Does patient ethnicity predict outcome following neck of femur fracture?

No further data will flow under this version of the agreement.

There are large variations in the outcomes of orthopaedic surgery across ethnic groups in England and Wales, with patients from Black and Minority Ethnic (BAME) backgrounds struggling to access care and having worse outcomes.

A pilot study at a University Teaching Hospital which serves a large BAME population assessed a locally held database of neck of femur (NOF) fractures. Fracture patterns and outcomes post-fracture in patients from different ethnic backgrounds were compared. South Asian patients were more likely to have the worse fracture types, more likely to be discharged to their own home, and tended to have more co-morbidities and worse mobility. Achievement of best practice tariff (BPT) targets during care was associated with preferentially better outcomes in White patients versus South Asian patients. These findings have not been demonstrated previously.

Traditionally accepted distribution for fracture types, patient demographics and patient mobility may be inconsistent across ethnic groups. As a result, targets for ideal treatment may not be reliably associated with a reduction in mortality in BAME groups. The aim of this data sharing agreement is to compare the outcomes following a NOF fracture across ethnic groups in the UK and explore reasons if any differences are observed which may include physiological differences between groups or other social determinants of health. This analysis is being completed in support of the National Hip Fracture Database service evaluation activity “to improve the quality and cost-effectiveness of hip fracture care; and to reduce the incidence of the injury by improving secondary prevention”. The NHFD has section 251 approval in support of its aim "To support audit, service evaluation and quality improvement activity by NHFD, NHFD participant sites and approved third parties". The analysis being undertaken by the University of Leeds is in the capacity of a service evaluation by an approved third party.

Pseudonymised data will be acquired from the NHFD. This will be linked with Hospital Episode Statistics (HES) Admitted Patient Care (APC) and Civil Registration (Deaths) data, as well as equivalent hospital episode data for patients attending Welsh hospitals using a unique study ID given to all patients in the NHFD. Ethnic groups deemed to have high stability from census data will be used to estimate ethnic group specific demographics, fracture types and risk factors for mortality. Ethnic groups assessed will be White (British/ Irish/ Other), South Asian (Indian, Pakistani, Bangladeshi), Asian (Chinese), Caribbean and African.

Data previously required from NHS Digital’s HES APC and Deaths datasets includes demographic data (e.g. Age on admission, Ethnic Category, Sex of patient, Index of Multiple Deprivation, Lower Super Output Area (LSOA)), as well as data relating to the diagnosis of, procedures relating to, and subsequent outcomes of patients with NOF fractures. These demographic covariates will be controlled for to study the effect of ethnicity.

The University of Leeds will link NHFD data with the pseudonymised NHS Digital data with the equivalent pseudonymised Patient Episode Database for Wales (PEDW) data to investigate outcomes following NOF fracture by ethnicity. Pseudonymised NHFD data includes variables relating to fracture side, classification and pathology; surgical variables; post-operative outcomes; BPT criteria (geriatrician/ specialist fall/ multi-disciplinary assessments); discharge variables (when and where) and follow-up variables (mobility, medications, residential status).

Selecting smaller geographical areas rather than requesting national data was considered but is not feasible because it would affect the robustness of the findings. The previous pilot study conducted in a geographical location with a high South Asian population included only 53 South Asian patients and even fewer from other ethnic backgrounds from a sample of 1,520 patients. A national data set is required to draw firmer conclusions and study patients from various BAME backgrounds. The geographical spread is required to be England and Wales specifically since this is the geographical coverage of the NHFD. The aim is to include all NOF fractures and the inclusion criteria are a hip fracture on the NHFD between 01/01/2009 and 01/02/2021. This timeframe is necessary to have robust, large scale data to be able to draw accurate conclusions. The data requested is the minimum data set required to achieve the aims of the service evaluation. There are no alternative, less intrusive ways of achieving the purpose.

