UK Biobank: A long-term prospective health research study

UK Biobank is the most detailed, long-term prospective health research study in the world, with a specific objective of following up the health of its participants through access to their health records. The resource, which is now one of the most-used research resource for health-related research, enables researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions. This in turn will improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses by contributing to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers (e.g., polygenic risk scores or imaging biomarkers of internal organs) and the development of new and improved treatments for patients with existing conditions, all of which are of considerable benefit to the public. NHS England health record data are essential for UK Biobank to achieve its aims as a prospective resource enabling researchers to follow up participants and establish risk factors and causality of disease (otherwise it is effectively an observational study).

- Explanation of Article 6 justification of processing the data

UK Biobank requires data from NHS England to pursue its legitimate interests, which are to set up and manage a major international research resource for health-related research in the public interest. UK Biobank's Article 6 lawful basis for processing the personal data relating to the participants is therefore Article 6 (1) (f) of the GDPR (legitimate interest).

- Public interest for processing

UK Biobank processes special category personal data provided by NHS England in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. As a major research resource for health-related research, UK Biobank falls clearly within this category as its purpose is to enable scientific discoveries that improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses. The data are required for research purposes in the public interest - meeting the conditions on the DPA Schedule 1 Part 1 (4) - which GDPR Recital 52 (2) determines is an appropriate derogation from prohibition on processing special categories of personal data; proportionate to the aim pursued (as above). The ways in which the processing of data will be of benefit to the public, thereby demonstrating that the processing is in the public interest, are described in section 5d.

In the design of the UK Biobank resource, the moral and ethical issues were specifically considered, both internally and externally, and are set out in the Ethics and Governance Framework (https://www.ukbiobank.ac.uk/media/0xsbmfmw/egf.pdf). UK Biobank is provided with relevant ethical input and oversight about the dissemination of participant data (which is all de-identified) from:

- the Access Sub Committee (ASC), which oversees all the data access applications to UK Biobank: the ASC is advised by its own ethical advisers on specific access applications, Ethox (https://www.ethox.ox.ac.uk/); and separately

- the Ethics Advisory Committee (EAC – https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/governance/ethics-advisory-committee), which is also a committee of the UK Biobank Board (and which superseded the original Ethics and Governance Council) and has a remit to provide advice to the Board on general ethical issues that arise during the maintenance, development and use of UK Biobank.

The data included in the proposed dissemination relates only to the consented cohort, rather than that of the general public. There is no risk of potential harm to the public by releasing the cohort’s data as described in the initial informed consent form and participant information leaflet.

Processing meets the requirement in point (g) of Article 9(2) of the General Data Protection Regulations for a basis in the law of the United Kingdom or a part of the United Kingdom. It also meets a condition in Part 2 of Schedule 1 of the Data Protection Act 2018.

- How the requested data achieves UK Biobank’s mission

UK Biobank’s mission is to enable scientific discoveries that improve human health. UK Biobank's operational objective is to combine extensive and precise measurement of exposures with detailed and rigorous follow-up for a wide range of health-related outcomes, and to promote innovative science by maximising secure research access to de-identified data.

Over the past two decades, UK Biobank has developed into a unique research resource, due to its very large size and the range and detail of participant-level data, including extensive phenotype and genotype information. As previously mentioned, it is now one of the most detailed and most-used research resources for health-related research in the world, enabling researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions that affect people in middle and older age, regardless of risk factors. The ability to link to participants’ health records from NHS England, in order to identify specific outcomes occurring during long-term participant follow-up to inform appropriate disease-based research, is critical and fundamental to UK Biobank's mission.

To date, there are more than 30,000 researchers registered with and approved by UK Biobank, from approximately 3,500 approved institutions. Approximately 18,883 researchers are named on projects that are actively using UKB data and there are currently around 3,700 active projects.

- Background to the request

UK Biobank is a detailed, long-term prospective health research study that was established in 2003. Its core funding is received from Wellcome, the Medical Research Council, Cancer Research UK, the British Heart Foundation and NIHR. Further funding for specific enhancements is detailed at https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/about-us/our-funding. No funder is permitted to influence or suppress any outcomes of research undertaking using the UK Biobank resource. Since 2012 UK Biobank has been open to external access applications and for the great majority of this time (nearly a decade), UK Biobank has been approved to receive linked NHS England data to achieve the aims of the study.

Between 2006 and 2010, UK Biobank recruited 503,000 people from the general population – known as “participants” – then aged between 40 and 69 to take part in the study. Inevitably, some of the participants are now deceased and there are approximately 450,000 living participants who currently remain in the study and are the data subjects for the purpose of this Agreement.

The methods used for the invitation and assessment of participants were developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All UK Biobank participants have given fully informed consent to UK Biobank for: access to and long-term storage of information from their medical and other health-related records; use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purposes; and re-contact by UK Biobank to invite them (but without any obligation) to answer further questions or to take part in further assessments.

It is considered very unlikely that participants - and to date UK Biobank has not received such an objection from any of its 503,000 participants since access commenced - would object to the processing activities involved in obtaining linked data from NHS England. Participants can withdraw from UK Biobank at any time (https://www.ukbiobank.ac.uk/media/fpoku2ac/withdrawal-protocol-030412-1.pdf), although since UK Biobank began recruitment in 2006 only a small number (around 1,500) have done so.

The online Research Analysis Platform (RAP) where the Data is held enables researchers to analyse UK Biobank data in-situ (i.e. without having to download the data directly onto their own computers). Access to the whole exome- and genome sequencing data is only possible via the RAP (owing to the sheer size of the data). Currently, about 20% of UK Biobank projects are performed using the RAP, which it is anticipated will increase over time given the imminent availability of whole-genome sequencing data for all 500,000 participants. In addition, early-career researchers and researchers from low- and middle-income countries are eligible for research credits to cover their compute costs, which will further increase uptake of the RAP. Future planned data releases are detailed on the UK Biobank website (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/future-data-release-timelines).

- Data required

UK Biobank requires access to the following datasets

Hospital Episode Statistics Critical Care and Hospital Episode Statistics Admitted Patient Care, Mental Health Services Data Set, Improved Access to Psychological Therapies Data Set, Civil Registrations of death, Cancer Registration data, National Diabetes Audit data, COVID-19 Vaccination Status (this dataset will only be used for COVID-19 research purposes and this restricted purpose is added explicitly to the terms of the Material Transfer Agreement), Personal Demographic Service (PDS) updates, which details changes to participants' demographics, postal addresses or registered GP practices, Medicines prescribed under primary care dataset, NDRS Cancer Registrations, NDRS Systemic Anti-Cancer (SACT), NDRS National Radiotherapy Dataset (RTDS) and NDRS Somatic Molecular Dataset.

- Justification for the datasets requested

UK Biobank is a large-scale biomedical database and research resource, containing in-depth genetic and health information from half a million UK participants. It is now the most detailed, long-term prospective health research study in the world, facilitating health-related research that is in the public interest and enabling the international scientific community to better understand a range of common and life-threatening diseases, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke and mental health.

All of the clinical datasets requested are essential components of the UK Biobank resource and enable epidemiological research into antecedent clinical events prior to diagnosis and the effects of multi-morbidity on future health outcomes. Without linked data from NHS England, researchers would not be able to ascertain the full burden of disease in the cohort nor the effects of particular treatment and management pathways on health. In order for prospective research to track health and the development of disease over time, longitudinal data must be updated with the latest-available healthcare records for the cohort.

Of key importance are datasets related to death, cancer registry notifications and hospital activity, as they provide crucial information about which participants have developed which diseases and at what time. Without access to these data, UK Biobank could not function as a prospective cohort study and therefore research on the associations between social or economic determinants of health and genetic factors and the corresponding risk of developing diseases over time would not be possible.

Linkage to the requested NHSE datasets facilitates research which draws upon data relating to: attendance at hospital for any reason, including hospitalisation for cancer care, diabetes and COVID-19 related complications; care received by participants who are in contact with mental health and learning disability services; and cause of death.

Data from PDS are required to enable UK Biobank to maintain accurate information about participants’ contact details and GP practice in order to enable direct communication with GPs. For example, for the purpose of the UK Biobank imaging assessment study (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study), which is by some distance the largest (by an order of magnitude) imaging study in the world to date. To expand, in the event that during the imaging assessment visit a participant has been flagged by a radiographer with a potentially serious incidental finding and this potentially serious finding has been confirmed by an expert radiologist (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study/incidental-findings) then UK Biobank contacts both the participant and their GP. These data also facilitate participant follow up which, in a cohort study, needs to be as complete as possible, to avoid both unnecessary reduction in cohort numbers (which reduces statistical power of data analyses) and to minimise loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on periodic re-contact with UK Biobank participants to invite them to complete brief questionnaires about health-related exposures or outcomes, or to other enhancement projects (such as to attend an imaging assessment), for which correct participant contact details are required. UK Biobank is also committed to ensuring that participants are informed of progress and developments. UK Biobank does this in part through its annual newsletter, sent via e-mail and/or postal mail, for which up-to-date contact details are required. For obvious reasons, UK Biobank prefers to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress. Although participants are encouraged to inform UK Biobank when they change address, this is not a failsafe way of ensuring that the most up-to-date contact information is held on all participants.

Incorporation of prescription data into UK Biobank facilitates research into the identification of disease occurrence in UK Biobank (e.g. defining who has diabetes through the use of metformin, insulin and other drugs); research that uses the genomic data to predict who will develop side effects prior to drug initiation (e.g. ACE-inhibitor-associated cough as assessed through GP data) and how genetic variation leads to variability in drug response, treatment duration and disease prognosis. It also enables research into the genetic, clinical and lifestyle determinants of treatment-resistant disease (e.g., what are the characteristics of treatment-resistant depression). The prescription data complements the other data sets held by UK Biobank and will inform further on lifestyle determinants, disease risk factors and treatment outcomes.

- Justification for the level of data (identifying, pseudonymised, unsuppressed aggregated)

Pseudonymised data are required in order to correctly link updated NHS records to the existing data held for individual participants within UK Biobank’s database. The process of pseudonymising UK Biobank participants has already been undertaken, and NHS England and UK Biobank both hold a set of pseudo IDs that can be used for linking health data to the correct individuals, meaning no identifying data is needed for this purpose. This is essential to creating and maintaining a longitudinal dataset. UK Biobank’s default position is not to make available to approved researchers any direct and indirect identifiers (such as full postcode, date of birth, name, NHS number, etc. - https://www.ukbiobank.ac.uk/media/3kmh2t1z/de-identification-protocol-v-2-1-29-06-22.pdf). Each approved research project is instead provided with a unique encrypted 7-digit identification number for participants. Hence, it should not be possible for a researcher to identify a UK Biobank participant using the data that UK Biobank releases to them.

- Justification for the number of years requested

UK Biobank requires the latest available data so that this can be added to existing historical data held on each UK Biobank participant, in order to enrich the long-term picture of participants’ health and enable exploration of multiple risk factors which might contribute to the development of a wide range of diseases over time.

- Justification for the geographical spread of the data requested

UK Biobank participants were recruited from across Great Britain – there was no recruitment from Northern Ireland - and participants still reside in all three nations (although there will have inevitably been some movement within and between countries since recruitment).

- Confirmation there are no alternative, less intrusive ways of achieving the purpose

As a very large cohort, it is not feasible for UK Biobank itself to follow up regularly with each of its participants on an individual basis and even if this were possible it would not generate the necessary clinical detail (or anything approaching it) that is contained within the health records. Further, self-reported data do not possess the needed coverage, reliability and consistency that health records contain. The data provided via NHS England - as the primary mechanism for follow-up of the cohort - are crucial to the viability of UK Biobank, particularly as they provide information on the ~450,000 living participants in an entirely unobtrusive manner.

- Data minimisation

For each of these datasets listed above, UK Biobank requires data only for the ~450,000 living UK Biobank participants. It should be noted that UK Biobank is only requesting individual-level identifiers to ensure that the linkage is implemented correctly. UK Biobank regularly informs NHSE of updates to the cohort to ensure that data is not received for participants who have withdrawn from the study.

- Data controller

The sole Data Controller is UK Biobank Limited.

- Data processors

In relation to the data supplied by NHSE there are two appointed data processors and one sub-processor:

1. The University of Oxford is the primary store (and reference copy) of the data that form the UK Biobank resource (including the data supplied by NHSE). The University of Oxford, as a data processor (subject to their agreement with UK Biobank) act only on the instructions of UK Biobank and are required to store participant health data (such as the health data supplied by NHSE) in a secure environment and separately to any identifiers (such as name or NHS number).

2. DNAnexus Inc. operate UK Biobank’s Research Analysis Platform which enables researchers to access the de-identified health data for the purposes of research. DNAnexus act only on the instructions of UK Biobank, as the data processor (again subject to an agreement with UK Biobank) and do not have access to any identifiable participant data.

3. The de-identified data made available through the Research Analysis Platform, is stored in secure cloud storage provided by Amazon Web Services (AWS) and hosted on the AWS node in London. AWS is an approved and contracted sub-processor of DNAnexus. NHSE have requested that AWS are included in the itemised list of “processors”.

- Funders

While UK Biobank receives core funding from Wellcome, the Medical Research Council, Cancer Research UK, the British Heart Foundation, the National Institute for Health Research and philanthropic donations as major funders, these organisations have no access to the data held by UK Biobank and exercise no control or influence in respect of its use.

UK Biobank charges researchers a fee for accessing data and thus is deemed to include a commercial purpose. The fee charged by UK Biobank is on a cost recovery basis and thus covers the incremental costs of servicing an access application. UK Biobank does not seek to recover the cost of development and management of the UK Biobank resource (which already runs into hundreds of millions of pounds). UK Biobank is accessible by all researchers - whether they are from academia or industry. UK Biobank approves applications only where provision of data is for the promotion of health (including epidemiological research), is consistent with participants' consent and the UK Biobank's Ethics and Governance Framework, and is consistent with its access criteria.

The national data opt-out does not apply where explicit consent has been obtained from the patient for the specific purpose.

