National Hip Fracture Database

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Datasets: − Demographics; − MRIS - List Cleaning Report

Objective for processing

[3 paragraphs unchanged] The National Hip Fracture Database (NHFD) is a clinically led, web-based quality [5 words unchanged] /NHS England and managed by the Royal College of Physicians (RCP). The National Hip Fracture Database (NHFD) NHFD looks at the care of people who break with a hip bone. fracture or other femoral fracture. It collects data from hospital-based services in order to see how effective [53 words unchanged] the quality of that care and helps to improve the services provided. All 175 eligible hospitals in England, Wales and Northern Ireland are now regularly submitting submit data to the NHFD, which is the largest hip fracture database in [74 words unchanged] datasets, which provide further information about their care and outcomes of care. The datasets and the level of data requested are justified in detail in both this section and 'Processing Activities'. The geographical spread is required to be England and Wales since this [14 words unchanged] requested but Northern Ireland participates in and is part of the NHFD. The years requested are ongoing since the audit is run on a continuous basis; the agreement is requested to 2023 since that is the NCAPOP current contract end date for the audit; though it is hoped extension will be sought and approved. There are no alternative, less intrusive ways of achieving the purpose stated [7 words unchanged] minimum that is required to carry out the purposes of the audit. [1 paragraph unchanged] To ensure The that the Royal College of Physicians (RCP) can account for all reasons for a [15 words unchanged] died, particularly when this occurs unexpectedly during a planned sequence of episodes. [2 paragraphs unchanged] HQIP and NHS England both rely on the Article 6 (1) (e) 6(1)(e) legal basis under UK GDPR - "processing is necessary for the performance of a task carried [30 words unchanged] national bodies with statutory responsibilities to improve quality of health care services. The linkage requested is necessary for the performance of a task carried out in the public interest; improving the quality of care for people living with inflammatory arthritis. HQIP rely on Article 9(2)(i) as the legal basis for processing under UK GDPR - "processing is necessary for reasons of public interest in the [72 words unchanged] the quality and safety of care and to improve outcomes for patients. NHS England rely on Article 9(2)(h) of the UK GDPR as the legal basis for processing. "Processing is necessary for the [72 words unchanged] provision of health and social care, and management of systems and compliance. COVID 19 and CASE ASCERTAINMENT: The NHFD’s aim is to promote best practice in the care and secondary prevention of hip fracture and to this end it: The COVID-19 crisis has had different effects on different hospitals, but the most obvious is that a number of hospitals partially or completely ceased NHFD data entry for a period. In previous years the very high case ascertainment achieved across the NHFD have meant that (RCP) did not need to adjust their figures for incomplete data, or for potential biases in missing cases and missing data. To examine this issue, and to help in deciding whether the audit need to change the analysis across a range of metrics and key performance indicators (KPIs), the audit team need to compare submitted data with HES/PEDW data which should provide a more complete dataset for units in which NHFD data entry has been most significantly affected by staff sickness and redeployment. For instance, a couple of units briefly paused all NHFD data entry in March and April 2020. These units have since restarted data submissions, but have no record of how many cases they might have missed. An extract of HES/PEDW data based on all presentations with ICD-10 codes indicative of femoral fracture (S72) would allow us to model the total cases these units received over 2020, and so ensure that estimates of their KPIs and mortality are as accurate as possible, and that figures for the NHFD as a whole are not affected by these units temporary problems. RCP will combine NHFD and HES/PEDW datasets to profile the impact of COVID-19 on total numbers of people presenting with hip fracture, changes in the case-mix of these individuals, and changes in performance and mortality over the course of 2020. Inclusion of HES/PEDW data will allow us to do these even for those units which have been most affected by hip fracture staff sickness and redeployment — the same units in which NHFD data is most likely to be incomplete and in which care, performance, outcome and mortality may have been most affected. In previous years RCP have also used HES/PEDW data to capture additional care spells that follow the initial trust admission — so allowing us to calculate “superspell”. This measure will be particularly important given the COVID-19 crisis, since it is the only way to fully capture the true bed-occupancy associated with care and recovery from hip fracture, and for us to aid NHS managers in modelling the cost and resource implications of a condition that carries and annual cost of over £1 billion . The additional COVID codes will allow for case-mix adjusted outcome assessment facilitating all of the above and in particular, case-mix adjusted outcomes for the mortality data for the outlier analyses. This information would help inform our model for case-mix adjustment. COVID-19 is the dominant factor affecting all hip fracture services around the country. Hospitals can already log this information themselves, but it is not easy for them to examine how COVID status affects the provision and outcome of care. The additional codes will allow the NHFD to monitor and with HQIP approval to publish trends in the prevalence of infection - among inpatients - among people presenting from the community, and - among people from care homes; the most challenging and least understood setting in the current pandemic. The aim is to promote best practice in the care and secondary prevention of hip fracture. Building on the experience of a range of hip fracture audits it: [1 paragraph unchanged] • Allows casemix-adjusted Casemix-adjusts outcome assessment to promote transparency of inter-hospital comparisons • Enables Allows