Evaluating the age extension of the NHS Breast Screening Programme - Age Extension Trial (AgeX)

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

Datasets: + NDRS Cancer Registrations; + NDRS National Radiotherapy Dataset (RTDS); + NDRS Systemic Anti-Cancer Therapy Dataset (SACT)

Objective for processing

[1 paragraph unchanged] The NHS Breast Screening Programme (NHSBSP) routinely invites women aged 50-70 years to come for three-yearly screening. Because of uncertainty about the effects of screening outside this age range, an England-wide cluster-randomised trial is underway was implemented to assess the risks benefits and benefits harms of additional invitations for screening at ages 47 to 49 and, separately, after age 71 (currently 71-73). at ages 71-73. Randomisation into AgeX ended in March 2020 and the AgeX database was consolidated 2020-2024. AgeX is now in a follow-up and analytical phase (2025-2033) which requires ongoing data linkage to NHS England data and is funded by Cancer Research UK (letter of support provided). [1 paragraph unchanged] • Amass and analyse sufficient evidence that will enable the final reports [11 words unchanged] the subject of controversy for over 15 years – specifically whether the harms benefits of breast screening outweigh the benefits. harms. • Assess the risks harms of screening (in particular, the chances of being diagnosed and treated for [9 words unchanged] of saving life) for women undergoing additional screening at ages 47-49, and separately, at ages 71-73. [5 paragraphs unchanged] • HES-ID to MPS-ID HES Admitted Patient Care • NDRS Cancer Registration data • MRIS – Flagging Current Status Report • NDRS Systemic Anti-Cancer Therapy • MRIS - Members and Postings Report • NDRS Radiotherapy Data Set (RTDS) • MRIS - Cohort Event Notification Report • MRIS - Cause of Death Report [1 paragraph unchanged] The level of the data is identifiable. This is necessary because, although [17 words unchanged] such as records from the National Breast Screening System and NHS England data held within the National Cancer Outcome datasets. Registration and Analysis Service. The use of identifiable data enables linkage of individual records from different sources. [2 paragraphs unchanged] • Limited to data between 2011 to end of 2026. 2031. For each individual patient, data will be provided from the date they joined the trial only. [7 paragraphs unchanged] The funding Funding for the follow up phase of AgeX is provided by the Medical Cancer Research Council, and the Department of Health (Public Health England). Funding is provided to UK as well contributions in kind from the Nuffield Department of Population Health (NDPH) at the University of Oxford, who allocate funding Oxford (letter from NDPH stating support to support this study. CRUK provided). The University of Oxford is the sole data processor. There are no other organisations acting in an advisory capacity or as part of an oversight or steering committee. [1 paragraph unchanged] There is no public involvement in the study design. Where individuals have opted out of disease registration by the National Disease Registration Service (NDRS), their data has been permanently removed from the registry and therefore will not be disseminated under this Data Sharing Agreement (DSA). https://digital.nhs.uk/ndrs/patients/opting-out

Processing activities

[2 paragraphs unchanged] NHS England will provide the relevant records from the Cancer Registrations, Civil Registrations of Death, Demographics, and Hospital Episode Statistics – Admitted Patient Care datasets dataset, NDRS Cancer Registration, NDRS RTDS and NDRS SACT to the University of Oxford. The data will contain directly identifying data [16 words unchanged] the data at record level with data already held by the recipient. [1 paragraph unchanged] Data provided by NHS England and its predecessors is stored at the University of Oxford only. The data will be held on servers at the Cancer Epidemiology Unit (CEU) within the Nuffield Department of Population Health (NDPH) at the University of Oxford. [6 paragraphs unchanged] There will be no requirement and no attempt to reidentify individuals when using the pseudonymised dataset. The identifying details will be stored in a separate database to the linked dataset used for analysis. All analyses will use a pseudonymised subset of the data provided. When using the pseudonymised data there will be no requirement or attempt to reidentify individuals. [2 paragraphs unchanged]

