Towards UK post Arthroplasty Follow-up rEcommendations (UK-SAFE) - Work Package 2a (RO-HES)

Most people now understand the need for a cost-effective NHS, but seek reassurance that this will not reduce the standard of care. This is particularly true of older people, who are the group most likely to be affected by this research.

Total joint replacement provides considerable improvement in quality of life to people suffering with severe joint damage and in 2013 over 150,000 total hip and knee replacements were conducted. Due to increasing ageing and obesity in the UK population, this number is likely to increase each year. Sometimes, problems can develop with the replaced joint over time and a small percentage of people require further surgery. Because joint failure is not always associated with symptoms, follow-up care is provided to ensure that problems are identified as early as possible.

Towards UK post Arthroplasty (the surgical reconstruction or replacement of a joint) Follow-up recommendations (UK-SAFE) is funded by the National Institute for Health Research (NIHR) - Health Services and Delivery Research (HS&DR) with reference 14/70/146

(see https://www.journalslibrary.nihr.ac.uk/programmes/hsdr/1470146).

The project will examine the requirements for arthroplasty follow-up and produce evidence and consensus-based recommendations as to how, when and on whom follow-up should be conducted. The University of Leeds have assembled a highly experienced multi-disciplinary team from the Leeds and Oxford NIHR Musculoskeletal Biomedical Research Units with the necessary academic expertise to tackle the methodological challenges in undertaking this work to ensure the project yields meaningful results that are accepted by key stakeholders.

UK-SAFE is a research project sponsored by the University of Leeds and funded by the National Institute of Health Research (NIHR) Health Services and Delivery Research Programme (see https://www.fundingawards.nihr.ac.uk/award/14/70/146).

UK-SAFE comprises of a number of Work Packages, which are led by co-applicants from different institutions, including the University of Leeds and the University of Oxford.

The specific work package to which this application relates - UK-SAFE WP2a (RO-HES) - is being undertaken at the University of Leeds. University of Leeds is the data controller in respect of the data within this work package. University of Oxford is not a data controller or processor in respect of any data for UK-SAFE WP2a (RO-HES).

Further information regarding the Work Packages can also be found on the UK-SAFE Project Web site (see https://leedsbrc.nihr.ac.uk/uk-safe-project/).

Work Package 2 of the UK-SAFE project focuses on 'determining when, which and how patients present for revision surgery' and is itself divided into two Work Packages:

1. Work Package 2a which focuses on analysis of data from routine datasets. Work Package 2a is further sub-divided into three Work Packages:

i) Work Package 2a (CPRD-HES) - led by the University of Oxford

ii) Work Package 2a (RO-HES) - led by the University of Leeds

iii) Work Package 2 (NJR-HES-PROMS) - led by the University of Oxford

2. Work Package 2b which focuses on analysis of a prospective cohort.

The 3 Work Packages within Work Package 2a have been designed to provide different, complementary views on 'when, which and how patients present for revision surgery' by drawing on different routine datasets. Work Package 2a (CPRD-HES) and Work Package 2a (RO-HES) provide different, complementary views that include consideration of primary care. Inclusion and exclusion criteria for cohorts and data items, and analytical techniques to apply to these data, have been determined by the research teams at the University of Oxford and the University of Leeds to ensure the comparability of results between Work Package 2a (CPRD-HES) and Work Package 2a (RO-HES).

This application (DARS-NIC-147997-R8B9S) relates only to Work Package 2a (RO-HES), which is led by the University of Leeds. University of Oxford is separately responsible for Work Package 2a (CPRD-HES) and the other Work Packages within the UK-SAFE project for which they provide the leadership.

The University of Leeds is the Data Controller for the data referenced in this application, and the data is processed at the locations listed in 'Section 1c - Data Processor(s)' of the application. Analysis of the data supplied from this application will be undertaken by a team of researchers based at the University of Leeds only. Aggregated results from analysis will then be reviewed by the Study Management group.

The UK SAFE project will examine the requirements for arthroplasty follow-up and produce evidence and consensus-based recommendations as to how, when and on whom follow-up should be conducted.