hospitals to compare care and outcomes against with national benchmarks and quality standards [3 paragraphs unchanged] • To support Supports audit, service evaluation and quality improvement activity by NHFD, NHFD activity: its own, those of participant sites and those of approved third parties (NB identifiable data will not be released to third [5 words unchanged] will only be released subject to appropriate data sharing agreements being agreed. • In addition, the NHFD has the potential to provide in future a robust framework for future large-scale formal clinical research projects, projects; organised and funded as such, and such; subject – in (in contrast to the above observational studies – studies) to research approval via the National Research Ethics Committees process. National Data Opt-Outs (NDOs) have been historically applied to the data disseminated under this agreement following support from the Confidentiality Advisory Group (CAG). The National Data Opt-Out (NDO) enables patients to opt-out from the use of their confidential patient information for research and planning purposes where the data flows rely upon Regulation 5 of the Health Service COPI (Control of Patient Information) Regulations 2002. It is a standard condition of support under Regulation 5 of the COPI Regulations 2002 that patient wishes are respected. In line with the National Data Opt-Out Operational Policy the Confidentiality Advisory Group (CAG) may exceptionally advise the decision-maker that the NDO should not apply to a specific data flow supported under Regulation 5 of the COPI Regulations 2002. In the case of the NCAPOP, this has been supported. The justification to not apply the NDOs is as below: 1. Deferral rationale: patient safety The NHFD is responsible for the identification and management of hospitals which are 'outliers’ for casemix-adjusted mortality 30-days after hip fracture. This process depends on the completeness of data from each hospital. Any casemix model will be sensitive to incomplete data, and geographical variation in the impact of the NDO means that hospitals in some areas will appear to perform better or less well, simply because of the extent of missing data that will arise with the application of the NDO. Some hospitals will therefore be falsely reassured of the quality of care they are providing, whereas patients and staff in other hospitals may be misidentified as a concern for the same reason. 2. Deferral rationale: Introduction of bias Excluding patients that have registered against the NDO will introduce a biased sampling frame due to non-random opt-out patterns. The data opt out figures from NHS England show that 50 to 70 year olds (~6%) opt out at a higher rate than the national average (5.4%). 3. Deferral rationale: technical impacts Applying the NDO would generate additional workload for hospital teams, which could lead to disengagement across the audits, either through delayed entry, reduced entry or complete disengagement from data entry due to the increased burden. This would ultimately impact the programme’s ability to deliver its remit in effectively measuring and providing high quality data across patient safety of an ever increasing, vulnerable population. COVID 19 and CASE ASCERTAINMENT: The COVID-19 crisis has had different effects on different hospitals, but the most obvious is that a number of hospitals partially or completely ceased NHFD data entry for a period. In previous years the very high case ascertainment achieved across the NHFD have meant that RCP did not need to adjust their figures for incomplete data, or for potential biases in missing cases and missing data. To examine this issue, and to help in deciding whether the audit need to change the analysis across a range of metrics and key performance indicators (KPIs), the audit team need to compare submitted data with HES/PEDW (Welsh HES) data which should provide a more complete dataset for units in which NHFD data entry has been most significantly affected by staff sickness and redeployment. For instance, a couple of units briefly paused all NHFD data entry in March and April 2020. These units have since restarted data submissions, but have no record of how many cases they might have missed. An extract of HES/PEDW data based on all presentations with ICD-10 codes indicative of femoral fracture (S72) would allow the audit to model the total cases these units received over 2020, and so ensure that estimates of their KPIs and mortality are as accurate as possible, and that figures for the NHFD as a whole are not affected by these units temporary problems. RCP will combine NHFD and HES/PEDW datasets to profile the impact of COVID-19 on total numbers of people presenting with hip fracture, changes in the case-mix of these individuals, and changes in performance and mortality over the course of 2020. Inclusion of HES/PEDW data will allow RCP to do these even for those units which have been most affected by hip fracture staff sickness and redeployment — the same units in which NHFD data is most likely to be incomplete and in which care, performance, outcome and mortality may have been most affected. In previous years RCP have also used HES/PEDW data to capture additional care spells that follow the initial trust admission — so allowing calculation of the “superspell”. This measure is particularly important given the COVID-19 crisis, since it is the only way to fully capture the true bed-occupancy associated with care and recovery from hip fracture, and for RCP to aid NHS managers in modelling the cost and resource implications of a condition that carries and annual cost of over £1 billion . The additional COVID codes will allow for case-mix adjusted outcome assessment facilitating all of the above and in particular, case-mix adjusted outcomes for the mortality data for the outlier analyses. This information would help inform our model for case-mix adjustment. COVID-19 has been the dominant factor affecting all hip fracture services around the country. Hospitals can already log this information themselves, but it is not easy for them to examine how COVID status affects the provision and outcome of care. The additional codes will allow the NHFD to monitor and with HQIP approval to publish trends in the prevalence of infection - among inpatients - among people presenting from the community, and - among people from care homes