Expected output

[1 paragraph unchanged] • Submissions Scientific reports submitted to peer reviewed open-access journals. • Reports for public contributors and relevant stakeholders. [2 paragraphs unchanged] • Publication of dashboards metadata on the AgeX website hosted by the University of Oxford’s website. Oxford (https://www.ceu.ox.ac.uk/research/agex-trial). The outputs will not contain identifiable NHS England data and will only contain aggregated information with small numbers [7 words unchanged] relevant disclosure rules for the dataset(s) from which the information was derived. [1 paragraph unchanged] • Journals Scientific journals [1 paragraph unchanged] • Trial specific website • AgeX website hosted by the University of Oxford [2 paragraphs unchanged] • Co-hosted events such as the Glasgow Symposium Mammographicum. • Relevant breast radiology and breast cancer meetings nationally and internationally [2 paragraphs unchanged] Patients will be followed up until 31st December 2026, therefore the earliest date for publication of results will be 2027/28. The planned publication schedule is outlined fully in the attached protocol for the follow-up phase of AgeX and includes the following reports: - The first report of breast cancer incidence and treatment will be published in 2026. - The first mortality report will be published in 2027/28 to include follow-up to December 2026. - The final mortality report will be published in 2032/33 to include follow-up to December 2031, and will be accompanied by updated analyses of breast cancer incidence and treatment. Each scientific report will be accompanied with lay summaries developed in collaboration with members of the AgeX public advisory group.

Expected measurable benefits

The aim of the trial AgeX is to assess reliably the risks benefits and benefits harms of additional NHS invitations for breast screening before age 50 in women aged 47-49 and, separately, after age 70. in women aged 71-73. The UK Health Security Agency will use the findings to inform government and influence policy. This will affect breast screening policy in the UK and influence policy in much of the rest of the world for decades after the results have published. All the findings from AgeX will be used by The UK National Screening Committee (letter of support for funding attached) to directly inform government and influence policy. The benefits of interim analyses are that these will be used to monitor the protection and safety of trial participants in addition to feeding into the wider trial aims. This will affect breast screening policy in the UK and influence policy in much of the rest of the world where similar screening programmes exist for decades after the results have published, and as such is a trial of significant public health importance.

Benefits reported

Stated in the previous version and removed here.

The first analysis of the data will not be complete until 2027/28, therefore there are no yielded benefits to date.

