joint replacement surgery differ before and after the start of the COVID pandemic by looking at pain and activity, death and repeat operation.

To do this, researchers will compare these measures over 2 years immediately before COVID and compare these to the 2 years after restart of routine operating. In making these analyses researchers will use a range of mathematical tools to best fit the information the study receive from the NJR. One of the approaches researchers will use is called “machine learning” in which the computer “learns” to predict the outcome based on the information put in.

The other methods researchers will use are more traditional statistical approaches. As part of the study researchers will look to see which method makes the best predictions. Researchers imagine that the extra waiting time might affect these outcomes, as might surgeon deskilling.

However, researchers cannot measure these things directly in the NJR, nor see if any patients had COVID around the time of their operation. Findings will be compared with 2 other registers that do record COVID positive tests. The work is exploratory, to see if there are differences in outcomes before and after COVID, and if these are found researchers will create “hypotheses” to explain the possible reason for the effects and test them in further patients

k. Periprosthetic fractures around primary total hip replacement (University of Leeds; NJR:CivReg)

Total Hip replacements (THR) are an effective treatment for managing end-stage hip arthritis. The implants used come in many different shapes and sizes, and work in different ways. The implants are designed to last as long as possible, but a small amount fail because of loosening, infection, instability or breakage of the thigh bone (PFF) amongst other causes.

In patients older than 70, PFF is the second most common cause of failure of THR.

PFF leads to significant problems for patients (more surgery, higher risk of medical complications including death) and their incidence is increasing. Unfortunately, PFF is difficult to prevent and one of the strongest predictors is the type of implants surgeons use. The NJR records when patients with PFF have their implants exchanged (revision).

Previously, researchers identified a number of implant designs which make PFF more likely. This is already making an impact to reduce the risk of PFF. Unfortunately, previous data does not fully explain what happens to patients because patients can also be treated with no surgery and fixation surgery, which are not recorded in the NJR. In order to get a full picture of how the implants affect the risk of PFF researchers need to get information from more sources, which includes implant data (from the NJR), hospital data (which tells us about PFF not recorded in the NJR) and death statistics. This will make sure researchers can work out the complete picture of how the implants used affect the risk of PFF and ultimately how practice can be changed to prevent PFF from harming patients.

This will enable researchers to look at the most complete picture of PFF; how and when it occurs, how it affects patients and identify things which w can be changed to make PFF less likely in future. This will be the largest study looking at PFF. Although this research does not seek to make a fail-safe conclusion about which implant is safest; it will help the surgeons, patients and other stakeholders make informed decision in choosing the right implant for the right patient.

The following benefits to date include, but are not limited to:

• Providing outcomes data to the Medicines and Healthcare products Regulatory Agency (MHRA) that has led to the withdrawal of a number of orthopaedic devices from the market.

• Quickly identifying to hospitals those patients affected by a Field Notice or Device Alert issued by the MHRA.

• Direct influences on changes to clinical practice include:

o Preventing the use of hip resurfacing procedures in women over the age of fifty (reported through the NJRs Annual Report).

o Getting it Right First Time (GIRFT) proscribing the use of cemented hip implants in patients over 70 years of age (reported in the NJR’s Annual Report and supplemented by further analysis).

o Undertaking analysis that reduced to almost zero, the amount of large head, metal-on-metal hip replacements being undertaken.

o Undertaking analysis that resulted in a significant reduction in the number of hip replacement using a metal-on-metal articulating bearing in hip replacement.

o Professional societies, including the British Orthopaedic Association (BOA), issuing guidance to members about the minimum number of procedure types of procedure to be undertaken in a year in order to remain ‘current’. This follows an analysis of outcomes based on the number of procedures carried out by surgeons.

o A realignment of services for shoulder, ankle, and elbow joint replacement to a ‘hub and spoke’ service delivery on advice from GIRFT, based on an analysis of outcomes and the effects of low volume surgeons.

• The provision of enhanced data and reporting services to all stakeholders enabling them to make decisions about the care provided to patients. Stakeholders include clinicians, device manufacturers, service providers and commissioners, patients, regulators, and advisory bodies such as GIRFT and National Institute for Health and Care Excellence (NICE).

• Reducing the cost of orthopaedic devices to many NHS trusts by benchmarking outcomes data against the price paid for implants by those Trusts.

• Using risk-adjusted outcomes analyses to identify suspected outlier performance in surgeons, hospitals, and Trusts. This has enabled the appropriate responsible organisations and individuals to take the action necessary to rectify individual or system issues affecting outcomes. Such resolution has included the re-training of surgeons in certain procedure types and surgeons agreeing not to undertake specific procedure types. The NJR, in conjunction with the BOA, provides support to hospitals requesting it.

• By publishing and sharing data on outcomes, the NJR has informed changes in clinical practice that have reduced the rate of revision surgery. A reduction in the rate of revision surgery evidences improvements in patient outcomes and safety and also reduces the cost to the NHS of such surgery.

The National Joint Registry (NJR) is commissioned by the Healthcare Quality Improvement Partnership (HQIP) on behalf of NHS England (NHSE) as part of the Clinical Audit and Patient Outcomes Programme (NCAPOP). The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety. A longstanding agreement has been in place between NHS Digital, HQIP and NHSE, under version 8 of this agreement, approval is being sought on HQIP and NHSE being able to sub-license NHS Digital data.

This agreement has Joint Data Controllership between HQIP and NHSE.

The NCAPOP is a large programme of circa 35 projects consisting of National Clinical Audits. HQIP is commissioned by NHSE to commission and manage the NCAPOP. NHSE is a controller of the NCAPOP jointly with HQIP as together both organisations determine the purposes and means of processing. NHSE is responsible for determining which projects/topics are included as part of the NCAPOP. HQIP, as commissioner of the NCAPOP, is responsible for project specification development, procurement and extension activities, contract management and authorising data sharing requests. NHS England, as a funder of the NCAPOP, participates within specification development, procurement and project extension activities and authorises the publication of project outputs.

NHSE is involved with developing the scope and purpose of the NCAPOP projects through participation within specification development activities and may authorise (as chair of the specification development meetings) the final project specifications. These specifications set out the purpose of the project, the patient groups and clinical services to evaluate and the types of data to collect. NHSE are a representative upon the HQIP Data access request group which authorises data sharing applications from third parties.

LEGAL BASIS JUSTIFICATION:

HQIP and NHSE both rely on the Article 6 (1) (e) legal basis under GDPR - "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". This is justified through commissioning arrangements which link back to NHS England and other national bodies with statutory responsibilities to improve quality of health care services.

HQIP rely on Article 9 (2) (i) as the legal basis for processing under GDPR - "processing is necessary for reasons of public interest in the area of public health, such as protecting against serious cross-border threats to health or ensuring high standards of quality and safety of health care and of medicinal products or medical devices, on the basis of Union or Member State law which provides for suitable and specific measures to safeguard the rights and freedoms of the data subject, in particular professional secrecy". This is justified as all projects aim to drive improvements in the quality and safety of care and to improve outcomes for patients.

NHSE rely on Article 9(2)(h) of the GDPR as the legal basis for processing. "Processing is necessary for the purposes of preventive or occupational medicine, for the assessment of the working capacity of the employee, medical diagnosis, the provision of health or social care or treatment or the management of health or social care systems and services on the basis of Union or Member State law or pursuant to contract with a health professional and subject to the conditions and safeguards referred to in paragraph 3". NHS England are responsible for provision of health and social care, and management of systems and compliance.