- Research applications and approved research projects

UK Biobank has developed one of the most detailed and most-used research resource for health-related research in the world, and enabled thousands of researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions. As a prospective resource, its success relies heavily on continued linkage to longitudinal health records from NHS England.

Over the past decade an increasingly large number of researchers have conducted and reported research on an extensive range of topics. The breadth of health conditions that it will be possible to study reliably using even more detailed information about exposures will increase further as the numbers of incident cases of different health outcomes amongst the ageing cohort increase during the next five to ten years. There are currently over 3,600 active approved research projects, a summary of which is available on UK Biobank’s website at https://www.ukbiobank.ac.uk/enable-your-research/approved-research.

Over coming years, UK Biobank will continue to enhance and improve the resource, not only through continued linkage to health-related records, but also through other enhancements to the resource. This will enable further research, leading to substantial patient and NHS benefit.

- Imaging (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study): UK Biobank's imaging study, which will run for the next five to six years, started in 2014 with the aim of collecting multi-modal MRI data on up to 100,000 UK Biobank participants. In 2022, it was announced that funding from the MRC, Calico Life Sciences and the Chan Zuckerberg Initiative will enable UK Biobank to perform up to 60,000 participants a few years later, creating the world’s largest longitudinal imaging dataset to assess disease progression. This dataset will be particularly valuable for researchers studying diseases such as dementia, as it will enable research into how early (i.e. pre-symptomatic) changes in the structure and function of the brain over time are related to dementia risk.

- Genetics (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/genetic-data): to complement the genome-wide genotyping data, in 2022, UK Biobank released whole-exome sequencing data to the global research community, creating the world’s largest exome (all the protein-encoding genes) dataset. In November 2023 the whole genome sequence data was released, creating the biggest whole genome dataset available in the world. This will transform the way in which scientists study the genetics determinants of a wide range of health outcomes. (https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/news/final-data-release-from-the-world-s-largest-whole-exome-sequencing-project).

- Assays (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/biomarker-data): UK Biobank is also undertaking further assays – proteomics, metabolomics, transcriptomics – which enable the disease pathway to be followed, connecting the phenotypic and the genetic data within UK Biobank. A first release of proteomic data took place in early 2023 and a second tranche of metabolomics data was made available to researchers in late 2023.

- Health record data (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/health-related-outcomes-data): As mentioned in section 5a linkage to health-related information is vital for researchers using the UK Biobank resource. UK Biobank will maintain its access to the health record data and it will seek to add other linkage data, particularly including (when available) the primary care data, socio-economic data and data to enhance cancer research.

To date, there are more than 30,000 researchers registered with and approved by UK Biobank, from approximately 3,500 approved institutions. Approximately 18,883 researchers are named on around 3,700 approved research projects t that aim to use UK Biobank data for health-related research that is in the public interest. The majority of applications are still led by academic researchers, but an increasing number are submitted from commercial entities. Summaries of all approved research projects that are currently underway can be found at https://www.ukbiobank.ac.uk/enable-your-research/approved-research. The nature of the research that is currently being conducted reflects the increasing breadth and depth of data that are now available, such as whole-exome and whole-genome sequence data, biochemical assay data and metabolomics data, as well as detailed imaging measures, and much larger numbers of incident cases of many different conditions identified from the NHS England linked healthcare data.

- Outputs from the data processing

Applications to use UK Biobank data are able to generate more than one publication/output. To date, over 11,300 publications have been produced, and the number of papers published each year is increasing rapidly (https://www.ukbiobank.ac.uk/enable-your-research/publications - note that this webpage is currently awaiting a refresh of data for 2022 and 2023). UK Biobank expects to see a continued increase in the number of research publications based upon analysis of UK Biobank data over the coming years.

- Data contained in the outputs

The resource contains individual-level data with associated pseudo IDs. Every dataset made available to approved researchers has a unique set of randomly-generated pseudo IDs. Summary data derived from UK Biobank data are used in an increasing number of journal articles every year but researchers are not permitted to publish individual-level data. UK Biobank has developed guidance about the publication of results about very rare diseases which is available here https://www.ukbiobank.ac.uk/media/cukhqxtp/uk-biobank-policy-on-protecting-confidentiality-in-public-browsers-final_.pdf.

- Approach to dissemination and communication

Researchers who use the UK Biobank resource are responsible for ensuring that the findings of their research are published and disseminated. Under the terms of UK Biobank’s Access procedures and Material Transfer Agreement, they are required to publish the findings of their research and return their results to UK Biobank so they can be made available for request by other researchers. In 2023, over 3,000 publications referenced UK Biobank data.

UK Biobank disseminates huge amounts of information to participants and researchers (plus the general public) via its website and other channels.

UK Biobank makes its governance, history, use, plans, and news available transparently through its website. The website also provides a broad range of information about the resource and its outputs, including hosting lists of approved research (https://www.ukbiobank.ac.uk/enable-your-research/approved-research), publications (https://www.ukbiobank.ac.uk/enable-your-research/publications), as well as more detailed case studies of how UK Biobank is helping scientists and researchers to evolve our understanding of human health and the most common life-threatening diseases (https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/our-impact).

It also has an X (Twitter) feed (https://twitter.com/uk_biobank/) and YouTube channel (https://www.youtube.com/@uk_biobank), releases newsletters annually to participants (e.g. https://www.ukbiobank.ac.uk/explore-your-participation/stay-involved/uk-biobank-newsletter-homepage-2023-24), hosts interactive events with participants at which enhancement projects and the results of research using the resource are discussed (see https://youtu.be/R1aHLOS1qoY for a recording of a participant event held in Newcastle upon Tyne and https://www.ukbiobank.ac.uk/explore-your-participation/stay-involved/a-unique-picture-of-health-the-uk-biobank-imaging-project for a UKB Imaging project online event), organises an annual Scientific Conference attended by an international delegation (see https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/scientific-conference-2023 for details of the conference held in December 2023), and engages with industry partners. UK Biobank presented details of its work at the following external conferences in the past year: the European Society of Human Genetics, the American Society of Human Genetics, European Society of Cardiology, Bio-IT world conference, the Festival of Genomics, the Life Arc MND conference, the World Congress of Psychiatric Genetics, and the American Heart Association annual conference.

Participants can contact the Participant Resource Centre with questions, or to provide feedback. The Centre answers over 20,000 calls and around 35,000 emails from participants annually. Feedback obtained about enhancement studies, either via the PRC or other feedback routes is relayed to project teams and appropriate actions taken to address any issues of concern raised. Participants who make enquiries about international research all receive a written response from UK Biobank and are offered the opportunity to speak directly to UK Biobank staff, including the Principal Investigator.

UK Biobank’s Participant Involvement Strategy sets out our plans to develop an expanded and embedded programme of participant involvement over the next three years. We will use different methods to facilitate both in-depth discussion and broad engagement, including establishing a Participant Advisory Group and a large pool of Participant Reviewers, involving participants in UK Biobank committees, and conducting cohort-wide participant surveys. This will help us deepen the involvement of participants in areas of our work such as communications, data enhancement and linkage, and governance. The programme will also work closely with the UK Biobank Ethics Advisory Committee to ensure participants are consulted about ethical issues that arise in the course of our work.

- Target dates for outputs

Publications have been generated using the resource for over a decade and increase in number year on year (https://www.ukbiobank.ac.uk/enable-your-research/publications - note that this webpage is currently awaiting a refresh of data for 2022 and 2023). Researchers are expected to publish their findings within six months of the completion of their project, or three years following approval to access the data. This is stipulated in the MTA. Additionally, within six months of publication, results must be provided to UK Biobank so that they may be incorporated into the resource where relevant, for the benefit of others.

To date no specific metrics relating to targets have been shared with NHS England. However, UK Biobank is now a major international research resource for health-related research and is, by some distance, the most widely used and published upon health research resource in the world. It enables thousands of researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions, leading to the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses and the number of researchers, applications and publications resulting from the resource continues to grow.

- Yielded benefits

UK Biobank has yielded numerous and wide-ranging benefits to date (also see https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/our-impact). In the past year, UK Biobank has released whole genome sequencing (WGS) of its 500,000 participants, for use alone or in combination with the existing wealth of data UK Biobank has collected over the past 15 years on lifestyle, whole body imaging scans, health information, and proteins found in the blood. Proteomics research has flourished after biological samples from UK Biobank were turned into data, helping to drive the development of new diagnostics and treatments for a wide range of health conditions . Proteomics data was used by University of Oxford to find links to cases of cancer, with some proteins being present in the blood more than seven years before participants went on to be diagnosed with cancer (Papier et al., 2024)., UK Biobank's contribution to the national and international response to the COVID-19 pandemic presents a good case study for the broad value and benefit of the resource. The existence of this large, detailed, prospective cohort with high degrees of participant engagement enabled rapid repurposing in 2020 for COVID-19 research. The frequency of hospitalisation and death data updates from NHS England was immediately increased, and linkage was extended to SARS-CoV-2 testing data and to primary care data under emergency legislation (obtained separately). Much of this work has already translated into policy impact during the pandemic. For example, ~400 publications on COVID-19 have been produced using UK Biobank data. These include: studies of the genetic determinants of severe COVID-19 (Pairo-Castineria et al., 2021); relationships of pre-existing adverse cardiac phenotypes with COVID-19, determined from imaging scans (Raisi-Estabragh et al., 2021); and associations of NMR-metabolomic biomarkers for low-grade inflammation and cardiometabolic disease with severe COVID-19 (Julkunen et al., 2021).

The UK Biobank resource is enabling unprecedented worldwide genomics research and is at the heart of the UK Government's strategy for incorporating genomics in healthcare (HM Government, 2020). This strategy document recognises UK Biobank's central role in the development of the polygenic risk score (PRS) concept, which involves using the combined effects of thousands of genetic variants (each of which has only a small effect) to improve risk prediction for common diseases. Using UK Biobank data, researchers from the Broad Institute were the first to show that polygenic risk scores (PRS) – which combine the effects of very large numbers of genetic variant – could identify people at a high risk of a particular disease (often equivalent to that of a single-gene disorder). In particular, they found that about 5% of the population had a PRS which put them at equivalent risk of developing heart disease to someone who had Familial Hypercholesterolaemia (Khera et al., 2018). Other research using the whole exome sequencing data found that rare genetic mutations in the GPR75 gene (which encodes a receptor in the brain that regulates energy balance) are associated with having half the risk of developing obesity. Protective ‘loss of function’ mutations were found in about one of every 3,000 people sequenced. Heralded as a “genetic superpower”. This discovery will directly lead to approaches (such as gene-editing and gene-silencing) to develop drugs that mimic the function of this mutation (Akbari, et al., 2021).

UK Biobank’s imaging study enables work which provides an excellent example of how a single study can be used to improve understanding of disease. Researchers from the University of Oxford carried out genome association studies of functional and structural brain imaging phenotypes and showed that many of the phenotypes are heritable. The results provide insight into the genetic architecture of the brain that are of relevance to brain development, ageing, dementia and other brain disorders (Elliott, et al., 2018). Research led by University College London used artificial intelligence to analyse the UK Biobank OCT images, uncovering differences in the retina that may help with diagnosing Parkinsons disease seven years before symptoms appear (Wagner et al., 2023). Moreover, using brain MRIs, Queen Mary University London scientists identified whether people would go on to develop dementia years before their actual diagnosis, with an accuracy of 80% (Ereira et al., 2024).

- References

Akbari P, et al. Sequencing of 640,000 exomes identifies GPR75 variants associated with protection from obesity Science, 2021: 373 (6550)

Elliott, L.T., Sharp, K., Alfaro-Almagro, F. et al. Genome-wide association studies of brain imaging phenotypes in UK Biobank. Nature 562, 210–216 (2018).

HM Government. Genome UK: The future of healthcare. 2020

Julkunen H, et al. Metabolic biomarker profiling for identification of susceptibility to severe pneumonia and COVID-19 in the general population. Elife. 2021; 10: e63033

Khera AV, et al. Genome-wide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. Nat Genet. 2018: 50; 1219-24

Pairo-Castineira, E., et al. Genetic mechanisms of critical illness in COVID-19. Nature; 2021; 591, 92–98.

Raisi-Estabragh Z, et al. Adverse cardiovascular magnetic resonance phenotypes are associated with greater likelihood of incident coronavirus disease 2019: findings from the UK Biobank. Aging Clin Exp Res; 2021; 33: 1133-1144

Papier, K., Atkins, J.R., Tong, T.Y.N. et al. Identifying proteomic risk factors for cancer using prospective and exome analyses of 1463 circulating proteins and risk of 19 cancers in the UK Biobank. Nat Commun 15, 4010 (2024). https://doi.org/10.1038/s41467-024-48017-6

Wagner et al., 2023 - https://www.neurology.org/doi/10.1212/WNL.0000000000207727

Ereira, S., Waters, S., Razi, A. et al. Early detection of dementia with default-mode network effective connectivity. Nat. Mental Health 2, 787–800 (2024). https://doi.org/10.1038/s44220-024-00259-5

- How the outputs will achieve the stated purpose

The UK Biobank resource enables researchers to make scientific discoveries that provide considerable benefit to human health, not just for people living in the UK but worldwide. Much of this work relies upon data made available by NHS England. In relation to health outcomes, there are over 1,300 published papers on cardiovascular disease, over 1,500 on cancer, over 500 on mental health, and over 900 on obesity. Other broader research projects have published on the identification of novel patterns of association of many different exposures with a range of health outcomes. The rapid availability of more frequent healthcare records for COVID-19 research has already led to over ~400 papers. A list of publications using UKB data can be accessed at https://www.ukbiobank.ac.uk/enable-your-research/publications (note that this webpage is currently awaiting a refresh of data for 2022 and 2023). These publications will improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses by contributing to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers and the development of new and improved treatments for patients with existing conditions, all of which are of considerable benefit to the public.