Processing activities

NHS Digital implemented a change in process in June 2021 where they performed a risk assessment on the situation of having two distinct data recipients for two distinct data flows under one agreement. The process going forward was agreed that this would not be possible, as the risk of data breaches is deemed high. The decision was that the original agreement needed to be split in two. One to send the mortality data to Crown Informatics, under DARS-NIC-468622-L9V82. DARS-NIC-10343-Z3M1B will then supply the HES data for the audit team to work on. Two distinct data flows to different data recipients are required for the National Hip Fracture Database (NHFD): As of 1 April 2021 the processing and analysis of the data will be completed by the University of Bristol by an individual who was previously based at the University of Oxford, both locations have been kept in the Data Sharing Agreement to facilitate the process, during the migration period. • DARS-NIC-468622-L9V82 allows mortality data to flow to Crown Informatics The data currently held by the University of Oxford will be transferred to the University of Bristol using a safe and secure data transfer method. In Oxford, data is held in the data safe room, on a hard drive, where the data on the hard drive is contained in an encrypted veracrypt folder. The data will be securely uploaded directly to an encrypted folder to the secure server on the Safe Haven filestore at Bristol, by using a remote desktop and virtual private network. • DARS-NIC-10343-Z3M1B will then supply the HES data to the audit’s statisticians at the University of Bristol. Under DARS-NIC-10343-Z3M1B, Crown Informatics Ltd send NHS Number, Date of Birth, First [9 words unchanged] study ID for the Falls and Fragility Fracture Audit Programme) to NHS Digital England on a quarterly basis. Crown Informatics will provide the identifiers of ~110,000 individuals initially, with subsequent updates of ~22,000 individuals each quarter. NHS Digital England link the data and provide Crown Informatics Ltd with: [2 paragraphs unchanged] NHS Digital England link the data and provide Bristol NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) with o • HES non sensitive data for cohort with FFFAP ID (Linked Cohort) at the end of quarter one every year (under Cohort under DARS-NIC-10343-Z3M1B) o • HES non sensitive data for falls patients who are not in cohort but have a diagnosis of a fall or fracture (Case (COVID-19 Case Ascertainment Cohort) at the end of quarter one every year in 2019/20 and 2020/21 only (under DARS-NIC-10343-Z3M1B) Crown Informatics receive only the Deaths data from NHS Digital England on a quarterly basis and combine with FFFAP data. [2 paragraphs unchanged] These other linkages will be performed separately to the NHS Digital England linkage. The linkages to central NHS held data will only involve the [8 words unchanged] nations, so for example there will be no data transferred to NHS Digital England for patients recruited in Welsh institution. [3 paragraphs unchanged] 3. To calculate superspell figures for hospitals and NHS trusts from the last two available financial years and compare the average lengths of stays with superspell figures in the current year to assess whether there has been a reduction in acute and/or post-acute lengths of hospital stay. 3. Use HES to evaluate the benefit of long-term outcome measures such as 6-month/1-year survival and readmission profiles by using the ability of HES to track patients and describe their patterns of care after an incident hip fracture. 4. Use HES to evaluate the benefit of long-term outcome measures such as 6-month/1-year survival and readmission profiles by using the ability of HES to track patients and describe their patterns of care after an incident hip fracture. 4. To validate and refine casemix risk adjustment models for the audit. 5. To validate and refine casemix risk adjustment models for the audit. 5. To examine life status at 30 days for statistical analysis. 6. To examine life status at 30 days for statistical analysis. Crown Informatics makes life-status 30 days available to trusts for local audit purposes – but only of patients treated at that trust. Life-status for patients not treated at a trust will not be made available to a trust. For statistical purposes, such as monitoring trends, registered individuals within a trust can access date of death for that trust’s patients which they submit to the audit derived from NHS England's mortality data. Charts and tables are also provided using 30 day survival. These data flows are included and approved as part of the decision to support this use of confidential data by the HRA Confidentiality Advisory Group (under s251 of the NHS Act 2006). Crown Informatics make life status 30 days - only of patients treated at that trust - available to trusts, for local audit purposes; life status for patients not treated at a trust will not be made available to a trust. For statistical purposes, such as monitoring trends, registered individuals at Trusts can access date of death for that trust’s patients which they submit to the audit derived from the NHS Digital's mortality data. Charts and tables are also provided using 30 day survival. These data flows are included and approved as part of the decision to support this use of confidential data by the HRA Confidentiality Advisory Group (under s251 of the NHS Act 2006). Bristol NIHR Biomedical Research Centre Musculoskeletal Research Unit statisticians are commissioned to perform complex statistical analysis for the purposes of the national clinical audit reporting. To this end they require HES linked data (for case ascertainment) and date of death from NHS England's mortality data (for casemix adjusted mortality analysis). RCP analysts perform all further ad hoc analysis to inform the conduct of the audit programme and the writing of annual reports – for the most part this takes place on NHFD data alone, but in some circumstances date of death from NHS England's mortality data is required to calculate patient outcome. Bristol NIHR Biomedical Research Centre Musculoskeletal Research Unit statisticians are commissioned to perform complex statistical analysis for the purposes of the national clinical audit reporting. To this end they require HES linked data (for case ascertainment, superspell length of stay) and date of death from NHS Digital's mortality data (for casemix adjusted mortality analysis). RCP analysts perform all further ad hoc analysis to inform the conduct of the audit programme and the writing of annual reports – for the most part this takes place on NHFD data alone, but in some circumstances date of death from NHS Digital's mortality data is required to calculate patient outcome. Patient identifiers are not sent to RCP or Bristol NIHR Biomedical Research Centre Musculoskeletal Research Unit from Crown Informatics. Only pseudonymised data is sent to Bristol NIHR Biomedical Research Centre Musculoskeletal Research Unit. HES data will flow directly to Bristol NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) from NHS England. Patient identifiers are not sent to RCP or Bristol NIHR Biomedical Research Centre Musculoskeletal Research Unit from Crown Informatics. Only pseudonymised data is sent to Bristol NIHR Biomedical Research Centre Musculoskeletal Research Unit. HES data will flow directly to Bristol NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) from NHS Digital. [1 paragraph unchanged] All individuals with access to record level data are employed by the Royal College of Physicians, University of Bristol or University of Oxford. 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). All individuals with access to record level data are employed by the Royal College of Physicians or University of Bristol. [2 paragraphs unchanged] Ethnicity data A recent study based on the NHFD (Sheehan et al. 2020) demonstrated different patient groups to experience very different care and outcome. Unsurprisingly older people do less well, but so do men. Older people from socioeconomically deprived patients receive different types of surgery from those from less deprived backgrounds, and a black patient admitted from their own home recovers as poorly after such surgery as does a white patient admitted from a care home. All of the national clinical audits have been tasked with examining and challenging such inequalities. However, the complex inter-relatedness of socioeconomic deprivation, ethnicity, age and sex mean that analysis of the potential sources of inequality is only possible if patients’ ethnicity is known. It is clearly inappropriate, and potentially offensive, for patients to be questioned about their ethnicity each time they present to the health services involved in one of the national clinical audits. Such audits therefore need to make use of the ethnicity data that is already recorded in HES.