The NJR was established in 2003 by the Department of Health following a National Audit Office (NAO) report into the higher than expected failure rate of the 3M hip replacement device. The NAO report concluded that, had a national register of hip replacements been in existence, the failure rate would have been detected earlier. Earlier identification would have meant less patients were affected and the costs of revision surgery for the NHS would have been considerably less. The NJR went live in April 2003. Since its establishment, the responsibility for delivering the NJR has passed to HQIP who deliver the National Joint Registry.

HQIP and NHSE are joint data controllers for the NJR and contract to two other organisations for the delivery of the programme, both of which are data processors:

• Northgate Public Services (NPS; a software and outsourcing business) is responsible for the NJR’s data collection and processing activities, including data storage and the provision of stakeholder reporting services.

• The University of Bristol (UoB)is also a data processor with responsibilities for statistical analysis and reporting.

HQIP and NHSE, have no access to the record level data. Northgate has access to record level data, including patient identifiers, in order to link the data provided to patient records held by the NJR. The linked dataset is pseudonymised by NPS before it is made available to the statistical analysis team at the UoB. Although the UoB team has access to record level data, it does not have access to patient identifiers. The work undertaken by the NJR is to monitor the outcomes of hip, knee, shoulder, ankle, and elbow joint replacement surgery with regards to the performance of devices, surgical teams, and Trusts and hospitals. The monitoring is necessary to ensure patient safety and improve patient outcomes.

The overall purpose of the NJR is summarised in its mission statement:

‘The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety. In a continuous drive to improve the quality of outcomes and ensure the quality and cost effectiveness of joint replacement surgery, the NJR will monitor and report on outcomes, and support and enable related research.’

The strategic goals of the NJR are as follows:

• To monitor the outcomes achieved by brand of prosthesis, hospital and surgeon, and highlight where these fall below an expected performance in order to allow prompt investigation and to support follow-up action.

• To inform patients, clinicians, providers and commissioners of healthcare, regulators and implant suppliers of the outcomes achieved in joint replacement surgery.

• To evidence variations in outcome achieved across surgical practice in order to inform best practice.

• To enhance patient awareness of joint replacement outcomes to better inform patient choice and patients’ quality of experience through engagement with patients and patient organisations.

• To support evidence-based purchasing of joint replacement implants for healthcare providers to support quality and cost effectiveness.

• To support suppliers in the routine post-market surveillance of implants and provide information to clinicians, patients, hospital management and the regulatory authorities.

The NJR has been collecting data since 2003 and is an ongoing audit with no planned end date. The NJR’s Steering Committee reviews the NJR’s mission statement and strategic goals annually to ensure that they remain relevant to existing stakeholder priorities, clinical guidance, and legislation.

The NJR requires record level data from Hospital Episode Statistics (HES), Patient Reported Outcome Measures (PROMS) and Civil Registration (Deaths) data from NHS Digital to help achieve some of NJRs strategic goals. The NJR creates annual, linked dataset comprising data from the NJR, HES, Patient Episode Database Wales (PEDW), PROMs (NHS England programme) and Civil Registration. This dataset is used primarily for the production of the NJR’s Annual Report, its associated outcomes analyses, and in some reporting services provided to stakeholders. The data is also used for specific research projects undertaken by the NJR. By linking to HES, PROMs and Civil Registration data, the NJR is able to enhance the quality and type of analyses that it undertakes. The priorities for NJR analyses and research are set by the NJR’s Research Committee and Editorial Board, both of which report to the NJR Steering Committee.

Research committee members (including lay members) will review all application statements on benefits to health and care. In all cases where the applicant and/or the study funding are commercial in nature, careful consideration will be given to weigh the expected benefits against any commercial gain to the commercial organisation.

The data sets requested by the NJR consist of HES Patient Admitted Care, Civil Registration (Deaths), and Patient Reported Outcomes Measures. Patient identifiers are required in order to link the data requested from NHSD to patient identifiable data held by the NJR. This linkage is necessary to undertake analysis at a record level when, for example, considering factors such as length of stay for different joint replacement types, PROMs associated with different devices/implants, or outcomes associated with deprivation.

Civil Registration Data (Deaths) is essential for any outcomes analyses: patients who have died with an implant intact, for example, have to be excluded from any cohort. Outcomes analyses also includes a consideration of time to death from operation, and the cause of death.

HES Patient Admitted Care data is required to look at potential factors affecting the outcomes of joint replacement surgery, such as length of stay, re-admission to theatre, associated diagnoses and co-morbidities, deprivation, geography and location.

The NJR currently uses two endpoints in outcomes analysis: death and revision. PROMs represents an invaluable, third endpoint, providing a patients view, as opposed to a clinical view, of the outcome of joint replacement surgery. The NHS England programme is the only source of PROMs data relating to hip and knee joint replacement surgery.

The NJR requires data from 2003 which is when it started its own data collection. The number of years being requested is justified on the need to undertake longitudinal analyses of the available data. Currently, recommended guidance on the expected survivorship of joint replacement (as published by NICE) is based on ten years of data. Long term outcomes analyses are required.

Patient identifiable data is required to link the record level HES/Civil Registration/PROMs data to the relevant patient procedure in the NJR. The primary data field used for linkage is the patient's NHS number, but other fields such as date of birth, local patient ID, postcode, and side of procedure are used to ensure a reliable linkage. Data sets used for analyses of the linked data set have patient identifiers removed.

In some circumstances, it is not possible for the NJR to determine whether a patient has consented or not (‘unknown’ cohort). This might be because their operation was the result of an emergency admission, or because the admitting hospital has been unable to ask the patient to consent for some other reason. For this cohort, the NJR has support under S251 to collect and process data. Research into the ‘unknown’ cohort comes under purpose 2 of the s251 support; and that there is sufficient oversight within the NJR internal assurance procedures to ensure that any research using the ‘unknown’ cohort’s data is within the parameters of purpose 2 of the s251 support.

The sub-cohort received from NHS Digital where the patient has declined consent do not get linked with NJR and are not shared with applicants under sublicensing, however, the sub-cohort received from NHSD where consent is unknown and there is S251 support for are linked with NJR and do form part of the dataset that are shared with applicants.

In order to achieve minimisation, the NJR is requesting data for those OPCS4 codes relating to hip, knee, ankle, elbow, and shoulder joint replacement only. This will also ensure that Civil Registration data relating to those procedures only will be provided. By requesting data related to specific OPCS4 codes will ensure that the data requested is appropriate to the NJR's needs. Minimisation will be achieved by focusing on specific procedure codes.

• Linkage to HES data enables the NJR to improve the type and quality of the analyses that it can undertake without having to collect the data itself. Such linkage may be length of stay linked to outcomes or co-morbidities linked to outcomes. The NJR has, for example, published a paper examining the increased risk of cancer to patients who have had hip replacement procedures involving the use of metal on metal bearing surfaces, i.e. the femoral head and the acetabular cup or liner are both made of metal.

• By linking to PROMs outcomes analyses will improve as poor performance will be detected even if the primary joint replacement procedure has not been revised. A patient’s view of the outcome of joint replacement is a key indicator to the reporting of outcomes and provides an additional endpoint in addition to revision and death. Linkage to PROMs, and analysis at the record level, enables the NJR to monitor the performance of hospitals, surgeons, and individual implants.

• The NJR is required to publish 90 day mortality rates for all those NHS England Trusts undertaking joint replacement surgery in its Annual Clinical Reports to Trusts, the NJR’s Annual Report and as part of the former NHS England Clinical Outcomes Programme (although COP does not exist in its original form, the NJR still publishes annual indicators. The data, originally shared with Choices, is now shared with the CQC). Date of death and any associated co-morbidities are essential for outcomes analyses and for risk-adjusting mortality analyses.

This will enable the NJR to continue to improve its analysis and data quality audits, activities which are expressly intended to improve patient outcomes and patient safety.