UK SAFE comprises of 3 complementary evidence synthesis work-packages to inform a final consensus process.

Work packages 1-3 will be discrete research projects, conducted independently. However, knowledge and information from each will feed into and inform the other work streams with some tasks being conducted in parallel.

This project aims to investigate whether good after-care is expensive, in terms of time and money on the part of both patient and hospital staff, and whether individual patient-centred follow-up can better identify potential problems in a timely fashion, to the benefit of all concerned.

Membership of the Study Management Group (SMG) comprises all co-investigators of the UKSAFE programme. The SMG provides advice on the conduct, delivery and dissemination across all work packages in the programme. Under the conditions of funding, the Principal Investigator, (University of Leeds) has responsibility from the National Institute of Health Research (NIHR) for the delivery of the UKSAFE programme. Within this programme, University of Leeds has the sole responsibility for the delivery of the particular work package of the UKSAFE programme to which this application relates (WP2a RO-HES). The purpose for, and means by which this work package is to be delivered have been specified and agreed in accordance with the NIHR funding process. Any comments received from the SMG in respect of the work package are advisory.

This agreement is for data that will be used on Work Package 2a (RO-HES) which is being undertaken at the University of Leeds.

Work Package 2a (RO-HES) has the following objective:

- To determine when, which and how patients present for revision surgery

To fulfil this objective, Work Package 2a (RO-HES) will undertake a retrospective cohort study using the following sources of routine NHS data:

1. Hospital Episode Statistics (HES) : a database provided by NHS Digital containing patient data relating to A&E, inpatient and outpatient episodes

2. ResearchOne : a research database containing de-identified patient data from primary care settings that use the SystmOne clinical information system.

For patients with an inpatient admission (Index Episode) to a hospital in England within the defined index period (1st April 2000 – 31st March 2015) for one of the following procedures:

i) hip replacement

ii) hip revision

iii) knee replacement

iv) knee revision,

The following data will be obtained from Hospital Episode Statistics (HES) for the period from 1st April 2000 to 31st March 2015:

- A&E attendances with

i) a diagnosis of "dislocation/fracture/joint injury/amputation"

ii) a diagnosis (anatomical area) of Hip, Groin, Thigh, Knee or Lower Leg.

- Inpatient admissions with a treatment specialty of:

i) Trauma and Orthopaedics

ii) Rheumatology.

- Outpatient appointments with a treatment specialty of:

i) Trauma and Orthopaedics

ii) Rheumatology.

For patients included within the HES data, the study will determine whether primary care data exists for the patients within TPP's ResearchOne database. If so, a defined set of primary care data will be obtained for these patients for the period prior to 1st April 2015. Additionally, a mapping file will be produced by NHS Digital and supplied to the University of Leeds to enable the HES data for a specific patient to be linked to the corresponding primary care data (if present).

From the linked data, longitudinal records covering primary and secondary care will be constructed for patients.

Based on data provided within the National Joint Registry (NJR) Annual Report, 800,683 primary hip replacements, 89,023 hip revisions, 875,585 primary knee replacements and 54,278 knee revisions were reported (subject to specific inclusion and exclusion criteria) the period from 1 April 2003 to 31 December 2015. Therefore, for a comparable period within the study, it is anticipated that around 900,000 hip replacements and 1,000,000 knee replacements (1,900,000 joint replacements in total) will be identified.

The scope of all hospitals in England has been chosen to enable variation in follow-up for hip and knee replacements and revisions between hospitals to be analysed. The time period of 15 years over which patients will be identified has been chosen to enable variation in follow-up for hip and knee replacements and revisions for patients over time to be analysed.

The legal basis under which data is processed within UK-SAFE Work Package 2a (RO-HES) is:

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

The University of Leeds (as it is formally and legally known) is an independent corporation established by Royal Charter. Its objects, powers and framework of governance are set out in the Charter and its supporting statutes, amendments. The University of Leeds received its Charter as an independent institution from Edward VII in 1904. The royal charter under section II states that The University shall have the powers following:

“To provide for instruction in such branches of learning as the University may think fit, and also to make provision for research and for the advancement and dissemination of knowledge”.