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

Objective for processing

This Data Sharing Agreement permits the retention of the data provided under previous iterations of this Agreement for an interim period. This is a pragmatic approach to provide an active Agreement whilst enabling The University of Oxford to complete the necessary actions to enable a subsequent application to amend and renew the Agreement meeting all applicable data sharing standards as published in NHS Digital’s website (see: https://digital.nhs.uk/services/data-access-request-service-dars/dars-guidance). The University of Oxford requires access to NHS England data for the purposes of the following research project: Evaluating the age extension of the NHS Breast Screening Programme - Age Extension Trial (AgeX). The University of Oxford requires data for the purpose of a clinical trial: 'Evaluating the age extension of the NHS Breast Screening Programme - Age Extension (AgeX)'. The NHS Breast Screening Programme (NHSBSP) routinely invites women aged 50-70 years to come for three-yearly screening. Because of uncertainty about the effects of screening outside this age range, an England-wide cluster-randomised trial is underway to assess the risks and benefits of additional invitations for screening at ages 47 to 49 and, separately, after age 71 (currently 71-73). The NHS Breast Screening Programme routinely invites women aged 50-70 years to come for three-yearly screening. Because of uncertainty about the effects of screening outside this age range, an England-wide cluster-randomised trial is under way to assess the risks and benefits of additional invitations for screening at ages 47 to 49 and, separately, after age 71 (currently 71-73). The following is a summary of the aims of the research project provided by the University of Oxford: Random allocation of small clusters of participants is used to determine (in a 50:50 ratio) which women are offered one additional screening invitation before age 50 and which are not, and which women are offered additional screening after age 71 and which are not. This trial involves about 65 of the 81 breast screening units in England and will randomize at least four million women aged 47-49 and two million aged 71-73 to be invited for additional screening. The Age X database contains cohort Patient identifiers for the 4,553,604 patients flagged. Women will be followed up by routine electronic linkage to NHS and Civil Registration mortality records to assess short-term and long-term effects of the additional screening on: patterns of investigation, detection and treatment of breast lesions; breast cancer incidence; breast cancer mortality; hospital admissions; and overall mortality. • Amass and analyse sufficient evidence that will enable the final reports to provide definitive unequivocal evidence on an issue that has been the subject of controversy for over 15 years – specifically whether the harms of breast screening outweigh the benefits. The principal outcome for screening before age 50 and, separately, after age 70 will be breast cancer mortality, eventually subdivided by 5-year time periods (0-4, 5-9, 10-14 years, etc.) since random allocation. Subsidiary analyses will assess effects on other outcomes. The trial began recruiting participants in 2010 and the main results are expected in the mid-2020s. • Assess the risks of screening (in particular, the chances of being diagnosed and treated for a non-life-threatening cancer) and benefits (in particular, the chances of saving life) for women undergoing additional screening at ages 47-49, and 71-73. Approvals have been obtained, and remain ongoing from London Harrow Research Ethics committee to conduct the trial and Confidentiality Advisory Group (CAG) to process participant’s data under section 251 of the NHS Act and permit linkage to the NHS databases for medical history and civil registration data. The following NHS England data will be accessed: The data collected for this trial is processed under the following articles of GDPR at the Cancer Epidemiology Unit in the Nuffield Department of Population Health at the University of Oxford who act as the Data Controller. • Cancer Registration Data This work is research in the public interest as it aims to improve care for all patients considering undergoing this type of process – informing clinicians and commissioners of variation and outcomes and complications to support work to improve and standardise treatment selection choices. • Civil Registrations of Death The University of Oxford is the sole Data Controller and Data Processor, and has determined that there are no moral or ethical issues from dissemination of data for this purpose. Once received from NHS Digital, the data being processed by the University of Oxford will be pseudonymised and will be compliant with the ICO's "Anonymisation: managing data protection risk" code of practice. • Demographics The data will be pseudonymised before analysis and will not be shared with any third parties. • Hospital Episode Statistics Admitted Patient Care (HES APC) When the main results of the trial are available results will be reported to the Advisory Committee on Breast Cancer Screening which oversees the NHSBSP and reports to government ministers. The NHSBSP sets the standards for the screening units and monitors performance through a national quality assurance network. • HES-ID to MPS-ID HES Admitted Patient Care The reason for this trial is to provide definitive evidence on an issue that has been the subject of controversy for over 20 years – specifically whether the harms of breast screening outweigh the benefits. Previous trials have evidenced the benefits without sufficiently evidencing the potential harms. Numerous claims and counter claims have been made and continue to be made emanating from such sources as patient groups, publications, scientific literature and in the media. The nature of the claims varies over time. For example, it may, at different times, be claimed that harms include increases in operations, hospital stays, related health episodes of specific types, etc. Such claims are typically varied with a lack of definitive supporting evidence. • MRIS – Flagging Current Status Report This trial aims to amass and analyse sufficient evidence that will enable the final reports to provide definitive unequivocal evidence. Where such claims of harm have been made, the trial will investigate the validity and tailor the content of planned reports to provide responses. Due to the potentially wide variety of possible harms which might be attributed to screening, the trial requires wide ranging hospital episode data not restricted to specific types of episode. • MRIS - Members and Postings Report The number of women involved in the trial is necessarily large in order to ensure that the findings of the trial are unequivocal. The sample size has been reviewed by numerous committees and accepted as being appropriate. • MRIS - Cohort Event Notification Report In terms of the number of years of data required, it is necessary to have data over such a long period because firstly, the screening programme currently offers women 7 screenings between the ages of 50 and 70 and the trial is investigating the effect of adding just one screen to the existing 7 routinely offered. Therefore, the screening period is already more than 20 years and effects of screening are likely to become visible at least 10 years after the first screening. Secondly, it is essential for the trial design to have information on morbidity prior to randomization. It is well established that women with prior co-morbidities are less likely to attend for screening, when invited. It is stated explicitly in the protocol that groups of women will be excluded from the main analyses where there are known (on the basis of information collected before randomization) to be unlikely to accept an invitation for screening. It is therefore a fundamental requirement that the trial analyses HES data going back as far as is possible. • MRIS - Cause of Death Report Access to these data sets is required to assess short-term and long-term effects of the additional screening on: patterns of investigation; detection and treatment of breast lesions; breast cancer incidence; breast cancer mortality; hospital admissions; and overall mortality. The level of the data is identifiable. This is necessary because, although the applicant will pseudonymise the data before analysis, the study requires collation of data from other sources such as records from the National Breast Screening System and NHS England Cancer Outcome datasets. The use of identifiable data enables linkage of individual records from different sources. The data will be minimised as follows: • Limited to data for a study cohort of 4,553,604 women identified by the University of Oxford as meeting the following criteria: Ages 47-49 or ages 71-73 at the time of inclusion in the study. • Limited to data between 2011 to end of 2026. For each individual patient, data will be provided from the date they joined the trial only. The University of Oxford is the sole data controller as the organisation responsible for ensuring that the data will only be processed for the purpose described above. The Department of Health commissioned the University of Oxford to undertake the work. The Department of Health does not specify what data are required nor how the data shall be processed to achieve the purpose. Such decisions are taken by the University of Oxford only. The processing described within this Agreement falls under Article 6(1)(e) of the General Data Protection Regulation (GDPR) processing is necessary for the performance of a task carried out in the public interest for research purposes is justified as: • The study is directly related to population health benefits which are in the public interest. • The University of Oxford is a public authority as defined in the Data Protection Act 2018. • Statute 1(3) of the Statutes of the University of Oxford states: "The principal objects of the University are the advancement of learning by teaching and research and its dissemination by every means. (http://www.admin.ox.ac.uk/statutes/1086-120.shtml)" The condition for processing special category personal data is met by the use of Article 9(2)j - processing is necessary for archiving purposes in the public interest, scientific or historical research purposes. This work is research in the public interest as it aims to improve care for all patients considering undergoing this type of process – informing clinicians and commissioners of variation and outcomes and complications to support work to improve and standardise treatment selection choices. The funding is provided by the Medical Research Council, and the Department of Health (Public Health England). Funding is provided to the Nuffield Department of Population Health (NDPH) at the University of Oxford, who allocate funding to support this study. The University of Oxford is the sole data processor. There are no other organisations acting in an advisory capacity or as part of an oversight or steering committee. The data provided by NHS England will be accessed by substantive employees of The University of Oxford only. There is no public involvement in the study design.