The NJR has developed a secure, online Data Access Portal which provides additional security to the data used by researchers and analysts outside of the NJR. The plan is to provide subsets of linked data to approved research projects such that it is processed on the Portal and cannot be removed from the NJR’s security domain. The data does not contain patient identifiable fields. In February 2019, the NJR submitted a Research Application via IRAS. Once approved, the NJR will enter into a sub-licensing agreement with NHS Digital such that linked datasets can be provided for approved projects. This will ensure that more use can be made of the wealth of data held by the NJR.

Data from Wales, Northern Ireland, the Isle of Man, and Guernsey are sourced separately.

SUB-LICENSING (this is the purpose of v8 of this agreement):

Researchers currently submit research applications to the NJR that require a linked cohort of HES, NJR, and Civil Registry data which requires the submission of two separate data applications to each data controller: NHS Digital and HQIP. Both data controllers must then undertake more processing to create the cohort required which includes the provision of patient identifiers for linkage and then adding the source data to those linked identifiers. The NJR would then supply patient identifiers to NHS Digital for the proposed cohort, the HES records and linked unique NJR ID would then be returned to the applicant, who must then supply the unique IDs to the NJR so that the NJR data can be supplied to the applicant and subsequently linked. This is time-consuming for controllers/processors and has deterred individuals and organisations from using the data to undertake much needed research and analyses. It also entails the transmission of patient identifiable data for each project.

The NJR already creates an annual linked dataset comprising of HES, Civil Registry, PROMs, and NJR data. Sub-licensing datasets extracted from this linked data set will save both NHS Digital and the NJR considerable effort, shorten timescales for applicants, make the data easily accessible for applicants and encourage greater use of the data for research. It will also reduce the processing and transmission of patient identifiable data. Such fully linked dataset is something NHS Digital do not routinely provide.

The NJR currently uses two endpoints for undertaking survivorship and outcomes analyses: revision and death. Whilst the NJR captures revision, the use of Civil Registry is essential to determine mortality status, data of death, age at death, and time to death following a joint replacement. Information about the cause of death is used to further refine the outcomes analysis.

The addition of PROMs data provides a third endpoint for the analysis of outcomes for knee and hip joint replacement and being able to measure health gain and improvements in quality of life can be used to support analyses into the outcomes associated with, for example, particular types of procedure or types of implant.

The NJR will not sub-licence HES/PROMs/Civil Registry data in its raw form to applicants. The data from NHS Digital will be linked to NJR data and used for approved research and analyses relating to joint replacement surgery and which support the NJR’s research priorities. The addition of HES/PROMS/Civil Registry data to data held by the NJR is necessary for all outcomes and survivorship analyses and provides a richer set of data: linking the data enables research that would otherwise be impossible. For example, NJR data provides detailed data on the medical devices used in the operation as well as specific information about surgical approach, throboprophylaxis and intra-operative complications. For shoulder surgery, NJR also collects a programme of Patient Recorded Outcome Measures (PROMs) which are not otherwise available from NHS Digital. The linked cohort is cleaned and is made ‘research ready’ prior to sub-licensing. This involves re-formatting the data in such a way that the meta-data is available to researchers and the data can easily be used in statistical analysis packages. It should be noted that only subsets of the cohort, relevant to the approved research, are made available to a researcher via a secure Data Access Portal. The cohort is agreed as part of the NJR’s approvals process.

It is anticipated that around 15 applications for sub-licensing would likely be approved each year. Of these, approximately 10 would require NJR data linked to Civil Registration data (date of death or time to death) and approximately 5 would require NJR data linked to HES and/or PROMS data. Sub-licensing, which will make the data more easily accessible, may lead to an increase in the number of applications for data.

The length of the licence is agreed as part of the approvals process and any extension to the agreement must be approved by the NJR’s Research Committee. The length of the sub-licence will depend upon the research being taken but would normally be for 12 months. The NJR has implemented a secure Data Access Portal (DAP) through which sub-licensees must access the data. Access to the data at the end of the licence period can be terminated via the DAP’s central administrative function. Cohorts of data used for research will be archived on the DAP, in line with current guidance, for a period of up to five years. This is necessary should the outcomes of, or the methods employed in, the research be challenged and the analysis needs to be repeated. Access to the archived data would be subject to approval by the Research Committee and DARG.

The approved organisations and researchers, who are granted an access to the linked data via the DAP, agree to terms and conditions of use, their rights and responsibilities as users of the linked data, as defined by the data controllers.

The NJR’s DAP will enable the implementation of some of NHSD’s requirements, especially with regard to the protection and security of the data. Researchers, for example, can only process the data on the DAP and all downloads are placed into a quarantine area prior to being approved for release by an appropriate system administrator. This ensures that the original data remains within the DAP’s security domain.

Sub license and approval groups:

• The NJR Research Committee is responsible to the NJR Steering Committee (NJRSC), and its objectives are:

- To protect NJR data (and any linked dataset) and strengthen its governance through safe effective, and efficient data management.

- To provide a single point of entry and management pathway for all research proposals and activity.

- Maximise access to the data for researchers.

- Align research activity to a framework of priority themes.

- To deliver the NJR’s research strategy through an expertise-based Research Committee.

• Specifically, the Research Committee is responsible for:

- The release of data for research using an impartial and objective protocol.

- Oversight of the use and reporting of data by research groups.

- Upholding the standard and consistency of work carried out using the data.

• All Expressions of Interest (EOI) are formally reviewed by the Research Committee and, following review, an applicant will either be invited to submit a formal application or informed that the NJR will not support the proposed project. Projects fall into two categories:

NJR Supported Project: this is an NJR supported project by a third part applicant. These will typically require aggregate or summary data, or pseudonymised or anonymised patient level data. These projects are external to, but supported by, the NJR. Requests requiring access to patient, surgeon, or unit personal identifiers, or implant batch number will not be granted. NJR will not provide personal identifiers or NJR linked datasets for external projects.

Recent examples include a study of the effect of cement type on the survivorship of cemented total hip replacement by University of Manchester; and examination of surgeon experience-related effects in patient outcomes in elective orthopaedic operations by Sussex University Hospitals NHS Trust.

NJR Partnership Project: this is a project delivered by a third party applicant with NJR involvement and oversight. These will be studies that require sensitive data items, flows of identifiable data, or data linked to external datasets, such as PROMS, HES, or Civil Registration data (requested through NHS Digital). These projects are undertaken in partnership with the NJR and require an identified named collaborator from the NJR Steering Committee or NJR Research Committee. NJR Partnership Projects require Principal Investigator attendance at a Research Committee meeting before an application can be approved.

Recent examples include an examination of the evaluation of complex hip arthroplasty in the United Kingdom by Northumbria Healthcare NHS Foundation Trust in collaboration with NJR; and the study of floor and ceiling effects of the Oxford Shoulder Score by University of Leicester in collaboration with NJR.

• Any projects requiring external approvals (e,g, CAG, NRES) must be channelled through the NJR application process in the first instance. Upon receipt of Research Committee and before data controller approval, the applicant will be required to provide all evidence of external approvals prior to data release.

• The Research Committee will consider the application in terms of its relevance to the NJR’s research priorities: whether the data requested is clinically appropriate to the proposed use; that the project is methodologically sound; whether aggregated or record level data is required; the applying organisation. In some instances, the Principal Investigator (PI) will be required to attend a Research Committee meeting in person. The NJR also stipulates that the PI must be a substantive employee of the requesting organisation and not be on a fixed-term or temporary contract.

• The use of the DAP will ensure the relevant security measures are adhered to for the sub-licensee’s processing and securing of the data. Data sets made for specific research projects can only be accessed and processed on the DAP by approved users. Physical and administrative processes prevent the download of the data and, if necessary, access to the data can be easily prevented by disabling the project area and its associated user accounts. All other due diligence for data requests (such as formal approval groups etc) will still be adhered to.