Processing of data concerning health (and other ‘special categories’ of data) requires the fulfilment of an additional condition under GDPR. The specific condition fulfilled by the processing of such data within UK-SAFE Work Package 2a (RO-HES) is:

Article 9 (2) (j) - "processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject."

The data are required for research purposes in the public interest - meeting the conditions in the DPA 2018 Schedule 1 Part 1 (4) - which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data.

A summary of the data flows, which are performed using either: i) NHS Digital's Secure File Transfer (SEFT) or ii) University of Leeds LICTR's Secure File Transfer (SFT), is provided below. This list below explains the TRANSFER MECHANISM - further details of the full flow are included later on in this section.

1. University of Leeds provides TPP (SystmOne) with the cryptographic salt (via SFT)

2. University of Leeds provides NHS Digital with the cryptographic salt (via SEFT)

3. NHS Digital provides the University of Leeds with HES A&E, Inpatient and Outpatient data according to specified criteria with patients referenced by HESID (via SEFT)

4. NHS Digital provides TPP (ResearchOne) with a set of patient pseudonyms (via SEFT).

5. TPP (ResearchOne) provides University of Leeds with primary care data for patients identified from the set of patient pseudonyms supplied by NHS Digital with patients referenced by ROID (via SFT).

6. TPP (ResearchOne) provides NHS Digital with a mapping from each patient pseudonym to a unique ROID (via SEFT)

7. NHS Digital provides University of Leeds with a mapping file that maps a unique HESID to a unique ROID identifier (via SEFT)

The data which will be accessed by the University of Leeds will be pseudonymised. They will have no access to identifiers. They are not permitted to use the data provided to try and re-identify any individual.

Importantly, University of Leeds do not receive any pseudonyms from NHS Digital or ResearchOne.

University of Leeds receives: i) data from NHS Digital where patients are referenced by a HESspecific identifier, ii) data from ResearchOne where patients are referenced by a RO-specific identifier, and iii) a file from NHS Digital which maps HES-specific identifiers to RO-specific identifiers. Linkage of the HES-specific identifier to RO-specific identifier is undertaken at NHS Digital.

ResearchOne has been separated out in the summary above to TPP (ResearchOne) and TPP (SystmOne). Project documentation (Protocol) refers only to ResearchOne to emphasise that the project will use only data from the ResearchOne database. The Phoenix Partnership (TPP) are responsible for SystmOne, ResearchOne, and for ensuring the necessary separation of duties for staff working on SystmOne and ResearchOne.

NHS Digital and TPP will transfer the set of patient pseudonyms (via SEFT) in line with the correct procedures . The ResearchOne database is stored and processed inside the data centres which host the SystmOne production environment. The databases are in Tier 3 data centres that have been accredited by the NHS for storing and processing clinical data. There is a disaster recovery site (hosted to exactly the same standards as the live site) which is used as a back-up location in the event of any failures at the main site. Documentation relating to ResearchOne can be found at:

http://www.researchone.org/documentation/.

University of Leeds generates and supplies a salt key that is used in the generation of pseudonyms by NHSD and TPP (SystemOne). To ensure that the identifier used to generate the pseudonyms (NHS number) cannot be recovered by the University of Leeds, the University of Leeds does not receive any pseudonyms generated using the supplied salt key.

The salt key is a random string. The salt key is generated for this specific project by the University of Leeds and securely transferred to NHS Digital and TPP. The key is then used by NHS Digital and TPP in conjunction with NHS number to generate pseudonyms for patients. These pseudonyms are used by TPP to determine whether patients are present in ResearchOne and by NHS Digital to generate a mapping file that is provided to the University of Leeds.