Expected output

Publications and outputs using the data: The expected outputs of processing will be: NHFD annual report 2016, 2017, 2018, 2019, 2020, 2021, 2022 • The annual National Hip Fracture Database report, available at https://www.nhfd.co.uk NHFD commissioners report 2016, 2017 • Quarterly updates to charts, key performance indicators, benchmark summaries and dashboards on https://www.nhfd.co.uk NHFD patient outputs • Quarterly updates to 30-day mortality charts which flags any hospitals which are outliers https://www.nhfd.co.uk • NHFD supplementary reports Further results and methods used to derive case-ascertainment, casemix adjusted mortality and HES super-spells have been published by the RCP in the NHFD annual report in September 2017 and were further published in November 2018. The results of the other analyses will be published in other NHFD supplementary reports and /or peer-reviewed articles. • Peer-reviewed articles The RCP previously published CCG and provider level outcomes for the Best Practice Tariff and to support the NHS Outcomes Framework and CCG Outcome Indicators Set. The outputs will not contain NHS England data and will only contain aggregated information with small numbers suppressed as appropriate in line with the relevant disclosure rules for the dataset(s) from which the information was derived. Crown Informatics make date of death & 30 day mortality flag to hospitals for statistical purposes such as monitoring trends for patients they submit to the audit. This allows registered Trust users of the system to be able to identify patients who have died within 30 days of hip fracture (including those who die after hospital discharge) and follow standard clinical governance processes to identify expected vs unexpected deaths, perform root cause analysis and assure whether good practices of care are followed. Sites can only access data for their own patients and these data flows are included and approved as part of the decision to support this use of confidential data by the HRA Confidentiality Advisory Group (under s251 of the NHS Act 2006). The outputs will be communicated to relevant recipients through the following dissemination channels: • Webinars open to the public – the latest webinar recording can be found at: https://www.nhfd.co.uk/20/hipfractureR.nsf/docs/2022Report#webinar • Social media • Public reports • Journals • Presentations/ conferences • Industry newsletters • Press/media engagement The NHFD will hold quarterly webinars advertising findings from the data and publishing quarterly newsletters signposting participants to check their data.

Expected measurable benefits

[4 paragraphs unchanged] • performance Performance evaluation to support Monitor’s the Best Practice Tariff (BPT) [6 paragraphs unchanged] The data requested in this application agreement will support: [4 paragraphs unchanged] What is hoped to be achieved with the HES and mortality data requested: • An investigation into the relationship between hip fracture care and health inequalities. What is hoped to be achieved with the HES requested: [2 paragraphs unchanged] • Continued production and quarterly updating of performance run charts of mortality to allow sites to monitor trends in patient outcome • Continued productions of annual quarterly summary of casemix adjusted mortality to detect and support outlying sites • Refinement of methodologies to detect superspell length of stay and detected assess persistent uncertainties in patient pathways for patients sent to rehabilitation units. • Investigation into the relationship between hip fracture care and health inequalities. Codes identifying COVID status will allow the NHFD to monitor and with HQIP approval to publish trends in the prevalence of infection Specific benefits to patients should be improved care and improved mortality outcomes. - among inpatients Quarterly updates to charts, benchmarks and mortality data will enable HQIP/ NHSE/ RCP/ individual hospitals to implement changes based on the data available. Additionally, publication of annual reports with national recommendations will inform hospitals and commissioners of any changes that should be made to improve care. - among people presenting from the community, and - among people from care homes; the most challenging and least understood setting in the current pandemic. In particular, the additional COVID codes will allow for case-mix adjusted outcomes for the mortality data for the outlier analyses. This information would help inform the audit teams model for case-mix adjustment. COVID-19 is the dominant factor affecting all hip fracture services around the country. Hospitals can already log this information themselves, but it is not easy for them to examine how COVID status affects the provision and outcome of care.

Benefits reported

There have been no further yielded benefits realised since version 6. [4 paragraphs unchanged] • performance Performance evaluation to support Monitor’s the Best Practice Tariff (BPT) [6 paragraphs unchanged] The data requested in these applications under this Agreement so far have has supported: [7 paragraphs unchanged] • Productions of annual summarys summaries of casemix adjusted mortality to detect and support outlying sites • Calculated superspell length of stay and detected uncertainties in patient pathways for patients sent to rehabilitation units. Hip fracture is the commonest serious injury in older people and the commonest reason for them to need emergency surgery and anaesthesia. The people who typically sustain hip fracture have multiple medical, psychological and social problems. This frailty means that they must receive a coordinated programme of care, with collaboration between surgeons, anaesthetists, orthogeriatricians, nurses and therapists in assessment and rehabilitation if they are to regain their previous independence and to return home. The NHFD seeks to ensure that hospitals provide the efficient/affordable coordinated multidisciplinary care of the Hip Fracture Programme recommended by the National Institute of Clinical Excellence (NICE), so that all patients receive: • comprehensive geriatric assessment when they present to a hospital • prompt and effective forms of anaesthesia and surgery • individualised care that minimises their risk of delirium • rehabilitation to help them return to their previous home. NICE estimated that such Hip Fracture Programme care could lead to a financial saving of £5,000 per patient, as a result of improved length of hospital stay and a reduced requirement for long term residential and domiciliary care. An independent evaluation has confirmed that since its establishment the NHFD has succeeded in reducing deaths in the month after hip fracture (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4501693/).