• Once a research project has been approved by the NJR Research Committee, the application is then submitted to HQIP’s Data Access Request Group (DARG) for review. Once DARG approval has been granted, the Principal Investigator will have an account set up on the DAP and the data will be loaded into a dedicated, private project area.

NJR Research Committee – Composition:

- NJRSC Public Health and Epidemiology Member (Chair).

- NJRSC Patient Representative Member.

- NJRSC Consultant Orthopaedic Surgeon Member (x4).

- Consultant Orthopaedic Surgeon (foot and ankle surgery expert).

- NJR Medical Director (Medical Advisory Committee representative).

- Expert in mechanical engineering.

- Health Economist.

- Professor of Clinical Epidemiology.

- Professor of Epidemiology.

- Consultant Senior Lecturer in Musculo-skeletal medicine

- NJR Director and Deputy Director of Operations.

- Data Quality Manager (data processing contractor representative)

- Principal Consultant (data processing contractor representative)

- HQIP Data Access Request Group - Composition

- Chief Executive HQIP

- Medical Director HQIP

- Data Protection Officer HQIP

- NHS England Representative

- Welsh Government Representative

- NJR Deputy Director of Operations

- NCAPOP Associate Director

The following Information Governance criteria are considered by the NJR Research Committee and the HQIP DARG:

• Could the project achieve its aims using already published data?

• Whether the application requires record-level data or whether it could achieve its aims using aggregated data.

• Are the data fields requested appropriate to the project or could they be reduced?

• The need, in certain circumstances, to suppress small numbers.

• Whether the project can achieve its aims using anonymised or pseudonymised data. Linked data provided by the DAP will not include data that could be used to identify a patient or surgeon.

• Ensuring that the requesting organisation adheres to appropriate processes, data protection, and security as evidenced by accreditation to standards such as, for example, ISO/IEC 27001, NHSP DSP Toolkit, Cyber Essentials. Whilst many IG requirements are met by the DAP implementation and processes, adherence to standards by requesting organisations demonstrates that staff will be aware of the need for IG and data protection and will have received appropriate training.

• The length of time that the data is required. This would normally be twelve months. Any extensions to that must be approved by the Research Committee.

• That any external approvals (CAG, NRES) have been granted. Evidence will be required before data is made available via the DAP and before the application is submitted to HQIP DARG for final approval.

The GDPR legal basis for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of data controller review ahead of approval. For most applications, no confidential data will be made available to applicants as datasets are pseudonymised prior to release (to the applicant or the DAP). In any application where data is not pseudonymised, the Common Law Duty of Confidentiality for each application will be provided by the applicant as part of their submission. This will be examined by HQIP DARG as part of the data controller’s review prior to approval. It is also required from the applicant to provide evidence of NHS ethics permissions or evidence that ethics is not required.

The primary consideration in the review of any application for data is that the aims of the project are commensurate with the research goals and priorities of the NJR. The current priority themes approved by the NJR Steering Committee are as follows:

- Device and technology assessment.

- Patient outcomes, safety, and risk management.

- Health economics and cost effectiveness.

- Surgeon education and training.

- Patient information and shared decision-making.

- Disease diagnosis and prognosis.

- Disease causation, prevention, and treatment.

- Injury or trauma in the setting of joint replacement.

- Healthy development and ageing in patients with joint disease and replacement.

All these themes are directly related to both the provision of healthcare and the promotion of health through the twin goals of improving patient safety and patient outcomes. An improvement in patient outcomes also leads to the provision of more cost effective services and reduces costs through a reduction in the need for revision surgery. If applications do not address these themes, they are likely to be rejected.

Any organisation with a legitimate use for the data can apply, the only stipulation being that any proposed project supports the goals of the NJR and that the organisation is capable of delivering a successful outcome within the terms and conditions set out be the NJR for the use of the data. The most likely sub-licensees will be:

- Universities in the UK.

- NHS England trusts.

- Research Fellows appointed by professional societies and supervised by universities.

- Orthopaedic device manufactures (for technology assessment only).

Sub-licences will be granted on a cost recovery basis only.

All outputs are based on aggregated data: the NJR specifically forbids the use of record level data for publication, even if that data is anonymised. This means that the privacy of all service users is maintained. The NJR consent form and patient information leaflet explain that NJR data will be linked to other, specified data sets and made available to other organisations without any personal details included.

The territory of use for sublicenses will be limited to England and Wales.

A public register of NJR data releases is available here - http://www.njrcentre.org.uk/njrcentre/Research/Research-Portfolio

This public register is populated to include information about where NHS Digital sourced datasets are shared.

The data controllers will take responsibility for the actions and omissions of all sub licensees and breach of a sub licence should automatically be regarded as breach of the Data Sharing Framework Contract with NHS Digital. In the event of termination or expiry of the Data Sharing Framework Contract between NHS Digital and the data controllers, all sub licences shall automatically terminate.

The outputs from the use of the data will vary each year, dependent upon the research and analysis priorities set by the NJR’s Research Committee and NJR’s Editorial Board and agreed by the NJR Steering Committee.

a) The main vehicle for publication is the NJR’s Annual Report which is published annually in September. In addition to the Annual Report, a Patient’s Guide to the Annual Report is also published and made available in both hard and electronic copy. The Annual Report and the patients guide can be accessed at http://www.njrreports.org.uk.

b) Additional analyses are undertaken throughout the year and these are published either on the NJR website or in relevant, professional medical journals. To date those journals have included The Lancet, the British Medical Journal, and the Journal of Bone and Joint Surgery. The NJR maintains a dedicated research section on its website and details of planned, approved, completed, and published analysis can be found at: http://www.njrcentre.org.uk/njrcentre/Research/ResearchPortfolio/tabid/313/Default.aspx.

c) The outcomes of analyses are also disseminated at conferences and meetings of professional societies. These include the British Orthopaedic Association, the British Hip Society, the British Association for Surgery of the Knee, the British Elbow and Shoulder Society, and the British Orthopaedic Foot and Ankle Society. The NJR also holds two regional events per year which are aimed at those hospital staff responsible for collecting the data, providing them with an update on the work of the NJR, including its outputs.

d) Some outputs are also published through the NJR’s secure online reporting services. These services include NJR Clinician Feedback, NJR Management Feedback, and NJR Supplier Feedback which provide information for clinicians, trust/hospital management, and the manufacturers/suppliers of orthopaedic devices respectively.

The NJR is constantly reviewing the ways in which it communicates with patients and the public more generally and are keen to improve its reach. Recent and planned activity include:

• working with NJR Patient Representatives to recruit new members to the NJR Patient Panel to broaden the scope of patient voices on the NJR

• work with relevant patient groups such as Versus Arthritis to ensure that messaging for people who have had or plan to have joint replacement surgery are well targeted

• produce a series of patient guides http://www.njrcentre.org.uk/njrcentre/Reports-Publications-and-Minutes/Public-and-Patient-Guide co-designed with patients

• to press release key developments to ensure appropriate media coverage. See example at https://www.dailymail.co.uk/health/article-7111723/Thinking-getting-new-hip-knee-Try-online-tool-tells-unique-risks.html and https://medicalxpress.com/news/2020-08-implant-choice-important-surgeon-skill.html

• to consult with the PPI group at the Royal National Orthopaedic Hospital, Stanmore about NJR’s data flows

• work with Understanding Patient Data to produce a video about our work https://understandingpatientdata.org.uk/case-study/monitoring-joint-replacement-surgery

• collaborating with HQIP and The Health Foundation as part of the Understanding Health Data Access (UHDA) programme which aims to improve information available publically about secondary data uses

• make a newsletter available via our website which the public are able to subscribe to. A blog posts from joint replacement patients will also be hosted on the NJR website.