UK Biobank is the most detailed, long-term prospective health research study in the world, with a specific objective of following up the health of its participants through access to their health records. The resource, which is now one of the most-used research resource for health-related research, enables researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions. This in turn will improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses by contributing to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers (e.g., polygenic risk scores or imaging biomarkers of internal organs) and the development of new and improved treatments for patients with existing conditions, all of which are of considerable benefit to the public. NHS England health record data are essential for UK Biobank to achieve its aims as a prospective resource enabling researchers to follow up participants and establish risk factors and causality of disease (otherwise it is effectively an observational study).

- Explanation of Article 6 justification of processing the data

UK Biobank requires data from NHS England to pursue its legitimate interests, which are to set up and manage a major international research resource for health-related research in the public interest. UK Biobank's Article 6 lawful basis for processing the personal data relating to the participants is therefore Article 6 (1) (f) of the GDPR (legitimate interest).

- Public interest for processing

UK Biobank processes special category personal data provided by NHS England in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. As a major research resource for health-related research, UK Biobank falls clearly within this category as its purpose is to enable scientific discoveries that improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses. The data are required for research purposes in the public interest - meeting the conditions on the DPA Schedule 1 Part 1 (4) - which GDPR Recital 52 (2) determines is an appropriate derogation from prohibition on processing special categories of personal data; proportionate to the aim pursued (as above). The ways in which the processing of data will be of benefit to the public, thereby demonstrating that the processing is in the public interest, are described in section 5d.

In the design of the UK Biobank resource, the moral and ethical issues were specifically considered, both internally and externally, and are set out in the Ethics and Governance Framework (https://www.ukbiobank.ac.uk/media/0xsbmfmw/egf.pdf). UK Biobank is provided with relevant ethical input and oversight about the dissemination of participant data (which is all de-identified) from:

- the Access Sub Committee (ASC), which oversees all the data access applications to UK Biobank: the ASC is advised by its own ethical advisers on specific access applications, Ethox (https://www.ethox.ox.ac.uk/); and separately

- the Ethics Advisory Committee (EAC – https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/governance/ethics-advisory-committee), which is also a committee of the UK Biobank Board (and which superseded the original Ethics and Governance Council) and has a remit to provide advice to the Board on general ethical issues that arise during the maintenance, development and use of UK Biobank.

The data included in the proposed dissemination relates only to the consented cohort, rather than that of the general public. There is no risk of potential harm to the public by releasing the cohort’s data as described in the initial informed consent form and participant information leaflet.

Processing meets the requirement in point (g) of Article 9(2) of the General Data Protection Regulations for a basis in the law of the United Kingdom or a part of the United Kingdom. It also meets a condition in Part 2 of Schedule 1 of the Data Protection Act 2018.

- How the requested data achieves UK Biobank’s mission

UK Biobank’s mission is to enable scientific discoveries that improve human health. UK Biobank's operational objective is to combine extensive and precise measurement of exposures with detailed and rigorous follow-up for a wide range of health-related outcomes, and to promote innovative science by maximising secure research access to de-identified data.

Over the past two decades, UK Biobank has developed into a unique research resource, due to its very large size and the range and detail of participant-level data, including extensive phenotype and genotype information. As previously mentioned, it is now one of the most detailed and most-used research resources for health-related research in the world, enabling researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions that affect people in middle and older age, regardless of risk factors. The ability to link to participants’ health records from NHS England, in order to identify specific outcomes occurring during long-term participant follow-up to inform appropriate disease-based research, is critical and fundamental to UK Biobank's mission.

To date, there are more than 30,000 researchers registered with and approved by UK Biobank, from approximately 3,500 approved institutions. Approximately 18,883 researchers are named on projects that are actively using UKB data and there are currently around 3,700 active projects.

- Background to the request

UK Biobank is a detailed, long-term prospective health research study that was established in 2003. Its core funding is received from Wellcome, the Medical Research Council, Cancer Research UK, the British Heart Foundation and NIHR. Further funding for specific enhancements is detailed at https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/about-us/our-funding. No funder is permitted to influence or suppress any outcomes of research undertaking using the UK Biobank resource. Since 2012 UK Biobank has been open to external access applications and for the great majority of this time (nearly a decade), UK Biobank has been approved to receive linked NHS England data to achieve the aims of the study.

Between 2006 and 2010, UK Biobank recruited 503,000 people from the general population – known as “participants” – then aged between 40 and 69 to take part in the study. Inevitably, some of the participants are now deceased and there are approximately 450,000 living participants who currently remain in the study and are the data subjects for the purpose of this Agreement.

The methods used for the invitation and assessment of participants were developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All UK Biobank participants have given fully informed consent to UK Biobank for: access to and long-term storage of information from their medical and other health-related records; use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purposes; and re-contact by UK Biobank to invite them (but without any obligation) to answer further questions or to take part in further assessments.

It is considered very unlikely that participants - and to date UK Biobank has not received such an objection from any of its 503,000 participants since access commenced - would object to the processing activities involved in obtaining linked data from NHS England. Participants can withdraw from UK Biobank at any time (https://www.ukbiobank.ac.uk/media/fpoku2ac/withdrawal-protocol-030412-1.pdf), although since UK Biobank began recruitment in 2006 only a small number (around 1,500) have done so.

The online Research Analysis Platform (RAP) where the Data is held enables researchers to analyse UK Biobank data in-situ (i.e. without having to download the data directly onto their own computers). Access to the whole exome- and genome sequencing data is only possible via the RAP (owing to the sheer size of the data). Currently, about 20% of UK Biobank projects are performed using the RAP, which it is anticipated will increase over time given the imminent availability of whole-genome sequencing data for all 500,000 participants. In addition, early-career researchers and researchers from low- and middle-income countries are eligible for research credits to cover their compute costs, which will further increase uptake of the RAP. Future planned data releases are detailed on the UK Biobank website (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/future-data-release-timelines).

- Data required

UK Biobank requires access to the following datasets

Hospital Episode Statistics Critical Care and Hospital Episode Statistics Admitted Patient Care, Mental Health Services Data Set, Improved Access to Psychological Therapies Data Set, Civil Registrations of death, Cancer Registration data, National Diabetes Audit data, COVID-19 Vaccination Status (this dataset will only be used for COVID-19 research purposes and this restricted purpose is added explicitly to the terms of the Material Transfer Agreement), Personal Demographic Service (PDS) updates, which details changes to participants' demographics, postal addresses or registered GP practices, Medicines prescribed under primary care dataset, NDRS Cancer Registrations, NDRS Systemic Anti-Cancer (SACT), NDRS National Radiotherapy Dataset (RTDS) and NDRS Somatic Molecular Dataset.

- Justification for the datasets requested

UK Biobank is a large-scale biomedical database and research resource, containing in-depth genetic and health information from half a million UK participants. It is now the most detailed, long-term prospective health research study in the world, facilitating health-related research that is in the public interest and enabling the international scientific community to better understand a range of common and life-threatening diseases, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke and mental health.

All of the clinical datasets requested are essential components of the UK Biobank resource and enable epidemiological research into antecedent clinical events prior to diagnosis and the effects of multi-morbidity on future health outcomes. Without linked data from NHS England, researchers would not be able to ascertain the full burden of disease in the cohort nor the effects of particular treatment and management pathways on health. In order for prospective research to track health and the development of disease over time, longitudinal data must be updated with the latest-available healthcare records for the cohort.

Of key importance are datasets related to death, cancer registry notifications and hospital activity, as they provide crucial information about which participants have developed which diseases and at what time. Without access to these data, UK Biobank could not function as a prospective cohort study and therefore research on the associations between social or economic determinants of health and genetic factors and the corresponding risk of developing diseases over time would not be possible.

Linkage to the requested NHSE datasets facilitates research which draws upon data relating to: attendance at hospital for any reason, including hospitalisation for cancer care, diabetes and COVID-19 related complications; care received by participants who are in contact with mental health and learning disability services; and cause of death.

Data from PDS are required to enable UK Biobank to maintain accurate information about participants’ contact details and GP practice in order to enable direct communication with GPs. For example, for the purpose of the UK Biobank imaging assessment study (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study), which is by some distance the largest (by an order of magnitude) imaging study in the world to date. To expand, in the event that during the imaging assessment visit a participant has been flagged by a radiographer with a potentially serious incidental finding and this potentially serious finding has been confirmed by an expert radiologist (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study/incidental-findings) then UK Biobank contacts both the participant and their GP. These data also facilitate participant follow up which, in a cohort study, needs to be as complete as possible, to avoid both unnecessary reduction in cohort numbers (which reduces statistical power of data analyses) and to minimise loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on periodic re-contact with UK Biobank participants to invite them to complete brief questionnaires about health-related exposures or outcomes, or to other enhancement projects (such as to attend an imaging assessment), for which correct participant contact details are required. UK Biobank is also committed to ensuring that participants are informed of progress and developments. UK Biobank does this in part through its annual newsletter, sent via e-mail and/or postal mail, for which up-to-date contact details are required. For obvious reasons, UK Biobank prefers to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress. Although participants are encouraged to inform UK Biobank when they change address, this is not a failsafe way of ensuring that the most up-to-date contact information is held on all participants.

Incorporation of prescription data into UK Biobank facilitates research into the identification of disease occurrence in UK Biobank (e.g. defining who has diabetes through the use of metformin, insulin and other drugs); research that uses the genomic data to predict who will develop side effects prior to drug initiation (e.g. ACE-inhibitor-associated cough as assessed through GP data) and how genetic variation leads to variability in drug response, treatment duration and disease prognosis. It also enables research into the genetic, clinical and lifestyle determinants of treatment-resistant disease (e.g., what are the characteristics of treatment-resistant depression). The prescription data complements the other data sets held by UK Biobank and will inform further on lifestyle determinants, disease risk factors and treatment outcomes.

- Justification for the level of data (identifying, pseudonymised, unsuppressed aggregated)

Pseudonymised data are required in order to correctly link updated NHS records to the existing data held for individual participants within UK Biobank’s database. The process of pseudonymising UK Biobank participants has already been undertaken, and NHS England and UK Biobank both hold a set of pseudo IDs that can be used for linking health data to the correct individuals, meaning no identifying data is needed for this purpose. This is essential to creating and maintaining a longitudinal dataset. UK Biobank’s default position is not to make available to approved researchers any direct and indirect identifiers (such as full postcode, date of birth, name, NHS number, etc. - https://www.ukbiobank.ac.uk/media/3kmh2t1z/de-identification-protocol-v-2-1-29-06-22.pdf). Each approved research project is instead provided with a unique encrypted 7-digit identification number for participants. Hence, it should not be possible for a researcher to identify a UK Biobank participant using the data that UK Biobank releases to them.

- Justification for the number of years requested

UK Biobank requires the latest available data so that this can be added to existing historical data held on each UK Biobank participant, in order to enrich the long-term picture of participants’ health and enable exploration of multiple risk factors which might contribute to the development of a wide range of diseases over time.

- Justification for the geographical spread of the data requested

UK Biobank participants were recruited from across Great Britain – there was no recruitment from Northern Ireland - and participants still reside in all three nations (although there will have inevitably been some movement within and between countries since recruitment).

- Confirmation there are no alternative, less intrusive ways of achieving the purpose

As a very large cohort, it is not feasible for UK Biobank itself to follow up regularly with each of its participants on an individual basis and even if this were possible it would not generate the necessary clinical detail (or anything approaching it) that is contained within the health records. Further, self-reported data do not possess the needed coverage, reliability and consistency that health records contain. The data provided via NHS England - as the primary mechanism for follow-up of the cohort - are crucial to the viability of UK Biobank, particularly as they provide information on the ~450,000 living participants in an entirely unobtrusive manner.

- Data minimisation

For each of these datasets listed above, UK Biobank requires data only for the ~450,000 living UK Biobank participants. It should be noted that UK Biobank is only requesting individual-level identifiers to ensure that the linkage is implemented correctly. UK Biobank regularly informs NHSE of updates to the cohort to ensure that data is not received for participants who have withdrawn from the study.

- Data controller

The sole Data Controller is UK Biobank Limited.

- Data processors

In relation to the data supplied by NHSE there are two appointed data processors and one sub-processor:

1. The University of Oxford is the primary store (and reference copy) of the data that form the UK Biobank resource (including the data supplied by NHSE). The University of Oxford, as a data processor (subject to their agreement with UK Biobank) act only on the instructions of UK Biobank and are required to store participant health data (such as the health data supplied by NHSE) in a secure environment and separately to any identifiers (such as name or NHS number).

2. DNAnexus Inc. operate UK Biobank’s Research Analysis Platform which enables researchers to access the de-identified health data for the purposes of research. DNAnexus act only on the instructions of UK Biobank, as the data processor (again subject to an agreement with UK Biobank) and do not have access to any identifiable participant data.

3. The de-identified data made available through the Research Analysis Platform, is stored in secure cloud storage provided by Amazon Web Services (AWS) and hosted on the AWS node in London. AWS is an approved and contracted sub-processor of DNAnexus. NHSE have requested that AWS are included in the itemised list of “processors”.

- Funders

While UK Biobank receives core funding from Wellcome, the Medical Research Council, Cancer Research UK, the British Heart Foundation, the National Institute for Health Research and philanthropic donations as major funders, these organisations have no access to the data held by UK Biobank and exercise no control or influence in respect of its use.

UK Biobank charges researchers a fee for accessing data and thus is deemed to include a commercial purpose. The fee charged by UK Biobank is on a cost recovery basis and thus covers the incremental costs of servicing an access application. UK Biobank does not seek to recover the cost of development and management of the UK Biobank resource (which already runs into hundreds of millions of pounds). UK Biobank is accessible by all researchers - whether they are from academia or industry. UK Biobank approves applications only where provision of data is for the promotion of health (including epidemiological research), is consistent with participants' consent and the UK Biobank's Ethics and Governance Framework, and is consistent with its access criteria.

The national data opt-out does not apply where explicit consent has been obtained from the patient for the specific purpose.

- Research applications and approved research projects

UK Biobank has developed one of the most detailed and most-used research resource for health-related research in the world, and enabled thousands of researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions. As a prospective resource, its success relies heavily on continued linkage to longitudinal health records from NHS England.