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Objective for processing

The National Clinical Audit and Patient Outcomes Programme (NCAPOP) is a large programme of circa 35 projects consisting of National Clinical Audits and Confidential Enquiries. HQIP The Healthcare Quality Improvement Partnership (HQIP) is commissioned by NHS England to commission and manage the NCAPOP. NHS [9 words unchanged] HQIP as together both organisations determine the purposes and means of processing. [2 paragraphs unchanged] The National Hip Fracture Database (NHFD) is a clinically led, web-based quality improvement initiative commissioned by the Healthcare Quality Improvement Partnership (HQIP)/NHS HQIP /NHS England and managed by the Royal College of Physicians (RCP). The National [80 words unchanged] the quality of that care and helps to improve the services provided. [1 paragraph unchanged] There are no moral or ethical issues raised by the proposed dissemination. There is no risk of potential harm to the public by the dissemination. The datasets and the level of data requested are justified in detail in both this section and 'Processing Activities'. The geographical spread is required to be England and Wales since this is the geographical coverage of the NHFD. Data for Northern Ireland is not being requested but Northern Ireland participates in and is part of the NHFD. The years requested are ongoing since the audit is run on a continuous basis; the agreement is requested to 2023 since that is the NCAPOP current contract end date for the audit; though it is hoped extension will be sought and approved. There are no alternative, less intrusive ways of achieving the purpose stated here and the data requested is the minimum that is required to carry out the purposes of the audit. The datasets and the level of data requested are justified in detail in both this section and section 5b. The geographical spread is required to be England and Wales since this is the geographical coverage of the NHFD. Data for Northern Ireland is not being requested but Northern Ireland participates in and is part of the NHFD. The years requested are ongoing since the audit is run on a continuous basis; the agreement is requested to 2023 since that is the NCAPOP current contract end date for the audit; though it is hoped extension will be sought and approved. There are no alternative, less intrusive ways of achieving the purpose stated here and the data requested is the minimum that is required to carry out the purposes of the audit. In addition to the information collected by the audit, Hospital Episode Statistics (HES) data is linked to the audit data to provide a richer data set. HES data is used in this audit to look at patient pathways for people who have experienced a hip fracture. The audit will look at HES episode data for patients and link them together into a ‘super-spell’: this is the whole period of hospitalisation for the patient regardless of which consultant they are under and crossing hospitals where a transfer has taken place. In addition to the information collected by the audit, Hospital Episode Statistics (HES) data is linked to the audit data to provide a richer data set. HES data is used in this audit to look at patient pathways for people who have experienced a hip fracture. The audit will look at HES episode data for patients and link them together into a ‘super-spell’: this is the whole period of hospitalisation for the patient regardless of which consultant they are under and crossing hospitals where a transfer has taken place List cleaning is used to find the correct NHS Number and the primary purpose is also to obtain date of death for those on the list. It is used to ensure the quality of the identifiers, prior to linkage to HES. Date of latest posting is provided which includes fact of death, date of death, exits, cancellations and embarkations. [4 paragraphs unchanged] HQIP rely on Article 9 (2) (i) 9(2)(i) as the legal basis for processing under GDPR - "processing is necessary [79 words unchanged] the quality and safety of care and to improve outcomes for patients. [3 paragraphs unchanged] To examine this issue, and to help in deciding whether the audit need to change the analysis across a range of metrics and KPIs, key performance indicators (KPIs), the audit team need to compare submitted data with HES/PEDW data which [11 words unchanged] data entry has been most significantly affected by staff sickness and redeployment. [1 paragraph unchanged] (RCP) RCP will combine NHFD and HES/PEDW datasets to profile the impact of COVID-19 [68 words unchanged] in which care, performance, outcome and mortality may have been most affected. In previous years (RCP) RCP have also used HES/PEDW data to capture additional care spells that follow [56 words unchanged] a condition that carries and annual cost of over £1 billion . [13 paragraphs unchanged] • In addition, has the potential to provide in future a robust [16 words unchanged] contrast to the above observational studies – to research approval via the NREC National Research Ethics Committees process.