• to remain active on social media platforms (Facebook, Twitter, Linkedin)

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

OUTPUTS FOR SUB-LICENSING:

The principal outputs relating to the sublicensing agreement will be publication in peer reviewed journals and presentation as speciality conferences. The nature of outputs is described by applicants in their application and reviewed by the NJR Research Committee and HQIP DARG. All outputs will be aggregated with small numbers suppressed in accordance with the HES anaylsis guide.

The exact nature of those outputs will vary depending on the project portfolio, but by way of example a list of the current applications in the pipeline are included below:.

a. The use of the Exeter Short Revision Stem in primary and revision total hip replacement (Royal Devon & Exeter NHS Foundation Trust; NJR:CivReg)

The Exeter Short Revision Stem which is a surgical implant used in hip replacement surgery, (44mm No00 L.125) was released in 2004 and has now been used in both primary and revision surgeries for more than 10 years. This stem is shorter and slimmer than other Exeter stems of the same offset.

Applicant Organisation is undertaking a local review of the survival of the Exeter Short Revision Stem (Exeter stem 44mm No00 L.125) in both primary and revision Total Hip Replacement (THR) in their unit. More importantly however, the performance of the stem in the National Joint Registry in both primary and revision hip arthroplasty needs to be assessed.

Researchers would like to investigate the national survival of prosthesis rates at 2 years, 5 years and 10 years compared with standard Exeter stems of the same offset.

b. Perioperative aspirin and prevention of prosthetic joint infection (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Some artificial joints become infected and need to be removed. This may occur within a short time such as a month after the surgical operation. Such infections may occur even though the skin is thoroughly cleansed and antibiotics are given prior to the operation. In some cases infections occur after a much longer period up to many years. When artificial joint infection does occur in the early post surgical period the bacteria that cause such infection typically come from the patients skin. These bacteria are implanted into the deep part of the surgical wound during the operation to insert the artificial joint. The skin bacteria that most commonly cause these infections are staphylococci. When these and other bacteria come into contact with an artificial joint they form a slime that makes it very difficult for the antibiotics used to treat infection to penetrate and kill them. This slime is called a biofilm. It has been shown in the test tube that aspirin can reduce biofilm formation. For this reason the applicant organisation believes that it may be possible that aspirin given to patients before their artificial joint surgery may help to prevent infections requiring removal of the joint. The applicant organisation is able to use the records kept in the National Joint registry to measure whether there is an effect of aspirin by comparing the numbers of infections in patients treated with this compared with other medicines to prevent development of clots in the leg veins that may complicate artificial joint surgery. If it seems that aspirin prevents these infections more studies including clinical trials could be done to work out if more patients having artificial joint surgery should be given this commonly used and very safe drug.

c. Distribution and determinants of American Society of Anesthesiologists ( ASA) grade amongst patients undergoing hip arthroplasty: Data from an International Consortium of Joint Registries (University of Oxford; NJR:CivReg:HES)

Total hip replacement is performed to relieve pain and restore function in thousands of patients with hip arthritis each year. Whether or not a total hip replacement has a successful outcome may be affected by a patient’s age, gender and body mass index at the time of surgery. The outcome of surgery may also be affected if a patient has other long-term health disorders, so called ‘comorbidities’, in addition to hip arthritis. For example, previous studies have shown that patients with more co-morbidities are more likely to have complications after total hip

replacement, such as infections. The outcomes of total hip replacements in different countries are often compared to provide insights into healthcare in different counties. So that the comparisons are fair, it is

necessary to account for differences in the patients undergoing surgery. To date, there have been a small number of studies that have investigated how patients undergoing surgery in different countries compare. One previous study found that the age and level of education of patients in Switzerland having total hip replacement was quite

different to that of patients having surgery in the USA. However, to date no study has investigated patients differ in terms of co-morbidities. In this study, we propose to investigate how patients undergoing total hip replacements in different countries worldwide differ in terms of how many co-morbidities they have. Joint

replacement registries will be used as the source of information: these are specialised databases that have been created in many countries worldwide and hold information on patients having total hip replacement.

d. The effect of “allowable” unmatched component size on revision rates and time to 1st revision following primary TKR (Nottingham University Hospitals NHS Trust; NJR:CivReg)

Total knee replacements (TKR) contain a femoral and tibial component. Accurate sizing of both components is necessary in order to improve the range of motion of the knee replacement, minimise any soft tissue irritation and avoid bony overhang of the components. It is common practice to choose a tibial component size that matches the femoral component size. However, this does not always guarantee an appropriate level of joint motion and therefore surgeons may choose a size of the tibial component that could be smaller or larger than that of the femoral component which is “allowable” by the manufacturers of knee replacements. This is not a rare event; a recent Australian study reported that in almost half of the TKRs the two components were not equal in size. This same study found that patients with a femoral component larger than the tibial component were more likely to undergo revision than those with components of equal size or those where the femoral component was smaller than the tibial component. The authors postulated that this might be due to excess stress being placed on the tibial component thus increasing the risk that the TKR would wear out earlier and need revising.

e. Outcomes of revision total hip replacement following ceramic bearing fracture: an avoidable patient safety issue? (University Hospitals of Leicester NHS Trust; NJR:CivReg)

Ceramic is a commonly used material for hip replacement surfaces, but rarely this material may break. When a ceramic bearing fractures this requires further surgery (a revision) to replace the damaged components, of which the surgeon has several materials available to choose from. There have been numerous reported cases of early failures and even death following revision to metal containing bearings after fractures of ceramic components. This is thought to be due to the abrasive effect of hard ceramic particles eroding the softer metal surfaces and subsequently causing metal poisoning. The risk of developing such complications are however unknown, therefore this study aims to quantify the risk of re-revision or early death following such an event.

f. Health Economics Evaluation of Primary Total Hip Arthroplasty by Bearing Type and Fixation Modality (The Royal Orthopaedic Hospital NHS Foundation Trust; NJR:HES:CivReg:PROMs)

The objective of this study is to determine which hip replacement implants have performed best as per the data held by the National Joint Registry. Researcher will use information about each patient undergoing joint revision. The information will include need for revision surgery and Patient Reported Outcome Measures (PROMs). In addition to this it is necessary to examine the PROMs data for those who have undergone revision surgery (revision is where the joint replacement has been removed & a new joint replacement implanted). This will allow researchers to examine patient characteristics which may have led to the revision surgery as well as seeing how well the joint replacement was performing for that individual patient prior to revision surgery. Having this full set of data including the Hospital Episode Statistics (HES) data for each patient (which is why traceability is essential for this data) will allow researchers to examine whether patient factors such as comorbidities or previous surgery influence the satisfaction and performance after hip replacement surgery. By grouping data in to specific hip replacements (of which there are many available on the UK market) researchers can then see whether there are any outliers either performing particularly well or poorly and whether the performance and satisfaction with these hip replacements may be related to type of implant, comorbidities and similar. This study is a comprehensive analysis of the data available for England, Wales, Northern Ireland and the Isle of Man.

g. Evaluation of complex hip arthroplasty using the NJR dataset (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

Hip replacement is one of the most common operations performed in the NHS and one of the most successful procedures in terms of improving quality of life for patients with arthritis. Whilst a hip replacement will last many years for the majority patients, some may run in to problems and require ‘re-do’ (revision) surgery such as when an infection develops, the components become loose or when the bone around the implant breaks. This ‘re-do’ surgery is often more complex than the original surgery and has a higher chance of running into problems including requiring further ‘re-do’ surgery. The aim of this project it to obtain a perspective of the current delivery of this complex surgery at a national, regional and local levels. We aim to report on the kinds of operations which are taking place, where they are happening, who is doing them and how many they are doing. This study forms the basis of a wider project aiming to generate data on the best way to structure provision of such surgeries in the future.