Over the past decade an increasingly large number of researchers have conducted and reported research on an extensive range of topics. The breadth of health conditions that it will be possible to study reliably using even more detailed information about exposures will increase further as the numbers of incident cases of different health outcomes amongst the ageing cohort increase during the next five to ten years. There are currently over 3,600 active approved research projects, a summary of which is available on UK Biobank’s website at https://www.ukbiobank.ac.uk/enable-your-research/approved-research.

Over coming years, UK Biobank will continue to enhance and improve the resource, not only through continued linkage to health-related records, but also through other enhancements to the resource. This will enable further research, leading to substantial patient and NHS benefit.

- Imaging (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study): UK Biobank's imaging study, which will run for the next five to six years, started in 2014 with the aim of collecting multi-modal MRI data on up to 100,000 UK Biobank participants. In 2022, it was announced that funding from the MRC, Calico Life Sciences and the Chan Zuckerberg Initiative will enable UK Biobank to perform up to 60,000 participants a few years later, creating the world’s largest longitudinal imaging dataset to assess disease progression. This dataset will be particularly valuable for researchers studying diseases such as dementia, as it will enable research into how early (i.e. pre-symptomatic) changes in the structure and function of the brain over time are related to dementia risk.

- Genetics (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/genetic-data): to complement the genome-wide genotyping data, in 2022, UK Biobank released whole-exome sequencing data to the global research community, creating the world’s largest exome (all the protein-encoding genes) dataset. In November 2023 the whole genome sequence data was released, creating the biggest whole genome dataset available in the world. This will transform the way in which scientists study the genetics determinants of a wide range of health outcomes. (https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/news/final-data-release-from-the-world-s-largest-whole-exome-sequencing-project).

- Assays (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/biomarker-data): UK Biobank is also undertaking further assays – proteomics, metabolomics, transcriptomics – which enable the disease pathway to be followed, connecting the phenotypic and the genetic data within UK Biobank. A first release of proteomic data took place in early 2023 and a second tranche of metabolomics data was made available to researchers in late 2023.

- Health record data (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/health-related-outcomes-data): As mentioned in section 5a linkage to health-related information is vital for researchers using the UK Biobank resource. UK Biobank will maintain its access to the health record data and it will seek to add other linkage data, particularly including (when available) the primary care data, socio-economic data and data to enhance cancer research.

To date, there are more than 30,000 researchers registered with and approved by UK Biobank, from approximately 3,500 approved institutions. Approximately 18,883 researchers are named on around 3,700 approved research projects t that aim to use UK Biobank data for health-related research that is in the public interest. The majority of applications are still led by academic researchers, but an increasing number are submitted from commercial entities. Summaries of all approved research projects that are currently underway can be found at https://www.ukbiobank.ac.uk/enable-your-research/approved-research. The nature of the research that is currently being conducted reflects the increasing breadth and depth of data that are now available, such as whole-exome and whole-genome sequence data, biochemical assay data and metabolomics data, as well as detailed imaging measures, and much larger numbers of incident cases of many different conditions identified from the NHS England linked healthcare data.

- Outputs from the data processing

Applications to use UK Biobank data are able to generate more than one publication/output. To date, over 11,300 publications have been produced, and the number of papers published each year is increasing rapidly (https://www.ukbiobank.ac.uk/enable-your-research/publications - note that this webpage is currently awaiting a refresh of data for 2022 and 2023). UK Biobank expects to see a continued increase in the number of research publications based upon analysis of UK Biobank data over the coming years.

- Data contained in the outputs

The resource contains individual-level data with associated pseudo IDs. Every dataset made available to approved researchers has a unique set of randomly-generated pseudo IDs. Summary data derived from UK Biobank data are used in an increasing number of journal articles every year but researchers are not permitted to publish individual-level data. UK Biobank has developed guidance about the publication of results about very rare diseases which is available here https://www.ukbiobank.ac.uk/media/cukhqxtp/uk-biobank-policy-on-protecting-confidentiality-in-public-browsers-final_.pdf.

- Approach to dissemination and communication

Researchers who use the UK Biobank resource are responsible for ensuring that the findings of their research are published and disseminated. Under the terms of UK Biobank’s Access procedures and Material Transfer Agreement, they are required to publish the findings of their research and return their results to UK Biobank so they can be made available for request by other researchers. In 2023, over 3,000 publications referenced UK Biobank data.

UK Biobank disseminates huge amounts of information to participants and researchers (plus the general public) via its website and other channels.

UK Biobank makes its governance, history, use, plans, and news available transparently through its website. The website also provides a broad range of information about the resource and its outputs, including hosting lists of approved research (https://www.ukbiobank.ac.uk/enable-your-research/approved-research), publications (https://www.ukbiobank.ac.uk/enable-your-research/publications), as well as more detailed case studies of how UK Biobank is helping scientists and researchers to evolve our understanding of human health and the most common life-threatening diseases (https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/our-impact).

It also has an X (Twitter) feed (https://twitter.com/uk_biobank/) and YouTube channel (https://www.youtube.com/@uk_biobank), releases newsletters annually to participants (e.g. https://www.ukbiobank.ac.uk/explore-your-participation/stay-involved/uk-biobank-newsletter-homepage-2023-24), hosts interactive events with participants at which enhancement projects and the results of research using the resource are discussed (see https://youtu.be/R1aHLOS1qoY for a recording of a participant event held in Newcastle upon Tyne and https://www.ukbiobank.ac.uk/explore-your-participation/stay-involved/a-unique-picture-of-health-the-uk-biobank-imaging-project for a UKB Imaging project online event), organises an annual Scientific Conference attended by an international delegation (see https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/scientific-conference-2023 for details of the conference held in December 2023), and engages with industry partners. UK Biobank presented details of its work at the following external conferences in the past year: the European Society of Human Genetics, the American Society of Human Genetics, European Society of Cardiology, Bio-IT world conference, the Festival of Genomics, the Life Arc MND conference, the World Congress of Psychiatric Genetics, and the American Heart Association annual conference.

Participants can contact the Participant Resource Centre with questions, or to provide feedback. The Centre answers over 20,000 calls and around 35,000 emails from participants annually. Feedback obtained about enhancement studies, either via the PRC or other feedback routes is relayed to project teams and appropriate actions taken to address any issues of concern raised. Participants who make enquiries about international research all receive a written response from UK Biobank and are offered the opportunity to speak directly to UK Biobank staff, including the Principal Investigator.

UK Biobank’s Participant Involvement Strategy sets out our plans to develop an expanded and embedded programme of participant involvement over the next three years. We will use different methods to facilitate both in-depth discussion and broad engagement, including establishing a Participant Advisory Group and a large pool of Participant Reviewers, involving participants in UK Biobank committees, and conducting cohort-wide participant surveys. This will help us deepen the involvement of participants in areas of our work such as communications, data enhancement and linkage, and governance. The programme will also work closely with the UK Biobank Ethics Advisory Committee to ensure participants are consulted about ethical issues that arise in the course of our work.

- Target dates for outputs

Publications have been generated using the resource for over a decade and increase in number year on year (https://www.ukbiobank.ac.uk/enable-your-research/publications - note that this webpage is currently awaiting a refresh of data for 2022 and 2023). Researchers are expected to publish their findings within six months of the completion of their project, or three years following approval to access the data. This is stipulated in the MTA. Additionally, within six months of publication, results must be provided to UK Biobank so that they may be incorporated into the resource where relevant, for the benefit of others.

To date no specific metrics relating to targets have been shared with NHS England. However, UK Biobank is now a major international research resource for health-related research and is, by some distance, the most widely used and published upon health research resource in the world. It enables thousands of researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions, leading to the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses and the number of researchers, applications and publications resulting from the resource continues to grow.

- Yielded benefits

UK Biobank has yielded numerous and wide-ranging benefits to date (also see https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/our-impact). In the past year, UK Biobank has released whole genome sequencing (WGS) of its 500,000 participants, for use alone or in combination with the existing wealth of data UK Biobank has collected over the past 15 years on lifestyle, whole body imaging scans, health information, and proteins found in the blood. Proteomics research has flourished after biological samples from UK Biobank were turned into data, helping to drive the development of new diagnostics and treatments for a wide range of health conditions . Proteomics data was used by University of Oxford to find links to cases of cancer, with some proteins being present in the blood more than seven years before participants went on to be diagnosed with cancer (Papier et al., 2024)., UK Biobank's contribution to the national and international response to the COVID-19 pandemic presents a good case study for the broad value and benefit of the resource. The existence of this large, detailed, prospective cohort with high degrees of participant engagement enabled rapid repurposing in 2020 for COVID-19 research. The frequency of hospitalisation and death data updates from NHS England was immediately increased, and linkage was extended to SARS-CoV-2 testing data and to primary care data under emergency legislation (obtained separately). Much of this work has already translated into policy impact during the pandemic. For example, ~400 publications on COVID-19 have been produced using UK Biobank data. These include: studies of the genetic determinants of severe COVID-19 (Pairo-Castineria et al., 2021); relationships of pre-existing adverse cardiac phenotypes with COVID-19, determined from imaging scans (Raisi-Estabragh et al., 2021); and associations of NMR-metabolomic biomarkers for low-grade inflammation and cardiometabolic disease with severe COVID-19 (Julkunen et al., 2021).

The UK Biobank resource is enabling unprecedented worldwide genomics research and is at the heart of the UK Government's strategy for incorporating genomics in healthcare (HM Government, 2020). This strategy document recognises UK Biobank's central role in the development of the polygenic risk score (PRS) concept, which involves using the combined effects of thousands of genetic variants (each of which has only a small effect) to improve risk prediction for common diseases. Using UK Biobank data, researchers from the Broad Institute were the first to show that polygenic risk scores (PRS) – which combine the effects of very large numbers of genetic variant – could identify people at a high risk of a particular disease (often equivalent to that of a single-gene disorder). In particular, they found that about 5% of the population had a PRS which put them at equivalent risk of developing heart disease to someone who had Familial Hypercholesterolaemia (Khera et al., 2018). Other research using the whole exome sequencing data found that rare genetic mutations in the GPR75 gene (which encodes a receptor in the brain that regulates energy balance) are associated with having half the risk of developing obesity. Protective ‘loss of function’ mutations were found in about one of every 3,000 people sequenced. Heralded as a “genetic superpower”. This discovery will directly lead to approaches (such as gene-editing and gene-silencing) to develop drugs that mimic the function of this mutation (Akbari, et al., 2021).

UK Biobank’s imaging study enables work which provides an excellent example of how a single study can be used to improve understanding of disease. Researchers from the University of Oxford carried out genome association studies of functional and structural brain imaging phenotypes and showed that many of the phenotypes are heritable. The results provide insight into the genetic architecture of the brain that are of relevance to brain development, ageing, dementia and other brain disorders (Elliott, et al., 2018). Research led by University College London used artificial intelligence to analyse the UK Biobank OCT images, uncovering differences in the retina that may help with diagnosing Parkinsons disease seven years before symptoms appear (Wagner et al., 2023). Moreover, using brain MRIs, Queen Mary University London scientists identified whether people would go on to develop dementia years before their actual diagnosis, with an accuracy of 80% (Ereira et al., 2024).

- References

Akbari P, et al. Sequencing of 640,000 exomes identifies GPR75 variants associated with protection from obesity Science, 2021: 373 (6550)

Elliott, L.T., Sharp, K., Alfaro-Almagro, F. et al. Genome-wide association studies of brain imaging phenotypes in UK Biobank. Nature 562, 210–216 (2018).

HM Government. Genome UK: The future of healthcare. 2020

Julkunen H, et al. Metabolic biomarker profiling for identification of susceptibility to severe pneumonia and COVID-19 in the general population. Elife. 2021; 10: e63033

Khera AV, et al. Genome-wide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. Nat Genet. 2018: 50; 1219-24

Pairo-Castineira, E., et al. Genetic mechanisms of critical illness in COVID-19. Nature; 2021; 591, 92–98.

Raisi-Estabragh Z, et al. Adverse cardiovascular magnetic resonance phenotypes are associated with greater likelihood of incident coronavirus disease 2019: findings from the UK Biobank. Aging Clin Exp Res; 2021; 33: 1133-1144

Papier, K., Atkins, J.R., Tong, T.Y.N. et al. Identifying proteomic risk factors for cancer using prospective and exome analyses of 1463 circulating proteins and risk of 19 cancers in the UK Biobank. Nat Commun 15, 4010 (2024). https://doi.org/10.1038/s41467-024-48017-6

Wagner et al., 2023 - https://www.neurology.org/doi/10.1212/WNL.0000000000207727

Ereira, S., Waters, S., Razi, A. et al. Early detection of dementia with default-mode network effective connectivity. Nat. Mental Health 2, 787–800 (2024). https://doi.org/10.1038/s44220-024-00259-5

- How the outputs will achieve the stated purpose

The UK Biobank resource enables researchers to make scientific discoveries that provide considerable benefit to human health, not just for people living in the UK but worldwide. Much of this work relies upon data made available by NHS England. In relation to health outcomes, there are over 1,300 published papers on cardiovascular disease, over 1,500 on cancer, over 500 on mental health, and over 900 on obesity. Other broader research projects have published on the identification of novel patterns of association of many different exposures with a range of health outcomes. The rapid availability of more frequent healthcare records for COVID-19 research has already led to over ~400 papers. A list of publications using UKB data can be accessed at https://www.ukbiobank.ac.uk/enable-your-research/publications (note that this webpage is currently awaiting a refresh of data for 2022 and 2023). These publications will improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses by contributing to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers and the development of new and improved treatments for patients with existing conditions, all of which are of considerable benefit to the public.

- Purpose of this application

UK Biobank is the most detailed, long-term prospective health research study in the world, with a specific objective of following up the health of its participants through access to their health records. The resource, which is now one of the most-used research resource for health-related research, enables researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions. This in turn will improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses by contributing to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers (e.g., polygenic risk scores or imaging biomarkers of internal organs) and the development of new and improved treatments for patients with existing conditions, all of which are of considerable benefit to the public. NHS England health record data are essential for UK Biobank to achieve its aims as a prospective resource enabling researchers to follow up participants and establish risk factors and causality of disease (otherwise it is effectively an observational study).