Processing activities

NHS Digital implemented a change in process in June 2021 where they [47 words unchanged] original agreement needed to be split in two. One to send the demographic files mortality data to Crown Informatics, under DARS-NIC-468622-L9V82. This agreement DARS-NIC-10343-Z3M1B will then supply the HES data for the audit team to work on. 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 i.e: employees, agents and contractors of the Data Recipient who may have access to that data). [2 paragraphs unchanged] Under DARS-NIC-10343-Z3M1B, Crown Informatics Ltd send NHS Number, Date of Birth, First Name, Surname, Gender, Full postcode and FFFAP ID (a study ID for the Falls and Fragility Fracture Audit Programme) to NHS Digital. Digital on a quarterly basis. Crown Informatics will provide the identifiers of ~110,000 initially, with subsequent updates of ~22,000 each quarter. NHS Digital link the data and provide Crown Informatics Ltd with: • Demographics File (validated identifiers) with FFFAP ID • Deaths File (validated identifiers) with FFFAP ID and NHS number (under DARS-NIC-468622-L9V82) - NHS number is requested to be returned in addition to study ID to help confirm process validity and ensure the correct events are allocated to the correct patient • Date of latest posting with FFFAP ID (historic data request - no longer supplied) NHS Digital link the data and provide Bristol NHIR NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) with o HES non sensitive data for cohort with FFFAP ID. ID (Linked Cohort) at the end of quarter one every year (under DARS-NIC-10343-Z3M1B) o HES non sensitive data for falls patients who are not in cohort but have a diagnosis of a fall or fracture (Case Ascertainment Cohort) at the end of quarter one every year (under DARS-NIC-10343-Z3M1B) Crown Informatics receive only the Demographic Deaths data from NHS Digital on a quarterly basis and combine with FFFAP data. Name, Date of Birth, and postcode, which are included in the cleaned identifiers returned from list cleaning (the demographic file) will be destroyed after use. [1 paragraph unchanged] Crown Informatics send FFFAP ID and date of death data to Bristol NHIR NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) Unit. [1 paragraph unchanged] Bristol NHIR NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) and The Royal College of Physicians (RCP) analyse data for audit purposes: [5 paragraphs unchanged] 6. To examine life status at 30 days for statistical analysis. Case- mix adjustment performed by NDORMS. Crown Informatics make life status 30 days - only of patients treated [45 words unchanged] that trust’s patients which they submit to the audit derived from the list cleaning NHS Digital's mortality data. Charts and tables are also provided using 30 day survival. These [18 words unchanged] by the HRA Confidentiality Advisory Group (under s251 of the NHS Act 2006) this dissemination will be done under DARS-NIC- 468622-L9V82 as explained previously. 2006). Bristol NHIR NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) statisticians are commissioned to perform complex statistical analysis for the purposes of [13 words unchanged] (for case ascertainment, superspell length of stay) and date of death from the list cleaning NHS Digital's mortality data (for casemix adjusted mortality analysis). RCP analysts perform all further ad [24 words unchanged] on NHFD data alone, but in some circumstances date of death from the list cleaning NHS Digital's mortality data is required to calculate patient outcome. Crown Informatics, the RCP and Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) will all have access to patient level HES and date of latest posting data. Patient identifiers are not sent to RCP or Bristol NHIR NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) from Crown Informatics. Only pseudonymised data is sent to Bristol NHIR NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol). Unit. HES data will flow directly to Bristol NHIR NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) from NHS Digital. The RCP commission Bristol NHIR NIHR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) to perform key statistical analysis so in effect this is a collaboration since the RCP will advise on clinical elements of the methodology and will interpret the findings. All individuals with access to record level data are employed by Crown Informatics, the Royal College of Physicians, University of Bristol or University of Oxford. All [38 words unchanged] contractors of the Data Recipient who may have access to that data). [2 paragraphs unchanged] A one-off data flow is planned, consisting of Mortality data only for 2011-2016 inclusive, to Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) This is required for the development of live casemix-adjusted mortality run-charts, which will allow sites to monitor their mortality rates much more accurately and in real-time. This work is still ongoing.

Expected output

[1 paragraph unchanged] NHFD annual report 2016, 2017, 2018, 2019, 2020, 2021 2021, 2022 [1 paragraph unchanged] NHFD patient report 2018 outputs [2 paragraphs unchanged] The RCP publish previously published CCG and provider level outcomes for the Best Practice Tariff and to support the NHS Outcomes Framework and CCG Outcome Indicators Set. The RCP publish provider level outcomes for the Best Practice Tariff. Crown Informatics make date of death & 30 day mortality flag to hospitals for statistical purposes such as monitoring trends for patients they submit to the audit. This allows registered Trust users of the system to be able to identify patients who have died within 30 days of hip fracture (including those who die after hospital discharge) and follow standard clinical governance processes to identify expected vs unexpected deaths, perform root cause analysis and assure whether good practices of care are followed. Sites can only access data for their own patients and these data flows are included and approved as part of the decision to support this use of confidential data by the HRA Confidentiality Advisory Group (under s251 of the NHS Act 2006). Crown Informatics make date of death & 30 day mortality flag to hospitals for statistical purposes such as monitoring trends (calculated from date of latest posting) for patients they submit to the audit. This allows registered Trust users of the system to be able to identify patients who have died within 30 days of hip fracture (including those who die after hospital discharge) and follow standard clinical governance processes to identify expected vs unexpected deaths, perform root cause analysis and assure whether good practices of care are followed. Sites can only access data for their own patients and these data flows are included and approved as part of the decision to support this use of confidential data by the HRA Confidentiality Advisory Group (under s251 of the NHS Act 2006).

Expected measurable benefits

[16 paragraphs unchanged] What is hoped to be achieved with the HES and list cleaning mortality data requested: [11 paragraphs unchanged]

Benefits reported

There have been no further yielded benefits realised since the previous version (v6.2). 6. [21 paragraphs unchanged]

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Data controllers: + NHS ENGLAND