h. What is the effect of the femoral head material on revision rate following primary total hip replacement (THR) using the Exeter cemented stem? (Northumbria Healthcare NHS Foundation Trust; NJR:CivReg)

There are several reasons why primary joint replacements fail. Most orthopaedic studies have looked at the type of fixation (cemented or uncemented) or type of articulations (metal or metal, metal on polyethylene or ceramic). What has not been looked at is the difference in the material between the stem and modular head combinations. There is greater understanding that metal ions can be released from this so called “trunnion wear” head-stem articulation and contribute to failure of joint replacements. This is due to corrosion in this area caused by difference in electrical charge of the materials. Researchers wish to take advantage of the data available from the National Joint Registry to compare revision rates of different combinations of stem and head in terms of material and explore if this has any effect on the long-term survival (prosthesis) of joint replacements.

i. A comparison of patient reported outcomes and adverse events following elective total shoulder arthroplasty and hemiarthroplasty – an exploratory study (Imperial College London; NJR:CivReg)

Shoulder pain is a common reason people visit their general practitioner and the number of shoulder replacements performed in England, Wales and Northern Ireland has doubled since 2012. A joint replacement is major surgery and it is associated with significant risks. It’s important that patients and surgeons know which type of replacement leads to the best results and the fewest complications. Young patients who are considering a shoulder replacement have a higher chance of further operations, this group need specific information to guide treatment. When people have a joint replacement pseudonymised information about the surgery is stored in the joint registry. In the years following surgery they are asked to complete questionnaires about their symptoms, function and quality of life. The aim is to use this large volume of data to compare the two main types of joint replacement performed for arthritis of the shoulder. The applicant organisation is requesting data for all shoulder replacements performed since the national joint registry began collection in 2012. This will be linked to the questionnaires given to patients after their procedure and to the NHS database of patient admissions to hospital. The information remains pseudonymised; a patient number is used only to ensure all the information corresponds to the same patient. Researchers will compare the results of patients who have received a half shoulder replacement and total shoulder replacement. The work will influence the decision making of patients and medical staff. The proposed studies will provide information about the expected recovery and the risks and benefits of a shoulder replacement. The work aims to improve patients’ shoulder function after surgery and reduce the number of additional procedures, complications and hospital admissions.

j. Effect of COVID-19 on NJR Surgical Practice and Outcomes (University of Sheffield; NJR:HES:PROMs:CivReg)

COVID 19 has had a great impact on all of our lives and changed society in ways that are yet to be identified. In this work, researchers will look at the outcomes of joint replacement surgery before COVID and then after COVID. This will allow researchers to see how outcomes of joint replacement surgery differ before and after the start of the COVID pandemic by looking at pain and activity, death and repeat operation. To do this, researchers will compare these measures over 2 years immediately before COVID, and compare these to the 2 years after restart of routine operating. In making these analyses researchers will use a range of mathematical tools to best fit the information we receive from the NJR. One of the approaches researchers will use is called “machine learning” in which the computer “learns” to predict the outcome based on the information put in. The other methods researchers will use are more traditional statistical approaches. As part of the study researchers will look to see which method makes the best predictions. Researchers imagine that the extra waiting time might affect these outcomes, as might surgeon deskilling. However, researchers cannot measure these things directly in the NJR, nor see if any patients had COVID around the time of their operation. Findings will be compared with 2 other registers that do record COVID positive tests. The work is exploratory, to see if there are differences in outcomes before and after COVID, and if these are found researchers will create “hypotheses” to explain the possible reason for the effects and test them in further patients

k. Periprosthetic fractures around primary total hip replacement (University of Leeds; NJR:CivReg)

Total Hip replacements (THR) are an effective treatment for managing end-stage hip arthritis. The implants used come in many different shapes and sizes, and work in different ways. The implants are designed to last as long as possible but a small amount fail because of loosening, infection, instability or breakage of the thigh bone (PFF) amongst other causes. In patients older than 70, PFF is the second most common cause of failure of THR.

PFF leads to significant problems for patients (more surgery, higher risk of medical complications including death) and their incidence is increasing. Unfortunately, PFF is difficult to prevent and one of the strongest predictors is the type of implants surgeons use. The NJR records when patients with PFF have their implants exchanged (revision). Previously, researchers identified a number of implant designs which make PFF more likely. This is already making an impact to reduce the risk of PFF. Unfortunately, previous data does not fully explain what happens to patients because patients can also be treated with no surgery and fixation surgery, which are not recorded in the NJR. In order to get a full picture of how the implants affect the risk of PFF researchers need to get information from more sources, which includes implant data (from the NJR), hospital data (which tells us about PFF not recorded in the NJR) and death statistics. This will make sure researchers can work out the complete picture of how the implants used affect the risk of PFF and ultimately how practice can be changed to prevent PFF from harming patients.

This will enable researchers to look at the most complete picture of PFF; how and when it occurs, how it affects patients and identify things which w can be changed to make PFF less likely in future. This will be the largest study looking at PFF. Although this research does not seek to make a fail-safe conclusion about which implant is safest; it will help the surgeons, patients and other stakeholders make informed decision in choosing the right implant for the right patient.

The following benefits to date include, but are not limited to:

• Providing outcomes data to the Medicines and Healthcare products Regulatory Agency (MHRA) that has led to the withdrawal of a number of orthopaedic devices from the market.

• Quickly identifying to hospitals those patients affected by a Field Notice or Device Alert issued by the MHRA.

• Direct influences on changes to clinical practice include:

o Preventing the use of hip resurfacing procedures in women over the age of fifty (reported through the NJRs Annual Report.

o Getting it Right First Time (GIRFT) proscribing the use of cemented hip implants in patients over 70 years of age (reported in the NJR’s Annual Report and supplemented by further analysis).

o Undertaking analysis that reduced to almost zero, the amount of large head, metal-on-metal hip replacements being undertaken.

o Undertaking analysis that resulted in a significant reduction in the number of hip replacement using a metal-on-metal articulating bearing in hip replacement.

o Professional societies, including the British Orthopaedic Association (BOA), issuing guidance to members about the minimum number of procedure types of procedure to be undertaken in a year in order to remain ‘current’. This follows an analysis of outcomes based on the number of procedure carried out by surgeons.

o A realignment of services for shoulder, ankle, and elbow joint replacement to a ‘hub and spoke’ service delivery on advice from GIRFT, based on an analysis of outcomes and the effects of low volume surgeons.

• The provision of enhanced data and reporting services to all stakeholders enabling them to make decisions about the care provided to patients. Stakeholders include clinicians, device manufacturers, service providers and commissioners, patients, regulators, and advisory bodies such as GIRFT and National Institute for Health and Care Excellence (NICE).

• Reducing the cost of orthopaedic devices to many NHS trusts by benchmarking outcomes data against the price paid for implants by those Trusts.

• Using risk-adjusted outcomes analyses to identify suspected outlier performance in surgeons, hospitals, and Trusts. This has enabled the appropriate responsible organisations and individuals to take the action necessary to rectify individual or system issues affecting outcomes. Such resolution has included the re-training of surgeons in certain procedure types and surgeons agreeing not to undertake specific procedure types. The NJR, in conjunction with the BOA, provides support to hospitals requesting it.

• By publishing and sharing data on outcomes, the NJR has informed changes in clinical practice that have reduced the rate of revision surgery. A reduction in the rate of revision surgery evidences improvements in patient outcomes and safety and also reduces the cost to the NHS of such surgery.