Processing activities

Under this Agreement, the data may be securely stored and processed. No new data will be provided by NHS Digital under this Agreement. The University of Oxford provided the following participant identifiers under previous iterations of this agreement: study ID, NHS Number, Name, Date of Birth, address, postcode and gender (if available). To date, cohort patient identifiers (nhs number, date of birth, postcode) have been supplied to NHS Digital (formerly known as HSCIC) in separate batches for flagging on NHS Digital's MIDAS system and NHS Digital has supplied periodic updates on deaths, cancers and exits from NHS registration. 4,553,604 cohort members have been flagged to date. Flagging is now complete and there will be no further participants added to the cohort. There will be no additions to the cohort and no further data will flow to NHS England for the purposes of this Agreement. In addition to NHS England will provide the ongoing supply relevant records from the Cancer Registrations, Civil Registrations of notifications, NHS Digital will perform further linkage to Death, Demographics, and Hospital Episode Statistics data and supply the linked data – Admitted Patient Care datasets to the Cancer Epidemiology Unit (CEU) within the Nuffield Department of Population Health (NDPH) at the University of Oxford. The data will contain directly identifying data items including Names; NHS Number; Date of Birth; Postcode; and Gender, which are required to link the data at record level with data already held by the recipient. All data received from NHS Digital is added to the core AgeX database, which also contains trial participants’ records from the National Breast Screening System and NHS Digital (formerly Public Health England) cancer outcome datasets. The data will not be transferred to any other location. The purpose for collecting the data is to assess the short-term and long-term effects of the additional screening on: patterns of investigation; detection and treatment of breast lesions; breast cancer incidence; breast cancer mortality; hospital admissions, and overall mortality. Data provided by NHS England and its predecessors is stored at the University of Oxford only. The data will be held exclusively at the Cancer Epidemiology Unit (CEU) within the Nuffield Department of Population Health (NDPH) at the University of Oxford and used solely for the purpose of this project. Backup storage is located in three separate secure rooms within NDPH buildings. Backups should be useable in the event of a disaster at the main site and will be accorded the same physical & electronic protections as live data. All data (including core trial data on women randomised, flagging information, deaths, cancers, externally linked data, some admin data) is held in a central database. Several internal ID's exist so that data can be appropriately structured. No statistical analyses of the data are performed directly on this database. Direct access to the entire database is limited to the senior systems analyst who is a substantive employee of the University of Oxford. Access to personal identifiers is required to process externally linked data, and to prepare datasets for external linkage to HES, Civil Registration Mortality, Cancer Registration, NHS Registration, and NHS Screening records. The data will be accessed onsite at the premises of the University of Oxford. Data will also be accessed remotely using University of Oxford Laptop computers using secure VPN. Secure access to laptops is enabled via password protection. Pseudonymised analysis data is extracted from the database, either as database views or via text files. When accessing views directly, appropriate roles exist to strictly limit access to the appropriate data only. Analysis datasets will never contain any personally identifiable data e.g. name, NHS number etc. ONLY linkage datasets will require this information. Where appropriate year and month will be used in place of full dates, similarly, other abstractions will be applied to non-date fields where such abstractions are possible and appropriate. The data will not leave England/Wales at any time. Access is restricted to individuals within the Cancer Epidemiology Unit in the Nuffield Department of Population Health at the University of Oxford, who have authorisation from the Principal Investigator. All such individuals are substantive employees of the University of Oxford. All personnel accessing the data have been appropriately trained in data protection and confidentiality. All data received from NHS England is linked to the core AgeX database, which also contains trial participants’ records from the National Breast Screening System and NHS England Cancer Outcome datasets. There will be no requirement and no attempt to reidentify individuals when using the pseudonymised dataset. All data (including core trial data on women randomised, flagging information, deaths, cancers, externally linked data, some admin data) is held in a central Age-X database. Several internal identifiers exist so that data can be appropriately structured. No statistical analyses of the data are performed directly on this database. Direct access to the entire database is limited to the senior systems analyst who is a substantive employee of the University of Oxford. Access to participant's personal identifiers is required to process externally linked data, and to prepare datasets for external linkage to HES APC, Civil Registration of Death, and Cancer Registrations data. Pseudonymised data for the purpose of analysis is extracted from the database, either as database views or via text files. When accessing views directly, appropriate roles exist to strictly limit access to the appropriate data only. Analysis datasets will never contain any personally identifiable data e.g. name, NHS number etc. Only linkage datasets will require this information. Where appropriate year and month will be used in place of full dates, similarly, other abstractions will be applied to non-date fields where such abstractions are possible and appropriate.