- Explanation of Article 6 justification of processing the data

UK Biobank requires data from NHS England to pursue its legitimate interests, which are to set up and manage a major international research resource for health-related research in the public interest. UK Biobank's Article 6 lawful basis for processing the personal data relating to the participants is therefore Article 6 (1) (f) of the GDPR (legitimate interest).

- Public interest for processing

UK Biobank processes special category personal data provided by NHS England in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. As a major research resource for health-related research, UK Biobank falls clearly within this category as its purpose is to enable scientific discoveries that improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses. The data are required for research purposes in the public interest - meeting the conditions on the DPA Schedule 1 Part 1 (4) - which GDPR Recital 52 (2) determines is an appropriate derogation from prohibition on processing special categories of personal data; proportionate to the aim pursued (as above). The ways in which the processing of data will be of benefit to the public, thereby demonstrating that the processing is in the public interest, are described in section 5d.

In the design of the UK Biobank resource, the moral and ethical issues were specifically considered, both internally and externally, and are set out in the Ethics and Governance Framework (https://www.ukbiobank.ac.uk/media/0xsbmfmw/egf.pdf). UK Biobank is provided with relevant ethical input and oversight about the dissemination of participant data (which is all de-identified) from:

- the Access Sub Committee (ASC), which oversees all the data access applications to UK Biobank: the ASC is advised by its own ethical advisers on specific access applications, Ethox (https://www.ethox.ox.ac.uk/); and separately

- the Ethics Advisory Committee (EAC – https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/governance/ethics-advisory-committee), which is also a committee of the UK Biobank Board (and which superseded the original Ethics and Governance Council) and has a remit to provide advice to the Board on general ethical issues that arise during the maintenance, development and use of UK Biobank.

The data included in the proposed dissemination relates only to the consented cohort, rather than that of the general public. There is no risk of potential harm to the public by releasing the cohort’s data as described in the initial informed consent form and participant information leaflet.

Processing meets the requirement in point (g) of Article 9(2) of the General Data Protection Regulations for a basis in the law of the United Kingdom or a part of the United Kingdom. It also meets a condition in Part 2 of Schedule 1 of the Data Protection Act 2018.

- How the requested data achieves UK Biobank’s mission

UK Biobank’s mission is to enable scientific discoveries that improve human health. UK Biobank's operational objective is to combine extensive and precise measurement of exposures with detailed and rigorous follow-up for a wide range of health-related outcomes, and to promote innovative science by maximising secure research access to de-identified data.

Over the past two decades, UK Biobank has developed into a unique research resource, due to its very large size and the range and detail of participant-level data, including extensive phenotype and genotype information. As previously mentioned, it is now one of the most detailed and most-used research resources for health-related research in the world, enabling researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions that affect people in middle and older age, regardless of risk factors. The ability to link to participants’ health records from NHS England, in order to identify specific outcomes occurring during long-term participant follow-up to inform appropriate disease-based research, is critical and fundamental to UK Biobank's mission.

To date, approximately 34,000 researchers are registered with and approved by UK Biobank, from over 3,500 approved institutions, and over 4,000 health-related research applications have been approved to date.

- Background to the request

UK Biobank is a detailed, long-term prospective health research study that was established in 2003. Its core funding is received from Wellcome, the Medical Research Council, Cancer Research UK, the British Heart Foundation and NIHR. Further funding for specific enhancements is detailed at https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/about-us/our-funding. No funder is permitted to influence or suppress any outcomes of research undertaking using the UK Biobank resource. Since 2012 UK Biobank has been open to external access applications and for the great majority of this time (nearly a decade), UK Biobank has been approved to receive linked NHS England data to achieve the aims of the study.

Between 2006 and 2010, UK Biobank recruited 503,000 people from the general population – known as “participants” – then aged between 40 and 69 to take part in the study. Inevitably, some of the participants are now deceased and there are approximately 466,000 living participants who currently remain in the study and are the data subjects for the purpose of this application.

The methods used for the invitation and assessment of participants were developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All UK Biobank participants have given fully informed consent to UK Biobank for: access to and long-term storage of information from their medical and other health-related records; use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purposes; and re-contact by UK Biobank to invite them (but without any obligation) to answer further questions or to take part in further assessments.

It is considered very unlikely that participants - and to date UK Biobank has not received such an objection from any of its original 503,000 participants (now 466,000 living participants) since access commenced - would object to the processing activities involved in obtaining linked data from NHS England. Participants can withdraw from UK Biobank at any time (https://www.ukbiobank.ac.uk/media/fpoku2ac/withdrawal-protocol-030412-1.pdf), although since UK Biobank began recruitment in 2006 only a small number (under 1,000) have done so.

- Data required

Under Data Sharing Agreement, DARS-NIC-08472-V9SK version 15.4, UK Biobank currently receives updates on participants' health and health-related data from NHS England.

UK Biobank requests updates to the following datasets (for which it currently has approval under DARS-NIC-08472-V9SK version 15.4): Hospital Episode Statistics Critical Care and Hospital Episode Statistics Admitted Patient Care, Mental Health Services Data Set, Improved Access to Psychological Therapies Data Set, Civil Registrations of death, Cancer Registration data, National Diabetes Audit data, COVID-19 Vaccination Status (this dataset will only be used for COVID-19 research purposes and this restricted purpose is added explicitly to the terms of the Material Transfer Agreement), Personal Demographic Service (PDS) updates, which details changes to participants' demographics, postal addresses or registered GP practices and the new addition of the Medicines prescribed under primary care dataset.

Additionally, UK Biobank requests updates to the following datasets (for which it currently has approval under DARS-NIC-656745 version 0.2): NDRS Cancer Registrations, NDRS Systemic Anti-Cancer (SACT), NDRS National Radiotherapy Dataset (RTDS).

- Justification for the datasets requested

UK Biobank is a large-scale biomedical database and research resource, containing in-depth genetic and health information from half a million UK participants. It is now the most detailed, long-term prospective health research study in the world, facilitating health-related research that is in the public interest and enabling the international scientific community to better understand a range of common and life-threatening diseases, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke and mental health.

Under this iteration of the data sharing agreement, The UK Biobank now seek to include the Medicines prescribed under primary care data. The justification of this additional request is that the incorporation of prescription data into UK Biobank would facilitate research into the identification of disease occurrence in UK Biobank (e.g. defining who has diabetes through the use of metformin, insulin and other drugs); research that uses the genomic data to predict who will develop side effects prior to drug initiation (e.g. ACE-inhibitor-associated cough as assessed through GP data) and how genetic variation leads to variability in drug response, treatment duration and disease prognosis. It would also enable research into the genetic, clinical and lifestyle determinants of treatment-resistant disease (e.g., what are the characteristics of treatment-resistant depression).

Researchers will use a whole range of data (including the newly requested prescriptions dataset) included in UK Biobank to develop algorithms to define disease outcomes. These algorithmically-defined health outcomes will then be used in a wide range of research – e.g., that investigates the genetic, lifestyle and environmental determinants of disease, risk prediction and early detection of disease. As such, the Medicines prescribed under primary care data complements many other data sets already received/held by UK Biobank and will inform further on lifestyle determinants, disease risk factors and treatment outcomes.

All of the clinical datasets requested are essential components of the UK Biobank resource and enable epidemiological research into antecedent clinical events prior to diagnosis and the effects of multi-morbidity on future health outcomes. Without linked data from NHS England, researchers would not be able to ascertain the full burden of disease in the cohort nor the effects of particular treatment and management pathways on health. In order for prospective research to track health and the development of disease over time, longitudinal data must be updated with the latest-available healthcare records for the cohort.

Of key importance are datasets related to death, cancer registry notifications and hospital activity, as they provide crucial information about which participants have developed which diseases and at what time. Without access to these data, UK Biobank could not function as a prospective cohort study and therefore research on the associations between social or economic determinants of health and genetic factors and the corresponding risk of developing diseases over time would not be possible.

Linkage to the requested NHSE datasets facilitates research which draws upon data relating to: attendance at hospital for any reason, including hospitalisation for cancer care and diabetes; care received by participants who are in contact with mental health and learning disability services; and cause of death.

Data from PDS are required to enable UK Biobank to maintain accurate information about participants’ contact details and GP practice in order to enable direct communication with GPs. For example, for the purpose of the UK Biobank imaging assessment study (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study), which is by some distance the largest (by an order of magnitude) imaging study in the world to date. To expand, in the event that during the imaging assessment visit a participant has been flagged by a radiographer with a potentially serious incidental finding and this potentially serious finding has been confirmed by an expert radiologist (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study/incidental-findings) then UK Biobank contacts both the participant and their GP. These data also facilitate participant follow up which, in a cohort study, needs to be as complete as possible, to avoid both unnecessary reduction in cohort numbers (which reduces statistical power of data analyses) and to minimise loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on periodic re-contact with UK Biobank participants to invite them to complete brief questionnaires about health-related exposures or outcomes, or to other enhancement projects (such as to attend an imaging assessment), for which correct participant contact details are required. UK Biobank is also committed to ensuring that participants are informed of progress and developments. UK Biobank does this in part through its annual newsletter, sent via e-mail and/or postal mail, for which up-to-date contact details are required. For obvious reasons, UK Biobank prefers to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress. Although participants are encouraged to inform UK Biobank when they change address, this is not a failsafe way of ensuring that the most up-to-date contact information is held on all participants.

- Justification for the level of data (identifying, pseudonymised, unsuppressed aggregated)

Pseudonymised data are required in order to correctly link updated NHS records to the existing data held for individual participants within UK Biobank’s database. The process of pseudonymising UK Biobank participants has already been undertaken, and NHS England and UK Biobank both hold a set of pseudo IDs that can be used for linking health data to the correct individuals, meaning no identifying data is needed for this purpose. This is essential to creating and maintaining a longitudinal dataset. UK Biobank’s default position is not to make available to approved researchers any direct and indirect identifiers (such as full postcode, date of birth, name, NHS number, etc. - https://www.ukbiobank.ac.uk/media/3kmh2t1z/de-identification-protocol-v-2-1-29-06-22.pdf). Each approved research project is instead provided with a unique encrypted 7-digit identification number for participants. Hence, it should not be possible for a researcher to identify a UK Biobank participant using the data that UK Biobank releases to them. However, noting the rich source of data the sub-licensees may be able to access, in the unlikely event sub-licensees may in be able to re-identify a data subject, sufficient contractual controls are in place to prohibit this.

- Justification for the number of years requested

UK Biobank has received data from NHS England (previously NHS Digital) for nearly a decade and all of the datasets requested in this application have previously been approved and provided to UK Biobank under Data Sharing Agreement, DARS-NIC-08472-V9SK version 15.4 (and DARS-NIC-656745 - ODR_2014_095 for NDRS data). UK Biobank is requesting the latest available data so that this can be added to existing historical data held on each UK Biobank participant, in order to enrich the long-term picture of participants’ health and enable exploration of multiple risk factors which might contribute to the development of a wide range of diseases over time.

- Justification for the geographical spread of the data requested

UK Biobank participants were recruited from across Great Britain – there was no recruitment from Northern Ireland - and participants still reside in all three nations (although there will have inevitably been some movement within and between countries since recruitment).

- Confirmation there are no alternative, less intrusive ways of achieving the purpose

As a very large cohort, it is not feasible for UK Biobank itself to follow up regularly with each of its participants on an individual basis and even if this were possible it would not generate the necessary clinical detail (or anything approaching it) that is contained within the health records. Further, self-reported data do not possess the needed coverage, reliability and consistency that health records contain. The data provided via NHS England - as the primary mechanism for follow-up of the cohort - are crucial to the viability of UK Biobank, particularly as they provide information on the ~466,000 living participants in an entirely unobtrusive manner.

- Data minimisation

For each of these datasets listed above, UK Biobank requires data only for the ~466,000 living UK Biobank participants. It should be noted that UK Biobank is only requesting individual-level identifiers to ensure that the linkage is implemented correctly. UK Biobank regularly informs NHSE of updates to the cohort to ensure that data is not received for participants who have withdrawn from the study.

- Data controller

The sole Data Controller is UK Biobank Limited.

- Data processors

In relation to the data supplied by NHSE there are two appointed data processors and one sub-processor:

1. The University of Oxford is the primary store (and reference copy) of the data that form the UK Biobank resource (including the data supplied by NHSE). The University of Oxford, as a data processor (subject to their agreement with UK Biobank) act only on the instructions of UK Biobank and are required to store participant health data (such as the health data supplied by NHSE) in a secure environment and separately to any identifiers (such as name or NHS number).

2. DNAnexus Inc. operate UK Biobank’s Research Analysis Platform which enables researchers to access the de-identified health data for the purposes of research. DNAnexus act only on the instructions of UK Biobank, as the data processor (again subject to an agreement with UK Biobank) and do not have access to any identifiable participant data.

3. The de-identified data made available through the Research Analysis Platform, is stored in secure cloud storage provided by Amazon Web Services (AWS) and hosted on the AWS node in London. AWS is an approved and contracted sub-processor of DNAnexus. NHSE have requested that AWS are included in the itemised list of “processors”.

-Research Analysis Platform (RAP)

The online Research Analysis Platform (RAP) enables researchers to analyse UK Biobank data in-situ (i.e. without having to download the data directly onto their own computers). Access to the whole exome- and genome sequencing data is only possible via the RAP (owing to the sheer size of the data). Currently, about 20% of UK Biobank projects are performed using the RAP, it’s anticipated this will increase over time given the imminent availability of whole-genome sequencing data for all 500,000 participants. In addition, early-career researchers and researchers from low- and middle-income countries are eligible for research credits to cover their compute costs, which will further increase uptake of the RAP. All data processing of linked dataset from NHSE is currently performed at the University of Oxford, although work to migrate to secured, cloud-based infrastructure is underway and will be subject to both ISO27001 and DSPT accreditation once implemented. Future planned data releases are detailed on the UK Biobank website (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/future-data-release-timelines).

- Funders

While UK Biobank receives core funding from Wellcome, the Medical Research Council, Cancer Research UK, the British Heart Foundation and the National Institute for Health Research as major funders, these organisations have no access to the data held by UK Biobank and exercise no control or influence in respect of its use.