The National Joint Registry (NJR) is commissioned by the Healthcare Quality Improvement Partnership (HQIP) on behalf of NHS England as part of the Clinical Audit and Patient Outcomes Programme (NCAPOP). The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety.

This agreement has Joint Data Controllership - consisting of the Healthcare Quality Improvement Partnership (HQIP) and NHS England.

The NCAPOP is a large programme of circa 35 projects consisting of National Clinical Audits. HQIP is commissioned by NHS England to commission and manage the NCAPOP. NHS England is a controller of the NCAPOP jointly with HQIP as together both organisations determine the purposes and means of processing.

NHS England is responsible for determining which projects/topics are included as part of the NCAPOP. HQIP, as commissioner of the NCAPOP, is responsible for project specification development, procurement and extension activities, contract management and authorising data sharing requests. NHS England, as a funder of the NCAPOP, participates within specification development, procurement and project extension activities and authorises the publication of project outputs.

NHS England is involved with developing the scope and purpose of the NCAPOP projects through participation within specification development activities and may authorise (as chair of the specification development meetings) the final project specifications. These specifications set out the purpose of the project, the patient groups and clinical services to evaluate and the types of data to collect. NHS England are a representative upon the HQIP Data access request group which authorises data sharing applications from third parties.

Legal Basis Justification:

HQIP and NHS England both rely on the Article 6 (1) (e) legal basis under GDPR - "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". This is justified through commissioning arrangements which link back to NHS England and other national bodies with statutory responsibilities to improve quality of health care services.

HQIP rely on Article 9 (2) (i) as the legal basis for processing under GDPR - "processing is necessary for reasons of public interest in the area of public health, such as protecting against serious cross-border threats to health or ensuring high standards of quality and safety of health care and of medicinal products or medical devices, on the basis of Union or Member State law which provides for suitable and specific measures to safeguard the rights and freedoms of the data subject, in particular professional secrecy". This is justified as all projects aim to drive improvements in the quality and safety of care and to improve outcomes for patients.

NHS England rely on Article 9(2)(h) of the GDPR as the legal basis for processing. "Processing is necessary for the purposes of preventive or occupational medicine, for the assessment of the working capacity of the employee, medical diagnosis, the provision of health or social care or treatment or the management of health or social care systems and services on the basis of Union or Member State law or pursuant to contract with a health professional and subject to the conditions and safeguards referred to in paragraph 3". NHS England are responsible for provision of health and social care, and management of systems and compliance.

The NJR was established in 2002 by the Department of Health following a National Audit Office (NAO) report into the higher than expected failure rate of the 3M hip replacement device. The NAO report concluded that, had a national register of hip replacements been in existence, the failure rate would have been detected earlier. Earlier identification would have meant less patients were affected and the costs of revision surgery for the NHS would have been considerably less. The NJR went live in April 2003. Since its establishment, the responsibility for delivering the NJR has passed to HQIP who are contracted by the Department of Health to deliver the National Joint Registry.

HQIP \and NHSE are joint data controllers for the NJR and contract to two other organisations for the delivery of the programme, both of which are data processors:

• Northgate Public Services (NPS) is responsible for the NJR’s data collection and processing activities, including data storage and the provision of stakeholder reporting services.

• The University of Bristol (UoB)is also a data processor with responsibilities for statistical analysis and reporting.

HQIP and NHSE, have no access to the record level data. Northgate has access to record level data, including patient identifiers, in order to link the data provided to patient records held by the NJR. The linked dataset is pseudo-anonymised by NPS before it is made available to the statistical analysis team at the UoB. Although the UoB team has access to record level data, it does not have access to patient identifiers.

The work undertaken by the NJR is to monitor the outcomes of hip, knee, shoulder, ankle, and elbow joint replacement surgery with regards to the performance of devices, surgical teams, and Trusts and hospitals. The monitoring is necessary to ensure patient safety, improve patient outcomes.

The overall purpose of the NJR is summarised in its mission statement:

‘The purpose of the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man is to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety. In a continuous drive to improve the quality of outcomes and ensure the quality and cost effectiveness of joint replacement surgery, the NJR will monitor and report on outcomes, and support and enable related research.’

The strategic goals of the NJR are as follows:

• To monitor in real time the outcomes achieved by brand of prosthesis, hospital and surgeon, and highlight where these fall below an expected performance in order to allow prompt investigation and to support follow-up action.

• To inform patients, clinicians, providers and commissioners of healthcare, regulators and implant suppliers of the outcomes achieved in joint replacement surgery.

• To evidence variations in outcome achieved across surgical practice in order to inform best practice.

• To enhance patient awareness of joint replacement outcomes to better inform patient choice and patients’ quality of experience through engagement with patients and patient organisations.

• To support evidence-based purchasing of joint replacement implants for healthcare providers to support quality and cost effectiveness.

• To support suppliers in the routine post-market surveillance of implants and provide information to clinicians, patients, hospital management and the regulatory authorities.

The NJR has been collecting data since 2002 and is an ongoing audit with no planned end date. The NJR’s Steering Committee reviews the NJR’s mission statement and strategic goals annually to ensure that they remain relevant to existing stakeholder priorities, clinical guidance, and legislation.

The NJR requires record level data from HES, PROMs and Civil Registration in order to help it achieve some of its strategic goals. The NJR creates of an annual, linked dataset comprising data from the NJR, HES, Patient Episode Database Wales (PEDW), PROMs (NHS England programme), and Civil Registration. This dataset is used primarily for the production of the NJR’s Annual Report, its associated outcomes analyses, and in some reporting services provided to stakeholders. The data is also used for specific research projects undertaken by the NJR.

By linking to HES, PROMs and Civil Registration data, the NJR is able to enhance the quality and type of analyses that it undertakes. The priorities for NJR analyses and research are set by the NJR’s Research Committee and Editorial Board, both of which report to the NJR Steering Committee.

The data sets requested by the NJR consist of HES Patient Admitted Care, Civil Registration (Deaths), and Patient Reported Outcomes Measures. Patient identifiers are required in order to link the data requested from NHSD to patient identifiable data held by the NJR. This linkage is necessary to undertake analysis at a record level when, for example, considering factors such as length of stay for different joint replacement types, PROMs associated with different devices/implants, or outcomes associated with deprivation.

Civil Registration Data (Deaths) is essential for any outcomes analyses: patients who have died with an implant intact, for example, have to be excluded from any cohort. Outcomes analyses also includes a consideration of time to death from operation, and the cause of death.

HES Patient Admitted Care data is required to look at potential factors affecting the outcomes of joint replacement surgery, such as length of stay, re-admission to theatre, associated diagnoses and co-morbidities, deprivation, geography and location.

The NJR currently uses two endpoints in outcomes analysis: death and revision. PROMs represents an invaluable, third endpoint, providing a patients view, as opposed to a clinical view, of the outcome of joint replacement surgery. The NHS England programme is the only source of PROMs data relating to hip and knee joint replacement surgery.

The NJR requires data from 2003 which is when it started its own data collection. The number of years being requested is justified on the need to undertake longitudinal analyses of the available data. Currently, recommended guidance on the expected survivorship of joint replacement (as published by NICE) is based on ten years of data. Long term outcomes analyses are required.

Patient identifiable data is required to link the record level HES/Civil Registration/PROMs data to the relevant patient procedure in the NJR. The primary data field used for linkage is the patient's NHS number, but other fields such as date of birth, local patient ID, postcode, and side of procedure are used to ensure a reliable linkage. Data sets used for analyses of the linked data set have patient identifiers removed.

Research into the ‘unknown’ cohort comes under purpose 2 of the s251 support; and that there is sufficient oversight within the NJR internal assurance procedures to ensure that any research using the ‘unknown’ cohort’s data is within the parameters of purpose 2 of the s251 support.