Expected output

This Agreement permits the secure retention and processing of data provided under a previous iteration of this agreement. No new outputs will be produced under this Data Sharing Agreement. The expected outputs of the processing will be: The primary output of the study will be findings disseminated by publications in peer-reviewed open-access journals and presented at medical conferences to academics, NHS national policy makers and on the web. The results are expected to help determine future NHS policy on breast cancer screening outside the 50-70 age group. The main results are not expected before the mid-2020s. • Submissions to peer reviewed open-access journals. Outputs will include only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide. • Presentations at medical conferences to academics. As with any clinical trial, it is necessary to gather evidence incrementally throughout the course of the trial in order to check that the trial itself is not exposing participants to increased risk of harm or, conversely, to check for evidence of benefits. The trial end date is not fixed and may be brought forward if at any stage, sufficient evidence is obtained. • Reports to NHS national policy makers. The outputs from the interim analyses will be annual reports to the Data Ethics and Monitoring Committee who may also request additional analyses within the scope of the clinical trial’s objectives if they identify the need. The Data Ethics and Monitoring Committee report any concerns about the trial or findings to the trial management group. • Publication of dashboards on the University of Oxford’s website. 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. The outputs will be communicated to relevant recipients through the following dissemination channels: • Journals • Social media • Trial specific website • Public reports developed in collaboration with NHS bodies • Briefing documents provided to national health agencies such as UK Health Security Agency and UK National Screening Committee • Co-hosted events such as the Glasgow Symposium Mammographicum. • Public events such as conferences • Press/media engagement Patients will be followed up until 31st December 2026, therefore the earliest date for publication of results will be 2027/28.

Expected measurable benefits

This Agreement permits the secure retention and processing of data provided under a previous iteration of this agreement. The aim of the trial is to assess reliably the risks and benefits of additional NHS invitations for breast screening before age 50 and, separately, after age 70. The aim of the trial is to assess reliably the risks and benefits of additional NHS invitations for breast screening before age 50 and, separately, after age 70. The results are expected to help determine future NHS policy on age at breast screening. The main results are not expected before the mid-2020s but interim results will be reviewed and the trial end date may be brought forward if the evidence is clear before then. The UK Health Security Agency will use the findings to inform government and influence policy. This will affect breast screening policy in the UK and influence policy in much of the rest of the world for decades after the results have published. Public Health England will use the findings to inform government and influence policy. This will affect breast screening policy in the UK and influence policy in much of the rest of the world for decades after the results have published. The benefits of interim analyses are that these will be used to monitor the protection and safety of trial participants in addition to feeding into the wider trial aims.

Benefits reported

The benefits of interim analyses are that these will be used to monitor the protection and safety of trial participants in addition to feeding into the wider trial aims. The first analysis of the data will not be complete until 2027/28, therefore there are no yielded benefits to date.