The “salt” file represents a random string of defined length that is used as an additional input to the pseudonym generation process to reduce the risk

that: i) NHS numbers could be recovered from pseudonyms, and ii) pseudonyms generated from NHS numbers for different projects (and any associated data) could be matched. Use of the salt key in the pseudonym generation process provides a safeguard against the potential recovery of the NHS number from a pseudonym by a brute-force attack – where pseudonyms are generated for all possible NHS numbers – as such an attack would also require to the attacker to have access to the salt key.

Data extracts from ResearchOne contain unique references for each patient that are project-specific – i.e. the reference for a given patient will be different in any two data extracts from ResearchOne. The terms ‘ROID’ or ‘ResearchOne ID’ are used for these references within the project documentation.

o For a given data extract, these project-specific unique references are created and assigned as part of the extract generation process.

o To facilitate linkage with data from NHS Digital for this project, TPP will also generate a pseudonym for each patient as part of the extract generation process, using the salt supplied by UoL and the NHS number contained in the SystmOne data (from which ResearchOne data is generated). (Point 1 + 2 + 3 above)

o NHS Digital will supply a list of pseudonyms to TPP which will represent those patients to include in the data extract from ResearchOne. (Point 4 above)

o TPP will determine those patients for whom data exists within ResearchOne by matching pseudonyms in this list with those created as part of the extract generation process.

o TPP will then supply NHS Digital with a file that maps pseudonyms to ROIDs for the subset of patients who are present in ResearchOne. (point 6)

o NHS Digital will use the file supplied by TPP to generate a file that maps HESIDs to corresponding ROIDs. The file generated by NHS Digital will be supplied to UoL to enable data relating to a given patient to be determined in both the HES and ResearchOne data. (Point 7 above)

o TPP will also supply UoL with a data extract containing the required data items where each patient is uniquely referenced by a ROID. TPP does not supply the pseudonyms to UoL. (Point 5)

The University of Leeds receives the following:

i) HES data referenced by a HES ID

ii) ResearchOne data referenced by a RO ID

iii) a mapping file from HES ID to RO ID.

The linkage methodology has been developed so that the University of Leeds is responsible for the generation and supply of the salt key, and subsequent data flows to the University of Leeds do not contain any data that could be re-identified by the knowledge of the salt key. Additionally, the salt key is used only a one-time basis, such that any pseudonyms generated cannot be matched across data sets obtained for different usages. Approvals for the project (including NHS Ethics) have been obtained on the basis of the proposed linkage methodology.

NHS Digital will identify patients based on the presence of an inpatient admission (Index Episode) to a hospital in England within the defined index period (1st April 2000 – 31st March 2015) for one of the following procedures: i) hip replacement, ii) hip revision, iii) knee replacement, or iv) knee revision. For these patients, NHS Digital will identify any of the following in the period for 1st April 2000 – 31st March 2015:

- A&E attendances with i) a diagnosis of "dislocation/fracture/joint injury/amputation" or ii) a diagnosis (anatomical area) of Hip, Groin, Thigh, Knee or Lower Leg.

- Inpatient admissions with a treatment specialty of i) Trauma and Orthopaedics or ii) Rheumatology.

- Outpatient appointments with a treatment specialty of i) Trauma and Orthopaedics or ii) Rheumatology.

This data will be provided directly to the University of Leeds with patients uniquely referenced by a unique HESID. Additionally, NHS Digital will generate a pseudonym for each included patient, according to the methodology described above, and supply this list of pseudonyms to TPP (ResearchOne).

TPP will generate pseudonyms for all patients in the ResearchOne database by applying the same methodology as NHS Digital and using the salt key supplied to TPP (SystmOne). TPP (ResearchOne) will then determine those patients whose pseudonyms are present in the list provided by NHS Digital. Selected primary care data will be extracted from ResearchOne for these patients for the period prior to and including 31st March 2015.

Primary care data will only be obtained where:

- the GP practice to which the patient is registered using TPP's SystmOne clinical information system and

- has opted-in to ResearchOne (and the patient has not individually opted-out).