In order to achieve minimisation, the NJR is requesting data for those OPCS4 codes relating to hip, knee, ankle, elbow, and shoulder joint replacement only. This will also ensure that Civil Registration data relating to those procedures only will be provided. By requesting data related to specific OPCS4 codes will ensure that the data requested is appropriate to the NJR's needs. Minimisation will be achieved by focusing on specific procedure codes.

• Linkage to HES data enables the NJR to improve the type and quality of the analyses that it can undertake without having to collect the data itself. Such linkage may be length of stay linked to outcomes or co-morbidities linked to outcomes. The NJR has, for example, published a paper examining the increased risk of cancer to patients who have had hip replacement procedures involving the use of metal on metal bearing surfaces, i.e. the femoral head and the acetabular cup or liner are both made of metal.

• By linking to PROMs outcomes analyses will improve as poor performance will be detected even if the primary joint replacement procedure has not been revised. A patient’s view of the outcome of joint replacement is a key indicator to the reporting of outcomes and provides an additional endpoint in addition to revision and death. Linkage to PROMs, and analysis at the record level, enables the NJR to monitor the performance of hospitals, surgeons, and individual implants.

• The NJR is required to publish 90 day mortality rates for all those NHS England Trusts undertaking joint replacement surgery in its Annual Clinical Reports to Trusts, the NJR’s Annual Report and as part of the former NHS England Clinical Outcomes Programme (although COP does not exist in its original form, the NJR still publishes annual indicators. The data, originally shared with Choices, is now shared with the CQC). . Date of death and any associated co-morbidities are essential for outcomes analyses and for risk-adjusting mortality analyses.

This will enable the NJR to continue to improve its analysis and data quality audits, activities which are expressly intended to improve patient outcomes and patient safety.

The recent S251 annual review application states

The NJR has also developed a secure, online Data Access Portal which provides additional security to the data used by researchers and analysts outside of the NJR. The plan is to provide subsets of linked data to approved research projects such that it is processed on the Portal and cannot be removed from the NJR’s security domain. The data does not contain patient identifiable fields. In February 2019, the NJR submitted a Research Application via IRAS. Once approved, the NJR will enter into a sub-licensing agreement with NHS Digital such that linked datasets can be provided for approved projects. This will ensure that more use can be made of the wealth of data held by the NJR.

There will be no sub licences permitted under this Agreement. Any future sub licencing would need the approval of HQIP as data controller and NHS Digital.

Data from Wales, Northern Ireland, the Isle of Man, and Guernsey are sourced separately.

The outputs from the use of the data will vary each year, dependent upon the research and analysis priorities set by the NJR’s Research Committee and NJR’s Editorial Board and agreed by the NJR Steering Committee.

a) The main vehicle for publication is the NJR’s Annual Report which is published annually in September. In addition to the Annual Report, a Patient’s Guide to the Annual Report is also published and made available in both hard and electronic copy. The Annual Report and the patients guide can be accessed at http://www.njrreports.org.uk.

b) Additional analyses are undertaken throughout the year and these are published either on the NJR website or in relevant, professional medical journals. To date those journals have included The Lancet, the British Medical Journal, and the Journal of Bone and Joint Surgery. The NJR maintains a dedicated research section on its website and details of planned, approved, completed, and published analysis can be found at: http://www.njrcentre.org.uk/njrcentre/Research/ResearchPortfolio/tabid/313/Default.aspx.

c) The outcomes of analyses are also disseminated at conferences and meetings of professional societies. These include the British Orthopaedic Association, the British Hip Society, the British Association for Surgery of the Knee, the British Elbow and Shoulder Society, and the British Orthopaedic Foot and Ankle Society. The NJR also holds two regional events per year which are aimed at those hospital staff responsible for collecting the data, providing them with an update on the work of the NJR, including its outputs.

d) Some outputs are also published through the NJR’s secure online reporting services. These services include NJR Clinician Feedback, NJR Management Feedback, and NJR Supplier Feedback which provide information for clinicians, trust/hospital management, and the manufacturers/suppliers of orthopaedic devices respectively.

The NJR is constantly reviewing the ways in which it communicates with patients and the public more generally, and are keen to improve its reach. Recent and planned activity include:

• working with our NJR Patient Representatives to recruit new members to the NJR Patient Panel to broaden the scope of patient voices on the NJR

• work with relevant patient groups such as Versus Arthritis to ensure that messaging for people who have had or plan to have joint replacement surgery are well targeted

• produce a series of patient guides http://www.njrcentre.org.uk/njrcentre/Reports-Publications-and-Minutes/Public-and-Patient-Guide co-designed with patients

• to press release our key developments to ensure appropriate media coverage. See example at https://www.dailymail.co.uk/health/article-7111723/Thinking-getting-new-hip-knee-Try-online-tool-tells-unique-risks.html

• to consult with the PPI group at the Royal National Orthopaedic Hospital, Stanmore about NJR’s data flows

• work with Understanding Patient Data to produce a video about our work https://understandingpatientdata.org.uk/case-study/monitoring-joint-replacement-surgery

• collaborating with HQIP and The Health Foundation as part of the Understanding Health Data Access (UHDA) programme which aims to improve information available publically about secondary data uses

• make a newsletter available via our website which the public are able to subscribe to. We also have blog posts from joint replacement patients hosted on our website.

• to remain active on social media platforms (Facebook, Twitter, Linkedin)

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

The following benefits to date include, but are not limited to:

• Providing outcomes data to the MHRA that has led to the withdrawal of a number of orthopaedic devices from the market.

• Quickly identifying to hospitals those patients affected by a Field Notice or Device Alert issued by the MHRA.

• Direct influences on changes to clinical practice include:

o Preventing the use of hip resurfacing procedures in women over the age of fifty (reported through the NJRs Annual Report.

o GIRFT proscribing the use of cemented hip implants in patients over 70 years of age (reported in the NJR’s Annual Report and supplemented by further analysis).

o Undertaking analysis that reduced to almost zero, the amount of large head, metal-on-metal hip replacements being undertaken.

o Undertaking analysis that resulted in a significant reduction in the number of hip replacement using a metal-on-metal articulating bearing in hip replacement.

o Professional societies, including the British Orthopaedic Association (BOA), issuing guidance to members about the minimum number of procedure types of procedure to be undertaken in a year in order to remain ‘current’. This follows an analysis of outcomes based on the number of procedure carried out by surgeons.

o A realignment of services for shoulder, ankle, and elbow joint replacement to a ‘hub and spoke’ service delivery on advice from GIRFT, based on an analysis of outcomes and the effects of low volume surgeons.

• The provision of enhanced data and reporting services to all stakeholders enabling them to make decisions about the care provided to patients. Stakeholders include clinicians, device manufacturers, service providers and commissioners, patients, regulators, and advisory bodies such as GIRFT and NICE.

• Reducing the cost of orthopaedic devices to many NHS trusts by benchmarking outcomes data against the price paid for implants by those Trusts.

• Using risk-adjusted outcomes analyses to identify suspected outlier performance in surgeons, hospitals, and Trusts. This has enabled the appropriate responsible organisations and individuals to take the action necessary to rectify individual or system issues affecting outcomes. Such resolution has included the re-training of surgeons in certain procedure types and surgeons agreeing not to undertake specific procedure types. The NJR, in conjunction with the BOA, provides support to hospitals requesting it.

• By publishing and sharing data on outcomes, the NJR has informed changes in clinical practice that have reduced the rate of revision surgery. A reduction in the rate of revision surgery evidences improvements in patient outcomes and safety and also reduces the cost to the NHS of such surgery.