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

Objective for processing

23/07/2019: addition of only 1 year complete HES data for 2018/2019. This Data Sharing Agreement permits the retention of the data provided under previous iterations of this Agreement for an interim period. This is a pragmatic approach to provide an active Agreement whilst enabling The University of Oxford to complete the necessary actions to enable a subsequent application to amend and renew the Agreement meeting all applicable data sharing standards as published in NHS Digital’s website (see: https://digital.nhs.uk/services/data-access-request-service-dars/dars-guidance). [2 paragraphs unchanged] Random allocation of small clusters of participants is used to determine (in [54 words unchanged] 47-49 and two million aged 71-73 to be invited for additional screening. The Age X database contains cohort Patient identifiers for the 4,553,604 patients flagged. Women will be followed up by routine electronic linkage to NHS and [23 words unchanged] lesions; breast cancer incidence; breast cancer mortality; hospital admissions; and overall mortality. [4 paragraphs unchanged] The University of Oxford is the sole Data Controller and Data Processor, and has determined that there are no moral or ethical issues from dissemination [25 words unchanged] compliant with the ICO's "Anonymisation: managing data protection risk" code of practice. [2 paragraphs unchanged] The reason for this trial is to provide definitive evidence on an issue that has been the subject of controversy for over 15 20 years – specifically whether the harms of breast screening outweigh the benefits. [71 words unchanged] Such claims are typically varied with a lack of definitive supporting evidence. [3 paragraphs unchanged]

Processing activities

To date, identifying details have been supplied to NHS Digital (formerly known as HSCIC) in separate batches for flagging on NHS Digital's MIDAS system and NHS Digital has supplied periodic updates on deaths, cancers and exits from NHS registration. Approximately 3.1 million cohort members have been flagged to date with the number to rise to 4.2m during 2019. Under this Agreement, the data may be securely stored and processed. No new data will be provided by NHS Digital under this Agreement. Further batches of trial participants will be supplied as new participants are randomised into the trial on an ongoing basis. To date, cohort patient identifiers (nhs number, date of birth, postcode) have been supplied to NHS Digital (formerly known as HSCIC) in separate batches for flagging on NHS Digital's MIDAS system and NHS Digital has supplied periodic updates on deaths, cancers and exits from NHS registration. 4,553,604 cohort members have been flagged to date. Flagging is now complete and there will be no further participants added to the cohort. [1 paragraph unchanged] All data received from NHS Digital is added to the core AgeX database, which also contains trial participants’ records from the National Breast Screening System and PHE NHS Digital (formerly Public Health England) cancer outcome datasets. [1 paragraph unchanged] The data will be held exclusively at the Cancer Epidemiology Unit (CEU) within the Nuffield Department of Population Health (NDPH) at the University of Oxford and used solely for the purpose of this project. All data (including core trial data on women randomized, randomised, flagging information, deaths, cancers, externally linked data, some admin data) is held [25 words unchanged] this database. Direct access to the entire database is limited to the database manager senior systems analyst who is a substantive employee of the University of Oxford. Access to [12 words unchanged] datasets for external linkage to HES, Civil Registration Mortality, Cancer Registration, NHS Registration Registration, and NHS Screening records. [1 paragraph unchanged]

Expected output

This Agreement permits the secure retention and processing of data provided under a previous iteration of this agreement. No new outputs will be produced under this Data Sharing Agreement. [4 paragraphs unchanged]

Expected measurable benefits

This Agreement permits the secure retention and processing of data provided under a previous iteration of this agreement. [2 paragraphs unchanged] The benefits of interim analyses are that these will be used to monitor the protection and safety of trial participants in addition to feeding into the wider trial aims.

Benefits reported

No benefits to date the study is due to be completed in the mid-2020s The benefits of interim analyses are that these will be used to monitor the protection and safety of trial participants in addition to feeding into the wider trial aims.