The NHFD itself is an audit which aims to improve the processes of care and the outcomes for patients with hip fractures, as well as ensure reliable information on quality of care and outcomes is available for patients and the public. The outcomes of this service evaluation align with the outcomes of the NHFD, but include a break down by ethnicity, including specific demographic factors not covered by the current NHFD. The primary outcome will be patient mortality at 365-days post fractured NOF for BAME patients as compared to the White population.

The secondary outcomes will be to assess:

- Differences in mortality at 30- and 120-days

- Associations of clinical comorbities with increased mortality

- Differences in fracture type

- Differences in mobility and residential status

- If BPT achievement is associated with a different outcome

The University of Leeds will undertake this work to support service evaluation and quality improvement activity by the NHFD. University of Leeds will be the controller and processor of this data. HQIP act as data controllers for the NHFD but do not have access to linked data analysed by researchers at University of Leeds. The project is funded by the Royal College of Surgeons of Edinburgh (RCSEd). RCSEd do not determine the purpose or the manner in which the data will be processed.

The justification for processing the data by University of Leeds is Article 6(1)(e) of the GDPR: (processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller).

The justification for the processing of the special category data (health data) by University of Leeds is Article 9(2)(j) of the GDPR: (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.

There is no wider project or collaboration planned.

No further data will be disseminated under this version of the Agreement.

Under version v0.9 of the Agreement, Crown Informatics sent NHS Number, date of birth, Full postcode, Name and FFFAP ID (a unique National Hip Fracture Database (NHFD) person identifier) for all patients in the NHFD database with a neck of femur (NOF) fracture to NHS Digital and NHS Wales Informatics Service (NWIS). Pseudonymised data with the same FFFAP ID flowed to researchers at University of Leeds, with NHS Digital providing HES APC and Civil Registrations (Deaths) data, and NWIS providing Patient Episode Database for Wales (PEDW) data. HES APC data was filtered to include only the spell relating to the neck of femur (NOF) fracture admission, based on the NOF fracture diagnosis code.

Crown Informatics also sent an encrypted version of pseudonymised NHFD data to the study chief investigator (CI) at the University of Leeds who transferred data to Leeds Institute of Data Analytics (LIDA). The CI nominated LIDA members to receive the password for data decryption from Crown Informatics.

The pseudonymised data from NHS Digital, NHS Wales Informatics Service and the NHFD will be linked using the FFFAP ID inside the LASER cloud platform at the University of Leeds as needed for analysis.

Patients in the ethnic minority groups will be matched independently to a cohort of the control population (White). Odds of dying at 30 days, 120 days and 365 days respectively will be analysed between White and ethnic minority groups with adjustment for all clinically relevant covariates.

Data processing is carried out by substantive employees and honorary contract holders from the University of Leeds who have been appropriately trained in data protection and confidentiality. Researchers who work on the pseudonymised data set do not have access to the list of patient identifiers sent to NHS digital for linkage purposes.

There is no requirement, and no attempts will be made to identify individual patients. The full Date of Death is required to be able to calculate Kaplan Meier survival curves. Researchers will not be linking NHS Digital data with any other datasets except those listed above. There will be no capacity for researchers to identify any individual patients.

Data will be held at the LIDA, University of Leeds on the LASER cloud platform which is supported by Microsoft Azure. Microsoft Limited supply Cloud Services for the University of Leeds and are therefore listed as a data processor. They supply support to the system, but do not access data. Therefore, any access to the data held under this agreement would be considered a breach of the agreement.

Access to the data is only granted to users who have an authorised platform account and the correct permissions to access the project-specific secure environment. To access this secure environment, those authorised users will log into the platform using their credentials and multifactor authentication. Once logged in, they can remotely connect to the secure environment. The secure environment does not permit internet connection, clipboard sharing (copy+paste between systems), nor any other mechanism to access/transfer files unless via means that are documented in the platform’s Information Security Management System.