UK Biobank charges researchers a fee for accessing data and thus is deemed to include a commercial purpose. The fee charged by UK Biobank is on a cost recovery basis and thus covers the incremental costs of servicing an access application. Commercial users of data would be applying for sub licenses, any commercial use or applications that raise particular concerns (e.g. involve potentially contentious uses of the resource) is addressed via the UK Biobank Access Sub Committee who conduct a set of standard checks and seeks advice on particular applications as required. UK Biobank does not seek to recover the cost of development and management of the UK Biobank resource (which already runs into hundreds of millions of pounds). UK Biobank is accessible by all researchers - whether they are from academia or industry. UK Biobank approves applications only where provision of data is for the promotion of health (including epidemiological research), is consistent with participants' consent and the UK Biobank's Ethics and Governance Framework, and is consistent with its access criteria.

The national data opt-out does not apply where explicit consent has been obtained from the patient for the specific purpose.

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

- Research applications and approved research projects

UK Biobank has developed one of the most detailed and most-used research resource for health-related research in the world, and enabled thousands of researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions. As a prospective resource, its success relies heavily on continued linkage to longitudinal health records from NHS England.

Over the past decade an increasingly large number of researchers have conducted and reported research on an extensive range of topics. The breadth of health conditions that it will be possible to study reliably using even more detailed information about exposures will increase further as the numbers of incident cases of different health outcomes amongst the ageing cohort increase during the next five to ten years. There are currently over 4,000 approved research applications, a summary of which is available on UK Biobank’s website at https://www.ukbiobank.ac.uk/enable-your-research/approved-research.

Over coming years, UK Biobank will continue to enhance and improve the resource, not only through continued linkage to health-related records, but also through other enhancements to the resource. This will enable further research, leading to substantial patient and NHS benefit.

- Imaging (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study): UK Biobank's imaging study, which will run for the next five to six years, started in 2014 with the aim of collecting multi-modal MRI data on up to 100,000 UK Biobank participants. In 2022, it was announced that funding from the MRC, Calico Life Sciences and the Chan Zuckerberg Initiative will enable UK Biobank to perform up to 60,000 participants a few years later, creating the world’s largest longitudinal imaging dataset to assess disease progression. This dataset will be particularly valuable for researchers studying diseases such as dementia, as it will enable research into how early (i.e. pre-symptomatic) changes in the structure and function of the brain over time are related to dementia risk.

- Genetics (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/genetic-data): to complement the genome-wide genotyping data, in 2022, UK Biobank released whole-exome sequencing data to the global research community, creating the world’s largest exome (all the protein-encoding genes) dataset. Later in 2023 the whole genome sequence data will be released, creating the biggest whole genome dataset available in the world. This will transform the way in which scientists study the genetics determinants of a wide range of health outcomes. (https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/news/final-data-release-from-the-world-s-largest-whole-exome-sequencing-project).

- Assays (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/biomarker-data): UK Biobank is also undertaking further assays – proteomics, metabolomics, transcriptomics – which enable the disease pathway to be followed, connecting the phenotypic and the genetic data within UK Biobank. A first release of proteomic data took place in early 2023 and a second tranche of metabolomics data will be made available to researchers in late 2023.

- Health record data (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/health-related-outcomes-data): As mentioned in section 5a linkage to health-related information is vital for researchers using the UK Biobank resource. UK Biobank will maintain its access to the health record data and it will seek to add other linkage data, particularly including (when available) the primary care data, socio-economic data and data to enhance cancer research.

To date, approximately 34,000 researchers have successfully registered to use the UK Biobank resource from across 3,500 research institutions worldwide. There are now over 4,000 approved research applications that that aim to use UK Biobank data for health-related research that is in the public interest. Based on the growth achieved to date, this is expected to increase by approximately 1,000, to a total of 5,000, over the next twelve months. The majority of applications are still led by academic researchers, but an increasing number are submitted from commercial entities. Summaries of all approved research projects that are currently underway can be found at https://www.ukbiobank.ac.uk/approved-research/. The nature of the research that is currently being conducted reflects the increasing breadth and depth of data that are now available, such as whole-exome and whole-genome sequence data, biochemical assay data and metabolomics data, as well as detailed imaging measures, and much larger numbers of incident cases of many different conditions identified from the NHS England linked healthcare data.

- Outputs from the data processing

Applications to use UK Biobank data are able to generate more than one publication/output. To date, over 10,000 publications have been produced, and the number of papers published each year is increasing rapidly (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/health-related-outcomes-data - note that this webpage is currently awaiting a refresh of data for 2022 and 2023). UK Biobank expects to see a continued increase in the number of research publications based upon analysis of UK Biobank data over the coming years.

- Data contained in the outputs

The resource contains individual-level data with associated pseudo IDs. Every download available to approved researchers has a unique set of randomly-generated pseudo IDs. Summary data derived from UK Biobank data are used in an increasing number of journal articles every year but researchers are not permitted to publish individual-level data. UK Biobank is currently developing guidance about the publication of results about very rare diseases which will be published soon.

- Approach to dissemination and communication

Researchers who use the UK Biobank resource are responsible for ensuring that the findings of their research are published and disseminated. Under the terms of UK Biobank’s Access procedures and Material Transfer Agreement, they are required to publish the findings of their research and return their results to UK Biobank so they can be made available for request by other researchers. In 2022, over 2,000 publications referenced UK Biobank data.

UK Biobank disseminates huge amounts of information to participants and researchers (plus the general public) via its website and other channels.

UK Biobank makes its governance, history, use, plans, and news available transparently through its website. The website also provides a broad range of information about the resource and its outputs, including hosting lists of approved research (https://www.ukbiobank.ac.uk/enable-your-research/approved-research), publications (https://www.ukbiobank.ac.uk/enable-your-research/publications), as well as more detailed case studies of how UK Biobank is helping scientists and researchers to evolve our understanding of human health and the most common life-threatening diseases (https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/our-impact).

It also has a Twitter feed (https://twitter.com/uk_biobank/) and YouTube channel (https://www.youtube.com/@uk_biobank), releases newsletters annually to participants (e.g. https://www.ukbiobank.ac.uk/explore-your-participation/stay-involved/2022-newsletter), hosts interactive events with participants at which enhancement projects and the results of research using the resource are discussed (see https://youtu.be/R1aHLOS1qoY for a recording of a recent participant event held in Newcastle upon Tyne), organises an annual Scientific Conference attended by an international delegation (see https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/winter-scientific-conference-2022 for details of the conference held in December 2022), and engages with industry partners. UK Biobank presented details of its work at the following external conferences in the past year: the European Society of Human Genetics, the American Society of Human Genetics, European Society of Cardiology, Bio-IT world conference, and the Festival of Genomics.

Participants can contact the Participant Resource Centre with questions, or to provide feedback. The Centre answers over 20,000 calls and around 35,000 emails from participants annually. Feedback obtained about enhancement studies, either via the PRC or other feedback routes is relayed to project teams and appropriate actions taken to address any issues of concern raised. Participants who make enquiries about international research all receive a written response from UK Biobank and are offered the opportunity to speak directly to UK Biobank staff, including the Principal Investigator.

UK Biobank has recently appointed a Head of Engagement part of whose role it will be to ensure that the way in which UK Biobank communicates with its participants (both in terms of engaging them in specific UK Biobank activities and more generally) and researchers is as effective as it can be. Examples of forthcoming activities include questionnaires and focus groups about the materials used for the UK Biobank imaging study, as well as focus groups about possible new data linkages and new questionnaires. Furthermore, the UK Biobank Ethics Advisory Committee leads a programme of research which includes consulting participant views about relevant issues.

- Target dates for outputs

Publications have been generated using the resource for over a decade and increase in number year on year (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/health-related-outcomes-data - note that this webpage is currently awaiting a refresh of data for 2022 and 2023). Researchers are expected to publish their findings within six months of the completion of their project, or three years following approval to access the data. This is stipulated in the MTA. Additionally, within six months of publication, results must be provided to UK Biobank so that they may be incorporated into the resource where relevant, for the benefit of others.

To date no specific metrics relating to targets have been shared with NHS England. However, UK Biobank is now a major international research resource for health-related research and is, by some distance, the most widely used and published upon health research resource in the world. It enables thousands of researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions, leading to the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses and the number of researchers, applications and publications resulting from the resource continues to grow.

- Yielded benefits

Of the numerous and wide-ranging benefits yielded to date (also see https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/our-impact), UK Biobank's contribution to the national and international response to the recent COVID-19 pandemic presents a good case study for the broad value and benefit of the resource. The existence of this large, detailed, prospective cohort with high degrees of participant engagement enabled rapid repurposing in 2020 for COVID-19 research. The frequency of hospitalisation and death data updates from NHS England was immediately increased, and linkage was extended to SARS-CoV-2 testing data and to primary care data under emergency legislation (obtained separately). Much of this work has already translated into policy impact during the pandemic. For example, ~450 publications on COVID-19 have been produced using UK Biobank data. These include: studies of the genetic determinants of severe COVID-19 (Pairo-Castineria et al., 2021); relationships of pre-existing adverse cardiac phenotypes with COVID-19, determined from imaging scans (Raisi-Estabragh et al., 2021); and associations of NMR-metabolomic biomarkers for low-grade inflammation and cardiometabolic disease with severe COVID-19 (Julkunen et al., 2021).

The UK Biobank resource is enabling unprecedented worldwide genomics research and is at the heart of the UK Government's strategy for incorporating genomics in healthcare (HM Government, 2020). This strategy document recognises UK Biobank's central role in the development of the polygenic risk score (PRS) concept, which involves using the combined effects of thousands of genetic variants (each of which has only a small effect) to improve risk prediction for common diseases. Using UK Biobank data, researchers from the Broad Institute were the first to show that polygenic risk scores (PRS) – which combine the effects of very large numbers of genetic variant – could identify people at a high risk of a particular disease (often equivalent to that of a single-gene disorder). In particular, they found that about 5% of the population had a PRS which put them at equivalent risk of developing heart disease to someone who had Familial Hypercholesterolaemia (Khera et al., 2018). Other research using the whole exome sequencing data found that rare genetic mutations in the GPR75 gene (which encodes a receptor in the brain that regulates energy balance) are associated with having half the risk of developing obesity. Protective ‘loss of function’ mutations were found in about one of every 3,000 people sequenced. Heralded as a “genetic superpower”. This discovery will directly lead to approaches (such as gene-editing and gene-silencing) to develop drugs that mimic the function of this mutation (Akbari, et al., 2021).

UK Biobank’s imaging study enables work which provides an excellent example of how a single study can be used to improve understanding of disease. Researchers from the University of Oxford carried out genome association studies of functional and structural brain imaging phenotypes and showed that many of the phenotypes are heritable. The results provide insight into the genetic architecture of the brain that are of relevance to brain development, ageing, dementia and other brain disorders (Elliott, et al., 2018).

- References

Akbari P, et al. Sequencing of 640,000 exomes identifies GPR75 variants associated with protection from obesity Science, 2021: 373 (6550)

Elliott, L.T., Sharp, K., Alfaro-Almagro, F. et al. Genome-wide association studies of brain imaging phenotypes in UK Biobank. Nature 562, 210–216 (2018).

HM Government. Genome UK: The future of healthcare. 2020

Julkunen H, et al. Metabolic biomarker profiling for identification of susceptibility to severe pneumonia and COVID-19 in the general population. Elife. 2021; 10: e63033

Khera AV, et al. Genome-wide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. Nat Genet. 2018: 50; 1219-24

Pairo-Castineira, E., et al. Genetic mechanisms of critical illness in COVID-19. Nature; 2021; 591, 92–98.

Raisi-Estabragh Z, et al. Adverse cardiovascular magnetic resonance phenotypes are associated with greater likelihood of incident coronavirus disease 2019: findings from the UK Biobank. Aging Clin Exp Res; 2021; 33: 1133-1144

- How the outputs will achieve the stated purpose

The UK Biobank resource enables researchers to make scientific discoveries that provide considerable benefit to human health, not just for people living in the UK but worldwide. Much of this work relies upon data made available by NHS England. In relation to health outcomes, there are over 2,000 published papers on cardiovascular disease, over 1,400 on cancer, over 500 on mental health, and over 900 on obesity. Other broader research projects have published on the identification of novel patterns of association of many different exposures with a range of health outcomes. The rapid availability of more frequent healthcare records for COVID-19 research has already led to over 450 papers. A list of publications using UKB data can be accessed at https://www.ukbiobank.ac.uk/enable-your-research/publications (note that this webpage is currently awaiting a refresh of data for 2022 and 2023). These publications will improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses by contributing to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers and the development of new and improved treatments for patients with existing conditions, all of which are of considerable benefit to the public.

- Purpose of this application

UK Biobank is the most detailed, long-term prospective health research study in the world, with a specific objective of following up the health of its participants through access to their health records. The resource, which is now one of the most-used research resource for health-related research, enables researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions. This in turn will improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses by contributing to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers (e.g., polygenic risk scores or imaging biomarkers of internal organs) and the development of new and improved treatments for patients with existing conditions, all of which are of considerable benefit to the public. NHS England health record data are essential for UK Biobank to achieve its aims as a prospective resource enabling researchers to follow up participants and establish risk factors and causality of disease (otherwise it is effectively an observational study).

- Explanation of Article 6 justification of processing the data

UK Biobank requires data from NHS England to pursue its legitimate interests, which are to set up and manage a major international research resource for health-related research in the public interest. UK Biobank's Article 6 lawful basis for processing the personal data relating to the participants is therefore Article 6 (1) (f) of the GDPR (legitimate interest).

- Public interest for processing

UK Biobank processes special category personal data provided by NHS England in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. As a major research resource for health-related research, UK Biobank falls clearly within this category as its purpose is to enable scientific discoveries that improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses. The data are required for research purposes in the public interest - meeting the conditions on the DPA Schedule 1 Part 1 (4) - which GDPR Recital 52 (2) determines is an appropriate derogation from prohibition on processing special categories of personal data; proportionate to the aim pursued (as above). The ways in which the processing of data will be of benefit to the public, thereby demonstrating that the processing is in the public interest, are described in section 5d.