Objective for processing

The National Hip Fracture Database (NHFD) is a clinically led, web-based quality improvement initiative commissioned by the Healthcare Quality Improvement Partnership (HQIP) and managed by the Royal College of Physicians of London (RCP). The National Hip Fracture Database (NHFD) looks at the care of people who break a hip bone. It collects data from hospital-based services in order to see how effective services are at providing appropriate treatment to patients across England, Wales and Northern Ireland. In some areas they are also collecting information from community-based NHS services where these are responsible for following up patients after discharge from hospital. Information gathered about care in hospital and about recovery afterwards enables the RCP to measure the quality of that care and helps to improve the services provided. The National Clinical Audit and Patient Outcomes Programme (NCAPOP) is a large programme of circa 35 projects consisting of National Clinical Audits and Confidential Enquiries. 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. The National Hip Fracture Database (NHFD) is a clinically led, web-based quality improvement initiative commissioned by the Healthcare Quality Improvement Partnership (HQIP)/NHS England and managed by the Royal College of Physicians (RCP). The National Hip Fracture Database (NHFD) looks at the care of people who break a hip bone. It collects data from hospital-based services in order to see how effective services are at providing appropriate treatment to patients across England, Wales and Northern Ireland. In some areas they are also collecting information from community-based NHS services where these are responsible for following up patients after discharge from hospital. Information gathered about care in hospital and about recovery afterwards enables the RCP to measure the quality of that care and helps to improve the services provided. [2 paragraphs unchanged] The datasets and the level of data requested are justified in detail [52 words unchanged] audit is run on a continuous basis; the agreement is requested to 2021 2023 since that is the NCAPOP current contract end date for the audit; [29 words unchanged] minimum that is required to carry out the purposes of the audit. [4 paragraphs unchanged] GDRP Legal Basis: Legal Basis Justification: The lawful basis for processing data under GDPR has been reviewed against the guidance provided by IGARD and has been assessed as acceptable. The details are: 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. The linkage requested is necessary for the performance of a task carried out in the public interest; improving the quality of care for people living with inflammatory arthritis. HQIP have provided the following justification for choosing Article 6(1)(e) and Article 9(2)(i): 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 has devolved responsibility (through our HQIP contract with them) for managing the national audits and other programmes we commission on their behalf. This includes managing data access applications, signing data sharing agreements (we are able to do this as we are still a data controller with direct contractual responsibility over the clinical audits, with the organisations managing them acting as our data processors) and exercising due diligence (for IG and data protection) over the commissioned programme. The national clinical audits are publicly funded programmes with dependencies upon the data produced to improve healthcare, monitor safety and inform CQC inspections (plus other dependencies), therefore implementing the right to erasure would have an impact on these functions. 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. Article 6 (1)(e) - "processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller". NHS England are a public authority. NHS England contract HQIP to commission the NCAPOP and the NHS standard contract with NHS providers of services to participate within the National Clinical Audit and Patient Outcomes Programme (NCAPOP). The Secretary of State has a legal duty under section 2 of the Health and Social Care Act 2012 to continuously improve the quality of health care services. NHS England has a legal duty under the NHS Act 2006 section 13e to improve quality of healthcare services. The Secretary of State and NHS England exercise these duties through the NHS England contract with HQIP to commission the National Clinical Audit and Patient Outcomes Programme (NCAPOP), and the NHS standard contract with NHS providers of services to mandate participation within the NCAPOP (to improve the quality of patient care and outcomes). HQIP are therefore performing a public task. COVID 19 and CASE ASCERTAINMENT: Additional information provided by the Associate Director, Research and Governance, National Joint Registry (NJR) HQIP: The COVID-19 crisis has had different effects on different hospitals, but the most obvious is that a number of hospitals partially or completely ceased NHFD data entry for a period. In previous years the very high case ascertainment achieved across the NHFD have meant that (RCP) did not need to adjust their figures for incomplete data, or for potential biases in missing cases and missing data. The Secretary of State has a legal duty under section 2 of the Health and Social Care Act 2012 to continuously improve the quality of health care services. NHS England has a legal duty under the NHS Act 2006 section 13e to improve quality of healthcare services. The secretary and NHS England exercise these duties through the appointment of an NHS England expert committee accountable to the national medical director to oversee the National Joint Registry; and the NHS standard contract with NHS providers of services to mandate participation in the NJR (to improve the quality of patient care and outcomes). We are therefore performing a public task. To examine this issue, and to help in deciding whether the audit need to change the analysis across a range of metrics and KPIs, the audit team need to compare submitted data with HES/PEDW data which should provide a more complete dataset for units in which NHFD data entry has been most significantly affected by staff sickness and redeployment. For instance, a couple of units briefly paused all NHFD data entry in March and April 2020. These units have since restarted data submissions, but have no record of how many cases they might have missed. An extract of HES/PEDW data based on all presentations with ICD-10 codes indicative of femoral fracture (S72) would allow us to model the total cases these units received over 2020, and so ensure that estimates of their KPIs and mortality are as accurate as possible, and that figures for the NHFD as a whole are not affected by these units temporary problems. (RCP) will combine NHFD and HES/PEDW datasets to profile the impact of COVID-19 on total numbers of people presenting with hip fracture, changes in the case-mix of these individuals, and changes in performance and mortality over the course of 2020. Inclusion of HES/PEDW data will allow us to do these even for those units which have been most affected by hip fracture staff sickness and redeployment — the same units in which NHFD data is most likely to be incomplete and in which care, performance, outcome and mortality may have been most affected. In previous years (RCP) have also used HES/PEDW data to capture additional care spells that follow the initial trust admission — so allowing us to calculate “superspell”. This measure will be particularly important given the COVID-19 crisis, since it is the only way to fully capture the true bed-occupancy associated with care and recovery from hip fracture, and for us to aid NHS managers in modelling the cost and resource implications of a condition that carries and annual cost of over £1 billion . The additional COVID codes will allow for case-mix adjusted outcome assessment facilitating all of the above and in particular, case-mix adjusted outcomes for the mortality data for the outlier analyses. This information would help inform our model for case-mix adjustment. COVID-19 is the dominant factor affecting all hip fracture services around the country. Hospitals can already log this information themselves, but it is not easy for them to examine how COVID status affects the provision and outcome of care. The additional codes will allow the NHFD to monitor and with HQIP approval to publish trends in the prevalence of infection - among inpatients - among people presenting from the community, and - among people from care homes; the most challenging and least understood setting in the current pandemic. The aim is to promote best practice in the care and secondary prevention of hip fracture. Building on the experience of a range of hip fracture audits it: • Collects data on patient casemix, care, outcomes and secondary prevention • Allows casemix-adjusted outcome assessment to promote transparency of inter-hospital comparisons • Enables hospitals to compare care and outcomes against national benchmarks and quality standards • Monitors performance over time • Measures the impact of changes in clinical care and service organisation • Has the capacity to support valuable observational studies on aspects of hip fracture care, through the use of routinely collected data and specifically designed sprint audits - together with casemix-adjusted outcomes. • To support audit, service evaluation and quality improvement activity by NHFD, NHFD participant sites and approved third parties (NB identifiable data will not be released to third parties and patient level data will only be released subject to appropriate data sharing agreements being agreed. • In addition, has the potential to provide in future a robust framework for large-scale formal clinical research projects, organised and funded as such, and subject – in contrast to the above observational studies – to research approval via the NREC process.