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

Datasets: + Cancer Registration Data; + Civil Registrations of Death; + Demographics; + HES-ID to MPS-ID HES Admitted Patient Care

Objective for processing

23/07/2019: addition of only 1 year complete HES data for 2018/2019. [14 paragraphs unchanged]

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

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

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

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

Processing activities

To date, identifiable identifying details have been supplied to NHS Digital (formerly known as HSCIC) in [31 words unchanged] flagged to date with the number to rise to 4.2m during 2019. [7 paragraphs unchanged]

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

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

Datasets: + Hospital Episode Statistics Admitted Patient Care (HES APC); + MRIS - Bespoke · − MRIS - Personal Demographics Service; − MRIS - Scottish NHS / Registration

Objective for processing

Evaluating the age extension of the NHS Breast Screening ProgrammeTo evaluate the effects of a) offering women one additional screen at age 47-49 years and b) offering women one additional screen at age 71-73 years, within the existing NHS Breast Screening Programme which currently offers women seven 3-yearly screens between the ages of 50 and 70 years.   To date there is limited evidence on the net benefit of extending the age range for breast screening; no trial has looked at the added value of one extra screen within an existing screening programme.This trial provides a unique opportunity to assess the effects of adding an extra screening invitation to an existing screening programme. The results have the potential to inform future screening policy in the UK and elsewhere. The University of Oxford requires data for the purpose of a clinical trial: 'Evaluating the age extension of the NHS Breast Screening Programme - Age Extension (AgeX)'. The NHS Breast Screening Programme routinely invites women aged 50-70 years to come for three-yearly screening. Because of uncertainty about the effects of screening outside this age range, an England-wide cluster-randomised trial is under way to assess the risks and benefits of additional invitations for screening at ages 47 to 49 and, separately, after age 71 (currently 71-73). Random allocation of small clusters of participants is used to determine (in a 50:50 ratio) which women are offered one additional screening invitation before age 50 and which are not, and which women are offered additional screening after age 71 and which are not. This trial involves about 65 of the 81 breast screening units in England and will randomize at least four million women aged 47-49 and two million aged 71-73 to be invited for additional screening. Women will be followed up by routine electronic linkage to NHS and Civil Registration mortality records to assess short-term and long-term effects of the additional screening on: patterns of investigation, detection and treatment of breast lesions; breast cancer incidence; breast cancer mortality; hospital admissions; and overall mortality. The principal outcome for screening before age 50 and, separately, after age 70 will be breast cancer mortality, eventually subdivided by 5-year time periods (0-4, 5-9, 10-14 years, etc.) since random allocation. Subsidiary analyses will assess effects on other outcomes. The trial began recruiting participants in 2010 and the main results are expected in the mid-2020s. Approvals have been obtained, and remain ongoing from London Harrow Research Ethics committee to conduct the trial and Confidentiality Advisory Group (CAG) to process participant’s data under section 251 of the NHS Act and permit linkage to the NHS databases for medical history and civil registration data. The data collected for this trial is processed under the following articles of GDPR at the Cancer Epidemiology Unit in the Nuffield Department of Population Health at the University of Oxford who act as the Data Controller. This work is research in the public interest as it aims to improve care for all patients considering undergoing this type of process – informing clinicians and commissioners of variation and outcomes and complications to support work to improve and standardise treatment selection choices. The University of Oxford has determined that there are no moral or ethical issues from dissemination of data for this purpose. Once received from NHS Digital, the data being processed by the University of Oxford will be pseudonymised and will be compliant with the ICO's "Anonymisation: managing data protection risk" code of practice. The data will be pseudonymised before analysis and will not be shared with any third parties. When the main results of the trial are available results will be reported to the Advisory Committee on Breast Cancer Screening which oversees the NHSBSP and reports to government ministers. The NHSBSP sets the standards for the screening units and monitors performance through a national quality assurance network. The reason for this trial is to provide definitive evidence on an issue that has been the subject of controversy for over 15 years – specifically whether the harms of breast screening outweigh the benefits. Previous trials have evidenced the benefits without sufficiently evidencing the potential harms. Numerous claims and counter claims have been made and continue to be made emanating from such sources as patient groups, publications, scientific literature and in the media. The nature of the claims varies over time. For example, it may, at different times, be claimed that harms include increases in operations, hospital stays, related health episodes of specific types, etc. Such claims are typically varied with a lack of definitive supporting evidence. This trial aims to amass and analyse sufficient evidence that will enable the final reports to provide definitive unequivocal evidence. Where such claims of harm have been made, the trial will investigate the validity and tailor the content of planned reports to provide responses. Due to the potentially wide variety of possible harms which might be attributed to screening, the trial requires wide ranging hospital episode data not restricted to specific types of episode. The number of women involved in the trial is necessarily large in order to ensure that the findings of the trial are unequivocal. The sample size has been reviewed by numerous committees and accepted as being appropriate. In terms of the number of years of data required, it is necessary to have data over such a long period because firstly, the screening programme currently offers women 7 screenings between the ages of 50 and 70 and the trial is investigating the effect of adding just one screen to the existing 7 routinely offered. Therefore, the screening period is already more than 20 years and effects of screening are likely to become visible at least 10 years after the first screening. Secondly, it is essential for the trial design to have information on morbidity prior to randomization. It is well established that women with prior co-morbidities are less likely to attend for screening, when invited. It is stated explicitly in the protocol that groups of women will be excluded from the main analyses where there are known (on the basis of information collected before randomization) to be unlikely to accept an invitation for screening. It is therefore a fundamental requirement that the trial analyses HES data going back as far as is possible.