In the design of the UK Biobank resource, the moral and ethical issues were specifically considered, both internally and externally, and are set out in the Ethics and Governance Framework (https://www.ukbiobank.ac.uk/media/0xsbmfmw/egf.pdf). UK Biobank is provided with relevant ethical input and oversight about the dissemination of participant data (which is all de-identified) from:

- the Access Sub Committee (ASC), which oversees all the data access applications to UK Biobank: the ASC is advised by its own ethical advisers on specific access applications, Ethox (https://www.ethox.ox.ac.uk/); and separately

- the Ethics Advisory Committee (EAC – https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/governance/ethics-advisory-committee), which is also a committee of the UK Biobank Board (and which superseded the original Ethics and Governance Council) and has a remit to provide advice to the Board on general ethical issues that arise during the maintenance, development and use of UK Biobank.

The data included in the proposed dissemination relates only to the consented cohort, rather than that of the general public. There is no risk of potential harm to the public by releasing the cohort’s data as described in the initial informed consent form and participant information leaflet.

Processing meets the requirement in point (g) of Article 9(2) of the General Data Protection Regulations for a basis in the law of the United Kingdom or a part of the United Kingdom. It also meets a condition in Part 2 of Schedule 1 of the Data Protection Act 2018.

- How the requested data achieves UK Biobank’s mission

UK Biobank’s mission is to enable scientific discoveries that improve human health. UK Biobank's operational objective is to combine extensive and precise measurement of exposures with detailed and rigorous follow-up for a wide range of health-related outcomes, and to promote innovative science by maximising secure research access to de-identified data.

Over the past two decades, UK Biobank has developed into a unique research resource, due to its very large size and the range and detail of participant-level data, including extensive phenotype and genotype information. As previously mentioned, it is now one of the most detailed and most-used research resources for health-related research in the world, enabling researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions that affect people in middle and older age, regardless of risk factors. The ability to link to participants’ health records from NHS England, in order to identify specific outcomes occurring during long-term participant follow-up to inform appropriate disease-based research, is critical and fundamental to UK Biobank's mission.

To date, approximately 34,000 researchers are registered with and approved by UK Biobank, from over 3,500 approved institutions, and over 4,000 health-related research applications have been approved to date.

- Background to the request

UK Biobank is a detailed, long-term prospective health research study that was established in 2003. Its core funding is received from Wellcome, the Medical Research Council, Cancer Research UK, the British Heart Foundation and NIHR. Further funding for specific enhancements is detailed at https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/about-us/our-funding. No funder is permitted to influence or suppress any outcomes of research undertaking using the UK Biobank resource. Since 2012 UK Biobank has been open to external access applications and for the great majority of this time (nearly a decade), UK Biobank has been approved to receive linked NHS England data to achieve the aims of the study.

Between 2006 and 2010, UK Biobank recruited 503,000 people from the general population – known as “participants” – then aged between 40 and 69 to take part in the study. Inevitably, some of the participants are now deceased and there are approximately 466,000 living participants who currently remain in the study and are the data subjects for the purpose of this application.

The methods used for the invitation and assessment of participants were developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All UK Biobank participants have given fully informed consent to UK Biobank for: access to and long-term storage of information from their medical and other health-related records; use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purposes; and re-contact by UK Biobank to invite them (but without any obligation) to answer further questions or to take part in further assessments.

It is considered very unlikely that participants - and to date UK Biobank has not received such an objection from any of its original 503,000 participants (now 466,000 living participants) since access commenced - would object to the processing activities involved in obtaining linked data from NHS England. Participants can withdraw from UK Biobank at any time (https://www.ukbiobank.ac.uk/media/fpoku2ac/withdrawal-protocol-030412-1.pdf), although since UK Biobank began recruitment in 2006 only a small number (under 1,000) have done so.

- Data required

Under Data Sharing Agreement, DARS-NIC-08472-V9SK version 15.4, UK Biobank currently receives updates on participants' health and health-related data from NHS England.

UK Biobank requests updates to the following datasets (for which it currently has approval under DARS-NIC-08472-V9SK version 15.4): Hospital Episode Statistics Critical Care and Hospital Episode Statistics Admitted Patient Care, Mental Health Services Data Set, Improved Access to Psychological Therapies Data Set, Civil Registrations of death, Cancer Registration data, National Diabetes Audit data, COVID-19 Vaccination Status (this dataset will only be used for COVID-19 research purposes and this restricted purpose is added explicitly to the terms of the Material Transfer Agreement), Personal Demographic Service (PDS) updates, which details changes to participants' demographics, postal addresses or registered GP practices and the new addition of the Medicines prescribed under primary care dataset.

Additionally, UK Biobank requests updates to the following datasets (for which it currently has approval under DARS-NIC-656745 version 0.2): NDRS Cancer Registrations, NDRS Systemic Anti-Cancer (SACT), NDRS National Radiotherapy Dataset (RTDS).

- Justification for the datasets requested

UK Biobank is a large-scale biomedical database and research resource, containing in-depth genetic and health information from half a million UK participants. It is now the most detailed, long-term prospective health research study in the world, facilitating health-related research that is in the public interest and enabling the international scientific community to better understand a range of common and life-threatening diseases, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke and mental health.

Under this iteration of the data sharing agreement, The UK Biobank now seek to include the Medicines prescribed under primary care data. The justification of this additional request is that the incorporation of prescription data into UK Biobank would facilitate research into the identification of disease occurrence in UK Biobank (e.g. defining who has diabetes through the use of metformin, insulin and other drugs); research that uses the genomic data to predict who will develop side effects prior to drug initiation (e.g. ACE-inhibitor-associated cough as assessed through GP data) and how genetic variation leads to variability in drug response, treatment duration and disease prognosis. It would also enable research into the genetic, clinical and lifestyle determinants of treatment-resistant disease (e.g., what are the characteristics of treatment-resistant depression).

Researchers will use a whole range of data (including the newly requested prescriptions dataset) included in UK Biobank to develop algorithms to define disease outcomes. These algorithmically-defined health outcomes will then be used in a wide range of research – e.g., that investigates the genetic, lifestyle and environmental determinants of disease, risk prediction and early detection of disease. As such, the Medicines prescribed under primary care data complements many other data sets already received/held by UK Biobank and will inform further on lifestyle determinants, disease risk factors and treatment outcomes.

All of the clinical datasets requested are essential components of the UK Biobank resource and enable epidemiological research into antecedent clinical events prior to diagnosis and the effects of multi-morbidity on future health outcomes. Without linked data from NHS England, researchers would not be able to ascertain the full burden of disease in the cohort nor the effects of particular treatment and management pathways on health. In order for prospective research to track health and the development of disease over time, longitudinal data must be updated with the latest-available healthcare records for the cohort.

Of key importance are datasets related to death, cancer registry notifications and hospital activity, as they provide crucial information about which participants have developed which diseases and at what time. Without access to these data, UK Biobank could not function as a prospective cohort study and therefore research on the associations between social or economic determinants of health and genetic factors and the corresponding risk of developing diseases over time would not be possible.

Linkage to the requested NHSE datasets facilitates research which draws upon data relating to: attendance at hospital for any reason, including hospitalisation for cancer care and diabetes; care received by participants who are in contact with mental health and learning disability services; and cause of death.

Data from PDS are required to enable UK Biobank to maintain accurate information about participants’ contact details and GP practice in order to enable direct communication with GPs. For example, for the purpose of the UK Biobank imaging assessment study (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study), which is by some distance the largest (by an order of magnitude) imaging study in the world to date. To expand, in the event that during the imaging assessment visit a participant has been flagged by a radiographer with a potentially serious incidental finding and this potentially serious finding has been confirmed by an expert radiologist (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study/incidental-findings) then UK Biobank contacts both the participant and their GP. These data also facilitate participant follow up which, in a cohort study, needs to be as complete as possible, to avoid both unnecessary reduction in cohort numbers (which reduces statistical power of data analyses) and to minimise loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on periodic re-contact with UK Biobank participants to invite them to complete brief questionnaires about health-related exposures or outcomes, or to other enhancement projects (such as to attend an imaging assessment), for which correct participant contact details are required. UK Biobank is also committed to ensuring that participants are informed of progress and developments. UK Biobank does this in part through its annual newsletter, sent via e-mail and/or postal mail, for which up-to-date contact details are required. For obvious reasons, UK Biobank prefers to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress. Although participants are encouraged to inform UK Biobank when they change address, this is not a failsafe way of ensuring that the most up-to-date contact information is held on all participants.

- Justification for the level of data (identifying, pseudonymised, unsuppressed aggregated)

Pseudonymised data are required in order to correctly link updated NHS records to the existing data held for individual participants within UK Biobank’s database. The process of pseudonymising UK Biobank participants has already been undertaken, and NHS England and UK Biobank both hold a set of pseudo IDs that can be used for linking health data to the correct individuals, meaning no identifying data is needed for this purpose. This is essential to creating and maintaining a longitudinal dataset. UK Biobank’s default position is not to make available to approved researchers any direct and indirect identifiers (such as full postcode, date of birth, name, NHS number, etc. - https://www.ukbiobank.ac.uk/media/3kmh2t1z/de-identification-protocol-v-2-1-29-06-22.pdf). Each approved research project is instead provided with a unique encrypted 7-digit identification number for participants. Hence, it should not be possible for a researcher to identify a UK Biobank participant using the data that UK Biobank releases to them. However, noting the rich source of data the sub-licensees may be able to access, in the unlikely event sub-licensees may in be able to re-identify a data subject, sufficient contractual controls are in place to prohibit this.

- Justification for the number of years requested

UK Biobank has received data from NHS England (previously NHS Digital) for nearly a decade and all of the datasets requested in this application have previously been approved and provided to UK Biobank under Data Sharing Agreement, DARS-NIC-08472-V9SK version 15.4 (and DARS-NIC-656745 - ODR_2014_095 for NDRS data). UK Biobank is requesting the latest available data so that this can be added to existing historical data held on each UK Biobank participant, in order to enrich the long-term picture of participants’ health and enable exploration of multiple risk factors which might contribute to the development of a wide range of diseases over time.

- Justification for the geographical spread of the data requested

UK Biobank participants were recruited from across Great Britain – there was no recruitment from Northern Ireland - and participants still reside in all three nations (although there will have inevitably been some movement within and between countries since recruitment).

- Confirmation there are no alternative, less intrusive ways of achieving the purpose

As a very large cohort, it is not feasible for UK Biobank itself to follow up regularly with each of its participants on an individual basis and even if this were possible it would not generate the necessary clinical detail (or anything approaching it) that is contained within the health records. Further, self-reported data do not possess the needed coverage, reliability and consistency that health records contain. The data provided via NHS England - as the primary mechanism for follow-up of the cohort - are crucial to the viability of UK Biobank, particularly as they provide information on the ~466,000 living participants in an entirely unobtrusive manner.

- Data minimisation

For each of these datasets listed above, UK Biobank requires data only for the ~466,000 living UK Biobank participants. It should be noted that UK Biobank is only requesting individual-level identifiers to ensure that the linkage is implemented correctly. UK Biobank regularly informs NHSE of updates to the cohort to ensure that data is not received for participants who have withdrawn from the study.

- Data controller

The sole Data Controller is UK Biobank Limited.

- Data processors

In relation to the data supplied by NHSE there are two appointed data processors and one sub-processor:

1. The University of Oxford is the primary store (and reference copy) of the data that form the UK Biobank resource (including the data supplied by NHSE). The University of Oxford, as a data processor (subject to their agreement with UK Biobank) act only on the instructions of UK Biobank and are required to store participant health data (such as the health data supplied by NHSE) in a secure environment and separately to any identifiers (such as name or NHS number).

2. DNAnexus Inc. operate UK Biobank’s Research Analysis Platform which enables researchers to access the de-identified health data for the purposes of research. DNAnexus act only on the instructions of UK Biobank, as the data processor (again subject to an agreement with UK Biobank) and do not have access to any identifiable participant data.

3. The de-identified data made available through the Research Analysis Platform, is stored in secure cloud storage provided by Amazon Web Services (AWS) and hosted on the AWS node in London. AWS is an approved and contracted sub-processor of DNAnexus. NHSE have requested that AWS are included in the itemised list of “processors”.

-Research Analysis Platform (RAP)

The online Research Analysis Platform (RAP) enables researchers to analyse UK Biobank data in-situ (i.e. without having to download the data directly onto their own computers). Access to the whole exome- and genome sequencing data is only possible via the RAP (owing to the sheer size of the data). Currently, about 20% of UK Biobank projects are performed using the RAP, it’s anticipated this will increase over time given the imminent availability of whole-genome sequencing data for all 500,000 participants. In addition, early-career researchers and researchers from low- and middle-income countries are eligible for research credits to cover their compute costs, which will further increase uptake of the RAP. All data processing of linked dataset from NHSE is currently performed at the University of Oxford, although work to migrate to secured, cloud-based infrastructure is underway and will be subject to both ISO27001 and DSPT accreditation once implemented. Future planned data releases are detailed on the UK Biobank website (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/future-data-release-timelines).

- Funders

While UK Biobank receives core funding from Wellcome, the Medical Research Council, Cancer Research UK, the British Heart Foundation and the National Institute for Health Research as major funders, these organisations have no access to the data held by UK Biobank and exercise no control or influence in respect of its use.

UK Biobank charges researchers a fee for accessing data and thus is deemed to include a commercial purpose. The fee charged by UK Biobank is on a cost recovery basis and thus covers the incremental costs of servicing an access application. Commercial users of data would be applying for sub licenses, any commercial use or applications that raise particular concerns (e.g. involve potentially contentious uses of the resource) is addressed via the UK Biobank Access Sub Committee who conduct a set of standard checks and seeks advice on particular applications as required. UK Biobank does not seek to recover the cost of development and management of the UK Biobank resource (which already runs into hundreds of millions of pounds). UK Biobank is accessible by all researchers - whether they are from academia or industry. UK Biobank approves applications only where provision of data is for the promotion of health (including epidemiological research), is consistent with participants' consent and the UK Biobank's Ethics and Governance Framework, and is consistent with its access criteria.