TPP (ResearchOne) will provide primary care data directly to University of Leeds. Each patient will be uniquely referenceable in this data using a unique ROID. No pseudonyms will be included in the data provided by ResearchOne to the University of Leeds. TPP (ResearchOne) will also provide NHS Digital with a file that maps each pseudonym present in the list provided by NHS Digital and in the ResearchOne database to its corresponding ROID.

NHS Digital will use the mapping file provided by TPP (ResearchOne) to map the HESID of each patient in the HES data to the ROID identifier of each patient in the ResearchOne data. NHS Digital will supply the resulting mapping file to the University of Leeds who will use it to match the data for a particular patient in HES data to the data for that patient in the RO data (and vice-versa).

Survival analysis will be used to model time to revision. To determine the follow up time of revision, a smoothed Nelson-Aalen cumulative hard rate will be examined to identify any peak in the mid-long term risk of revision. Survival analysis is a statistical technique used to analyse the time until a particular event happens. In this case, the particular event would be a hip/knee revision.

To identify patients most likely to require revision, proportional hazards regression will be used to identify pre-, peri- and post- operative predictors of mid-late term revision.

The project seeks to determine those factors that predict mid-late term revision. The factors are categorised by the time at which their value is determined relative to the operation: pre (before) peri (during) and post (after). Cox proportional hazards regression modelling is a statistical technique used to determine associations between the time of an event happening (e.g. revision) and these factors.

The date of the first incidence of a subject's hip or knee replacement will be used as the start time. The event of interest in all time-to-event models will be the first recorded revision operation. Cox proportional hazards regression modelling will be used to identify pre-, peri- and post- operative predictors of midlate term revision (defined as more than 5y post primary surgery). Should testing reveal that the proportional hazards assumption is not valid, then parametric modelling will be used instead. Shared frailty will be modelled with a random effect for hospitals/providers and another for general practice, since it is anticipated that both hospital practice regarding follow-up and GP behaviour regarding referral will influence the survival time of the primary joint replacement. Competing risks, including mortality and comorbidities following primary surgery will be considered. Linearity of continuous predictors will be assessed using fractional polynomial regression modelling or splines. Analysis will assess the extent to which the relationship between the value of continuous predictors (i.e. predictors whose values might be, for instance, decimal numbers) and time can be represented by a linear function (straight line).

Missing data will be handled by using multiple imputation methods using the ICE (Imputation by Chained Equations) procedure.

Inclusion and exclusion criteria for patients and data items were determined by members of the project team who provided clinical, technical and statistical input. Only those patients and data items determined to be necessary and sufficient to answer the research question were included.

UK-SAFE WP2a (RO-HES) is not part of a PhD project.

All persons accessing the record level data provided by NHS Digital under this agreement are substantive employees of University of Leeds.

Employees of University of Leeds receive training in data protection and confidentiality.

University of Leeds will not link the data further and the only data linkages are those permitted under this application / Data Sharing Agreement.

This project may deliver the first research-supported, best-for-patient, joint-specific, cost-effective recommendations for follow-up care, providing a gold standard for clinical excellence, and follow-up advice for patients, surgeons, purchasers and health services. Value is not limited to the UK, but has massive global potential.

Nationally, (if successful) the outputs, in the form of an executive summary statement of the agreed pathway/s will be disseminated through appropriate NHS Networks, the NHS England Elective Orthopaedics Subcommittee, the NHS Institute for Innovation and Improvement and professional societies. Dissemination will be key to developing a culture of ‘finding the best way of doing something and doing it everywhere’ to significantly reduce wastage of clinical resources and optimise NHS spend. The University of Leeds have support of the British Hip Society (BHS), British Orthopaedic Association (BOA), British Association for Surgery of the Knee (BASK), NHS England, Arthroplasty Care Practitioners Association (ACPA), the National Joint Registry and three Leeds-based CCGs for this research and for dissemination activities. This includes individuals from these organisations sitting on the project Independent Advisory Group and attendance of representatives from each organisation at the final dissemination meeting in order to promote further cascading of study results. They will not influence the results in any way - just advise on how is best to cascade accordingly.