Processing activities

Not stated in the previous version; added here.

To date, identifiable details have been supplied to NHS Digital (formerly known as HSCIC) in separate batches for flagging on NHS Digital's MIDAS system and NHS Digital has supplied periodic updates on deaths, cancers and exits from NHS registration. Approximately 3.1 million cohort members have been flagged to date with the number to rise to 4.2m during 2019.

Further batches of trial participants will be supplied as new participants are randomised into the trial on an ongoing basis.

In addition to the ongoing supply of notifications, NHS Digital will perform further linkage to Hospital Episode Statistics data and supply the linked data to the Cancer Epidemiology Unit (CEU) within the Nuffield Department of Population Health (NDPH) at the University of Oxford.

All data received from NHS Digital is added to the core AgeX database, which also contains trial participants’ records from the National Breast Screening System and PHE cancer outcome datasets.

The purpose for collecting the data is to assess the short-term and long-term effects of the additional screening on: patterns of investigation; detection and treatment of breast lesions; breast cancer incidence; breast cancer mortality; hospital admissions, and overall mortality.

The data will be held exclusively at the Cancer Epidemiology Unit (CEU) within the Nuffield Department of Population Health at the University of Oxford and used solely for the purpose of this project.

All data (including core trial data on women randomized, flagging information, deaths, cancers, externally linked data, some admin data) is held in a central database. Several internal ID's exist so that data can be appropriately structured. No statistical analyses of the data are performed directly on this database. Direct access to the entire database is limited to the database manager who is a substantive employee of the University of Oxford. Access to personal identifiers is required to process externally linked data, and to prepare datasets for external linkage to HES, Civil Registration Mortality, Cancer Registration, NHS Registration and NHS Screening records.

Pseudonymised analysis data is extracted from the database, either as database views or via text files. When accessing views directly, appropriate roles exist to strictly limit access to the appropriate data only. Analysis datasets will never contain any personally identifiable data e.g. name, NHS number etc. ONLY linkage datasets will require this information. Where appropriate year and month will be used in place of full dates, similarly, other abstractions will be applied to non-date fields where such abstractions are possible and appropriate.

Expected output

Not stated in the previous version; added here.

The primary output of the study will be findings disseminated by publications in peer-reviewed open-access journals and presented at medical conferences to academics, NHS national policy makers and on the web. The results are expected to help determine future NHS policy on breast cancer screening outside the 50-70 age group. The main results are not expected before the mid-2020s.

Outputs will include only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide.

As with any clinical trial, it is necessary to gather evidence incrementally throughout the course of the trial in order to check that the trial itself is not exposing participants to increased risk of harm or, conversely, to check for evidence of benefits. The trial end date is not fixed and may be brought forward if at any stage, sufficient evidence is obtained.

The outputs from the interim analyses will be annual reports to the Data Ethics and Monitoring Committee who may also request additional analyses within the scope of the clinical trial’s objectives if they identify the need. The Data Ethics and Monitoring Committee report any concerns about the trial or findings to the trial management group.

Expected measurable benefits

Not stated in the previous version; added here.

The aim of the trial is to assess reliably the risks and benefits of additional NHS invitations for breast screening before age 50 and, separately, after age 70. The results are expected to help determine future NHS policy on age at breast screening. The main results are not expected before the mid-2020s but interim results will be reviewed and the trial end date may be brought forward if the evidence is clear before then.

Public Health England will use the findings to inform government and influence policy. This will affect breast screening policy in the UK and influence policy in much of the rest of the world for decades after the results have published.

The benefits of interim analyses are that these will be used to monitor the protection and safety of trial participants in addition to feeding into the wider trial aims.

Benefits reported

Yielded Benefits is not a requirement for new applications. No benefits to date the study is due to be completed in the mid-2020s