In order to protect patient confidentiality, when presenting results calculated from HES record level data, outputs will contain only aggregate level data with small numbers suppressed in line with HES Analysis Guide. When publishing HES data, the University of Leeds must ensure that:

· cell values from 1 to 7 are suppressed at a local level to prevent possible identification of individuals from small counts within the table.

· Zeros (0) do not need to be suppressed.

· All other counts will be rounded to the nearest 5.

Data will not be made available to any third parties other than those specified except in the form of aggregated outputs with small numbers suppressed in line with the HES Analysis Guide.

All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract ie: employees, agents and contractors of the Data Recipient who may have access to that data)”

There will be no data linkage undertaken with NHS Digital data provided under this agreement that is not already noted in the agreement.

The NHFD produce an annual report in which the findings of this evaluation are expected to be included. A report to the Royal College of Surgeons of Edinburgh (RCSEd) as the funding body will also be prepared.

To reach a wider scientific audience, submission to peer reviewed journals are intended to be prepared including to The New England Journal of Medicine, The Lancet, British Medical Journal, and Bone and Joint Journal. It is planned that presentations at Orthopaedic conferences including the American Academy of Orthopaedic Surgeons and British Orthopaedic Association will be prepared.

All outputs will be aggregated with small numbers suppressed in line with the HES analysis guide.

This evaluation is the start of a process of ensuring all patients receive optimal care and treatment following a neck of femur fracture. Members of the study team previously conducted a patient involvement focus group with Bradford residents. The overwhelming message was that care should be tailored to meet individual needs. Patient and Public Involvement (PPI) and focus groups are intended to be consulted to determine the best way to reach patients from various backgrounds with the results of this evaluation. Focus groups consulted in the lead up to this project were surprised that hip fracture care is not more patient specific already. 'Ask the Researcher' events will be held to reach a wider public audience. The aim is to ensure that knowledge developed by this evaluation improves outcome following hip fracture.

Outputs are planned to be realised December 2023.

This evaluation is hoped to lay the foundations in ensuring health services are allocated appropriately and fairly to all accessing treatment in the NHS. The evaluation aims to assess the difference in outcome following neck of femur (NOF) fracture for patients from Black and Minority Ethnic (BAME) backgrounds, and inform National Hip Fracture Database (NHFD) policy on data collection. It intends to provide evidence for the requirement of ethnicity data collection, therefore allowing monitoring to ensure all patients from all backgrounds receive optimal care and treatment following a neck of femur fracture. Trusts will be able to assess the percentages of ethnic minority patients and allocate resources as appropriate.

Through publication in the NHFD Annual Report, Trusts will have easy access to the findings of this evaluation. Publishing in peer-reviewed journals is also anticipated to allow greater discussion of the strengths and weaknesses of the results and will provide the benefit of peer-review of the work from third parties. The University of Leeds will identify risk factors as a means of identifying future areas for investigation.

An example of a future area for investigation that may be generated from this evaluation is the identification of a variation in the incidence of different fracture types between ethnicities. If such a variation does exist, it is anticipated that it will be possible to identify morphological differences in the hip that may account for this. Currently, it is recommended all patients with a displaced intracapsular femoral neck fracture receive a cemented implant. However, morphological differences in the hip may mean certain ethnicities would benefit from a different type of implant. Therefore, patients from BAME backgrounds may be suffering from a higher rate of complications from the use of certain implants, which is likely to lead to costly revision surgery. If the data demonstrates such an effect, it will provide useful information for Trusts to allocate healthcare resources.

It is also known that patients from BAME backgrounds are more likely to suffer from deficiency of Vitamin D, as such bisphosphonate treatment may be less effective in these patients. If the data shows an unexpected effect of bisphosphonate therapy in BAME patients, this may enable future research studies to identify the causes for this which could in turn lead to a trialling of alternative therapies.

There have been significant delays to this project with data acquisition. Nevertheless preliminary data analysis shows a variation in survival following NOF by ethnicity. The study are liaising closely with the NHFD who await further analysis to determine further policy on ethnicity data collection. An extension is required to allow for complete data analysis up until publication.