Support from these organisations in this context refers to broad agreement with the aims and objectives of the project and an interest in the outputs generated by the project. These organisations have no data controller or processor responsibilities in respect of the data from NHS Digital.

The study may put forward the consensus statement to each society’s AGM for adoption as a resolution (if successful).

Internationally, dissemination platforms are already in place through the International Society of Arthroplasty Registers (ISAR) and the European Federation of National Associations of Orthopaedics and Traumatology (EFORT).

Overall and individual work-packages findings may be disseminated through a variety of media. Abstracts may be submitted to major British and international orthopaedic conferences, and separate relevant meetings, including the Health Economics Study Group and Exploiting Existing Data for Health Research conference. The University of Leeds will look to present at the NIHR Methodology Conference and NHS Management conferences and events. Manuscripts will be submitted to appropriate peer-reviewed journals, including general medical, orthopaedic and management journals if appropriate.

Patient dissemination will be supported through the Leeds Biomedical Research Centre Patient and Public Involvement (LBRC PPI) forum and website. Patients with relevant lived experience of the conditions will be involved throughout the programme of work, including one PPI co-applicant and two lay members of the study Independent Advisory Group (both patients with relevant lived experience). The University of Leeds (if appropriate) will hold a PPI conference at the end of the study and will encourage the PPI representatives to be involved in presentations, with support from research staff, to help ensure conference material is appropriate. With lay representatives the University of Leeds will write a lay summary for publication in a patient publication such as The Patient or Inspire (if the project is successful).

To minimise any risk that data published by the research team discloses the identity or attributes of specific patients.

All published outputs from WP2a (RO-HES) will be assessed against the NHS Anonymisation Standard (see http://www.isb.nhs.uk/documents/isb-1523/amd-20-2010/1523202010sp ec.pdf) and ICO's Anonymisation: managing data protection risk - code of practice (see https://ico.org.uk/media/1061/anonymisation-code.pdf) and approved by the Principal Investigator prior to publication.

Any published outputs generated using the data from NHS Digital will contain only aggregated data with small numbers suppressed in line with the HES Analysis Guide

Benefits as described in the NIHR funding application.

Upon completion, this research will have major immediate effect on national NHS planning and budgeting and patient well-being. The outputs will be evidence-based support for timing of followup and identification of the most cost-effective follow-up model. This fits directly within the NHS framework for improving outcomes from elective procedures. Rationalising current diversity of followup practices should enable substantial savings for the NHS. Novel follow-up strategies, such as creating a rapid access pathway after joint replacement for symptomatic patients will be examined. The University of Leeds envisage outputs to be readily applicable to the wider NHS, not only hip and knee but also other joint replacements.

The impact will be to reduce the burden on patients and the NHS in terms of outpatient visits and clinical tests that do not add benefit, while optimising detection of potential problems. From an NHS perspective, this work will provide NHS managers with economic and clinical information on arthroplasty follow-up to inform service planning and delivery, and the role of arthroplasty practitioners in this service; provide orthopaedic surgeons with guidance on follow-up, including patient and economic considerations of factors involved; produce arthroplasty follow-up guidelines for adoption by the relevant specialist societies and inclusion with information for their members. From a patient perspective, this work will help to inform patients about follow-up practice and empower them to make choices about future healthcare relating to their joint arthroplasty.

At the end of the project, a policy document will be created with support of the relevant societies, NHS England, CCGs and patient representation. It is anticipated that this will include a stratification algorithm to determine appropriate follow-up for an individual patient, taking into account, for example, implant type and patient factors, and that recommended follow-up pathways for hip may differ to those for knee. This advisory document will be disseminated to all stakeholders, including orthopaedic surgeons, arthroplasty surveillance professionals and NHS managers. With the committed support of these key organisations, it is anticipated that these guidelines will be positively received and that implementation will be widespread. It is the University of Leeds's ambition for the recommended follow-up pathway/s defined by this programme of work to be adopted for all hip and knee replacement patients in the UK and internationally.

Not stated in the register.