Processing activities

NHS Digital implemented a change in process in June 2021 where they performed a risk assessment on the situation of having two distinct data recipients for two distinct data flows under one agreement. The process going forward was agreed that this would not be possible, as the risk of data breaches is deemed high. The decision was that the original agreement needed to be split in two. One to send the demographic files to Crown Informatics, under DARS-NIC-468622-L9V82. This agreement will then supply the HES data for the audit team to work on. 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 i.e: employees, agents and contractors of the Data Recipient who may have access to that data). As of 1 April 2021 the processing and analysis of the data will be completed by the University of Bristol by an individual who was previously based at the University of Oxford, both locations have been kept in the Data Sharing Agreement to facilitate the process, during the migration period. The data currently held by the University of Oxford will be transferred to the University of Bristol using a safe and secure data transfer method. In Oxford, data is held in the data safe room, on a hard drive, where the data on the hard drive is contained in an encrypted veracrypt folder. The data will be securely uploaded directly to an encrypted folder to the secure server on the Safe Haven filestore at Bristol, by using a remote desktop and virtual private network. [2 paragraphs unchanged] • List cleaning file Demographics File (validated identifiers) with FFFAP ID [1 paragraph unchanged] NHS Digital link the data and provide Nuffield Department of Orthopaedic, Rheumatology & Bristol NHIR Biomedical Research Centre Musculoskeletal Science (NDORMS, part Research Unit (part of the University of Oxford) Bristol) with HES non sensitive data for cohort with FFFAP ID. Crown Informatics receive only the MRIS List Cleaning data from NHS Digital and combine with FFFAP data. o HES non sensitive data for cohort with FFFAP ID. (Linked Cohort) Name, Date of Birth, and postcode, which are included in the cleaned identifiers returned from list cleaning will be destroyed after use. o HES non sensitive data for falls patients who are not in cohort but have a diagnosis of a fall or fracture (Case Ascertainment Cohort) Crown Informatics receive only the Demographic data from NHS Digital and combine with FFFAP data. Name, Date of Birth, and postcode, which are included in the cleaned identifiers returned from list cleaning (the demographic file) will be destroyed after use. [1 paragraph unchanged] Crown Informatics send FFFAP ID and date of death data to NDORMS Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Oxford). Bristol) [1 paragraph unchanged] NDORMS Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Oxford) Bristol) and The Royal College of Physicians of London analyse data for audit purposes: [6 paragraphs unchanged] Crown Informatics make life status 30 days - only of patients treated [89 words unchanged] by the HRA Confidentiality Advisory Group (under s251 of the NHS Act 2006). 2006) this dissemination will be done under DARS-NIC- 468622-L9V82 as explained previously. NDORMS Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Oxford) Bristol) statisticians are commissioned to perform complex statistical analysis for the purposes of [74 words unchanged] death from the list cleaning data is required to calculate patient outcome. Crown Informatics, the RCP and NDORMS Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Oxford) Bristol) will all have access to patient level HES and date of latest posting data. Patient identifiers are not sent to RCP or NDORMS Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Oxford) Bristol) from Crown Informatics. Only pseudonymised data is sent to NDORMS. Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol). HES data will flow directly to NDORMS Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Oxford) Bristol) from NHS Digital. The RCP commission NDORMS Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Oxford) Bristol) to perform key statistical analysis so in effect this is a collaboration since the RCP will advise on clinical elements of the methodology and will interpret the findings. All individuals with access to record level data are employed by Crown Informatics, Royal College of Physicians Physicians, University of Bristol or University of Oxford. All organisations party to this agreement must comply [31 words unchanged] contractors of the Data Recipient who may have access to that data). [2 paragraphs unchanged] A one-off data flow is planned, consisting of Mortality data only for 2011-2016 inclusive, to NDORMS. Bristol NHIR Biomedical Research Centre Musculoskeletal Research Unit (part of the University of Bristol) This is required for the development of live casemix-adjusted mortality run-charts by NDORMS, run-charts, which will allow sites to monitor their mortality rates much more accurately and in real-time. This work is still ongoing.

Expected output

Publications and outputs using the data to date: data: NHFD annual report 2016, 2017, 2018 2018, 2019, 2020, 2021 [2 paragraphs unchanged] https://www.nhfd.co.uk [4 paragraphs unchanged]

Expected measurable benefits

[22 paragraphs unchanged] Codes identifying COVID status will allow the NHFD to monitor and with HQIP approval to publish trends in the prevalence of infection - among inpatients - among people presenting from the community, and - among people from care homes; the most challenging and least understood setting in the current pandemic. In particular, the additional COVID codes will allow for case-mix adjusted outcomes for the mortality data for the outlier analyses. This information would help inform the audit teams model for case-mix adjustment. COVID-19 is the dominant factor affecting all hip fracture services around the country. Hospitals can already log this information themselves, but it is not easy for them to examine how COVID status affects the provision and outcome of care.

Benefits reported

As part of the Falls and Fragility Fracture Audit Programme (FFFAP) within the Care Quality Improvement Department (CQID) at the Royal College of Physicians of London (RCP), the NHFD has now developed into a comprehensive quality improvement initiative and combines several elements: There have been no further yielded benefits realised since the previous version (v6.2). As part of the Falls and Fragility Fracture Audit Programme (FFFAP) within the Care Quality Improvement Department (CQID) at the Royal College of Physicians (RCP), the NHFD has now developed into a comprehensive quality improvement initiative and combines several elements: [14 paragraphs unchanged] Achievements from the HES and list cleaning demographic data already held: [5 paragraphs unchanged]

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Datasets: + Demographics

Unchanged: Objective for processing, Processing activities, Expected output, Expected measurable benefits, Benefits reported.