The national data opt-out does not apply where explicit consent has been obtained from the patient for the specific purpose.

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

- Research applications and approved research projects

UK Biobank has developed one of the most detailed and most-used research resource for health-related research in the world, and enabled thousands of researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions. As a prospective resource, its success relies heavily on continued linkage to longitudinal health records from NHS England.

Over the past decade an increasingly large number of researchers have conducted and reported research on an extensive range of topics. The breadth of health conditions that it will be possible to study reliably using even more detailed information about exposures will increase further as the numbers of incident cases of different health outcomes amongst the ageing cohort increase during the next five to ten years. There are currently over 4,000 approved research applications, a summary of which is available on UK Biobank’s website at https://www.ukbiobank.ac.uk/enable-your-research/approved-research.

Over coming years, UK Biobank will continue to enhance and improve the resource, not only through continued linkage to health-related records, but also through other enhancements to the resource. This will enable further research, leading to substantial patient and NHS benefit.

- Imaging (https://www.ukbiobank.ac.uk/explore-your-participation/contribute-further/imaging-study): UK Biobank's imaging study, which will run for the next five to six years, started in 2014 with the aim of collecting multi-modal MRI data on up to 100,000 UK Biobank participants. In 2022, it was announced that funding from the MRC, Calico Life Sciences and the Chan Zuckerberg Initiative will enable UK Biobank to perform up to 60,000 participants a few years later, creating the world’s largest longitudinal imaging dataset to assess disease progression. This dataset will be particularly valuable for researchers studying diseases such as dementia, as it will enable research into how early (i.e. pre-symptomatic) changes in the structure and function of the brain over time are related to dementia risk.

- Genetics (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/genetic-data): to complement the genome-wide genotyping data, in 2022, UK Biobank released whole-exome sequencing data to the global research community, creating the world’s largest exome (all the protein-encoding genes) dataset. Later in 2023 the whole genome sequence data will be released, creating the biggest whole genome dataset available in the world. This will transform the way in which scientists study the genetics determinants of a wide range of health outcomes. (https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/news/final-data-release-from-the-world-s-largest-whole-exome-sequencing-project).

- Assays (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/biomarker-data): UK Biobank is also undertaking further assays – proteomics, metabolomics, transcriptomics – which enable the disease pathway to be followed, connecting the phenotypic and the genetic data within UK Biobank. A first release of proteomic data took place in early 2023 and a second tranche of metabolomics data will be made available to researchers in late 2023.

- Health record data (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/health-related-outcomes-data): As mentioned in section 5a linkage to health-related information is vital for researchers using the UK Biobank resource. UK Biobank will maintain its access to the health record data and it will seek to add other linkage data, particularly including (when available) the primary care data, socio-economic data and data to enhance cancer research.

To date, approximately 34,000 researchers have successfully registered to use the UK Biobank resource from across 3,500 research institutions worldwide. There are now over 4,000 approved research applications that that aim to use UK Biobank data for health-related research that is in the public interest. Based on the growth achieved to date, this is expected to increase by approximately 1,000, to a total of 5,000, over the next twelve months. The majority of applications are still led by academic researchers, but an increasing number are submitted from commercial entities. Summaries of all approved research projects that are currently underway can be found at https://www.ukbiobank.ac.uk/approved-research/. The nature of the research that is currently being conducted reflects the increasing breadth and depth of data that are now available, such as whole-exome and whole-genome sequence data, biochemical assay data and metabolomics data, as well as detailed imaging measures, and much larger numbers of incident cases of many different conditions identified from the NHS England linked healthcare data.

- Outputs from the data processing

Applications to use UK Biobank data are able to generate more than one publication/output. To date, over 10,000 publications have been produced, and the number of papers published each year is increasing rapidly (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/health-related-outcomes-data - note that this webpage is currently awaiting a refresh of data for 2022 and 2023). UK Biobank expects to see a continued increase in the number of research publications based upon analysis of UK Biobank data over the coming years.

- Data contained in the outputs

The resource contains individual-level data with associated pseudo IDs. Every download available to approved researchers has a unique set of randomly-generated pseudo IDs. Summary data derived from UK Biobank data are used in an increasing number of journal articles every year but researchers are not permitted to publish individual-level data. UK Biobank is currently developing guidance about the publication of results about very rare diseases which will be published soon.

- Approach to dissemination and communication

Researchers who use the UK Biobank resource are responsible for ensuring that the findings of their research are published and disseminated. Under the terms of UK Biobank’s Access procedures and Material Transfer Agreement, they are required to publish the findings of their research and return their results to UK Biobank so they can be made available for request by other researchers. In 2022, over 2,000 publications referenced UK Biobank data.

UK Biobank disseminates huge amounts of information to participants and researchers (plus the general public) via its website and other channels.

UK Biobank makes its governance, history, use, plans, and news available transparently through its website. The website also provides a broad range of information about the resource and its outputs, including hosting lists of approved research (https://www.ukbiobank.ac.uk/enable-your-research/approved-research), publications (https://www.ukbiobank.ac.uk/enable-your-research/publications), as well as more detailed case studies of how UK Biobank is helping scientists and researchers to evolve our understanding of human health and the most common life-threatening diseases (https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/our-impact).

It also has a Twitter feed (https://twitter.com/uk_biobank/) and YouTube channel (https://www.youtube.com/@uk_biobank), releases newsletters annually to participants (e.g. https://www.ukbiobank.ac.uk/explore-your-participation/stay-involved/2022-newsletter), hosts interactive events with participants at which enhancement projects and the results of research using the resource are discussed (see https://youtu.be/R1aHLOS1qoY for a recording of a recent participant event held in Newcastle upon Tyne), organises an annual Scientific Conference attended by an international delegation (see https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/winter-scientific-conference-2022 for details of the conference held in December 2022), and engages with industry partners. UK Biobank presented details of its work at the following external conferences in the past year: the European Society of Human Genetics, the American Society of Human Genetics, European Society of Cardiology, Bio-IT world conference, and the Festival of Genomics.

Participants can contact the Participant Resource Centre with questions, or to provide feedback. The Centre answers over 20,000 calls and around 35,000 emails from participants annually. Feedback obtained about enhancement studies, either via the PRC or other feedback routes is relayed to project teams and appropriate actions taken to address any issues of concern raised. Participants who make enquiries about international research all receive a written response from UK Biobank and are offered the opportunity to speak directly to UK Biobank staff, including the Principal Investigator.

UK Biobank has recently appointed a Head of Engagement part of whose role it will be to ensure that the way in which UK Biobank communicates with its participants (both in terms of engaging them in specific UK Biobank activities and more generally) and researchers is as effective as it can be. Examples of forthcoming activities include questionnaires and focus groups about the materials used for the UK Biobank imaging study, as well as focus groups about possible new data linkages and new questionnaires. Furthermore, the UK Biobank Ethics Advisory Committee leads a programme of research which includes consulting participant views about relevant issues.

- Target dates for outputs

Publications have been generated using the resource for over a decade and increase in number year on year (https://www.ukbiobank.ac.uk/enable-your-research/about-our-data/health-related-outcomes-data - note that this webpage is currently awaiting a refresh of data for 2022 and 2023). Researchers are expected to publish their findings within six months of the completion of their project, or three years following approval to access the data. This is stipulated in the MTA. Additionally, within six months of publication, results must be provided to UK Biobank so that they may be incorporated into the resource where relevant, for the benefit of others.

To date no specific metrics relating to targets have been shared with NHS England. However, UK Biobank is now a major international research resource for health-related research and is, by some distance, the most widely used and published upon health research resource in the world. It enables thousands of researchers worldwide to study the genetic, social or economic determinants of both common and rare conditions, leading to the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses and the number of researchers, applications and publications resulting from the resource continues to grow.

- Yielded benefits

Of the numerous and wide-ranging benefits yielded to date (also see https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/our-impact), UK Biobank's contribution to the national and international response to the recent COVID-19 pandemic presents a good case study for the broad value and benefit of the resource. The existence of this large, detailed, prospective cohort with high degrees of participant engagement enabled rapid repurposing in 2020 for COVID-19 research. The frequency of hospitalisation and death data updates from NHS England was immediately increased, and linkage was extended to SARS-CoV-2 testing data and to primary care data under emergency legislation (obtained separately). Much of this work has already translated into policy impact during the pandemic. For example, ~450 publications on COVID-19 have been produced using UK Biobank data. These include: studies of the genetic determinants of severe COVID-19 (Pairo-Castineria et al., 2021); relationships of pre-existing adverse cardiac phenotypes with COVID-19, determined from imaging scans (Raisi-Estabragh et al., 2021); and associations of NMR-metabolomic biomarkers for low-grade inflammation and cardiometabolic disease with severe COVID-19 (Julkunen et al., 2021).

The UK Biobank resource is enabling unprecedented worldwide genomics research and is at the heart of the UK Government's strategy for incorporating genomics in healthcare (HM Government, 2020). This strategy document recognises UK Biobank's central role in the development of the polygenic risk score (PRS) concept, which involves using the combined effects of thousands of genetic variants (each of which has only a small effect) to improve risk prediction for common diseases. Using UK Biobank data, researchers from the Broad Institute were the first to show that polygenic risk scores (PRS) – which combine the effects of very large numbers of genetic variant – could identify people at a high risk of a particular disease (often equivalent to that of a single-gene disorder). In particular, they found that about 5% of the population had a PRS which put them at equivalent risk of developing heart disease to someone who had Familial Hypercholesterolaemia (Khera et al., 2018). Other research using the whole exome sequencing data found that rare genetic mutations in the GPR75 gene (which encodes a receptor in the brain that regulates energy balance) are associated with having half the risk of developing obesity. Protective ‘loss of function’ mutations were found in about one of every 3,000 people sequenced. Heralded as a “genetic superpower”. This discovery will directly lead to approaches (such as gene-editing and gene-silencing) to develop drugs that mimic the function of this mutation (Akbari, et al., 2021).

UK Biobank’s imaging study enables work which provides an excellent example of how a single study can be used to improve understanding of disease. Researchers from the University of Oxford carried out genome association studies of functional and structural brain imaging phenotypes and showed that many of the phenotypes are heritable. The results provide insight into the genetic architecture of the brain that are of relevance to brain development, ageing, dementia and other brain disorders (Elliott, et al., 2018).

- References

Akbari P, et al. Sequencing of 640,000 exomes identifies GPR75 variants associated with protection from obesity Science, 2021: 373 (6550)

Elliott, L.T., Sharp, K., Alfaro-Almagro, F. et al. Genome-wide association studies of brain imaging phenotypes in UK Biobank. Nature 562, 210–216 (2018).

HM Government. Genome UK: The future of healthcare. 2020

Julkunen H, et al. Metabolic biomarker profiling for identification of susceptibility to severe pneumonia and COVID-19 in the general population. Elife. 2021; 10: e63033

Khera AV, et al. Genome-wide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. Nat Genet. 2018: 50; 1219-24

Pairo-Castineira, E., et al. Genetic mechanisms of critical illness in COVID-19. Nature; 2021; 591, 92–98.

Raisi-Estabragh Z, et al. Adverse cardiovascular magnetic resonance phenotypes are associated with greater likelihood of incident coronavirus disease 2019: findings from the UK Biobank. Aging Clin Exp Res; 2021; 33: 1133-1144

- How the outputs will achieve the stated purpose

The UK Biobank resource enables researchers to make scientific discoveries that provide considerable benefit to human health, not just for people living in the UK but worldwide. Much of this work relies upon data made available by NHS England. In relation to health outcomes, there are over 2,000 published papers on cardiovascular disease, over 1,400 on cancer, over 500 on mental health, and over 900 on obesity. Other broader research projects have published on the identification of novel patterns of association of many different exposures with a range of health outcomes. The rapid availability of more frequent healthcare records for COVID-19 research has already led to over 450 papers. A list of publications using UKB data can be accessed at https://www.ukbiobank.ac.uk/enable-your-research/publications (note that this webpage is currently awaiting a refresh of data for 2022 and 2023). These publications will improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses by contributing to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers and the development of new and improved treatments for patients with existing conditions, all of which are of considerable benefit to the public.

Aim and purpose of this application

UK Biobank is a major national and international health resource, and a registered charity in its own right, with the aim of improving the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye conditions, depression and forms of dementia. UK Biobank recruited 500,000 people aged between 40-69 years in 2006-2010 from across the UK to take part. Of these, 466,818 participants remain. They have answered detailed questionnaires, undergone a range of physical measures, provided blood, urine and saliva samples for future analysis, and agreed to have their health followed-up over time through linkage to their health and health-related records. The UK Biobank resource is available to researchers worldwide who can apply to use the de-identified data within the resource to undertake health-related research that is in the public interest.

UK Biobank requires data from NHS Digital to pursue its legitimate interests, which are to set up and manage a major international research resource for health-related research that is in the public interest. UK Biobank’s Article 6 lawful basis for processing the personal data relating to the participants in the UK Biobank study is therefore Article 6 (1) (f) of the GDPR (legitimate interest). UK Biobank processes special category personal data provided by NHS Digital in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. UK Biobank falls clearly within this category as its purpose is to improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses. The data are required for research purposes in the public interest – meeting the conditions on the DPA Schedule 1 Part 1 (4) – which GDPR Recital 52 (2) determines is an appropriate derogation from prohibition on processing special categories of personal data; proportionate to the aim pursued (as above). The ways in which the processing of data will be of benefit to the public – thereby demonstrating that the processing is in the public interest – are described in section ‘5d. ii.

UK Biobank is a prospective study with the specific objective of following up the health of its participants through access to their health records. This resource enables researchers to study the genetic, social or economic determinants of both common and rare conditions. This in turn, will lead to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers (e.g., polygenic risk scores or imaging biomarkers of internal organs) and the development of new and improved treatments for patients with existing conditions, all of which will be of considerable benefit to the public. Without the health record data, UK Biobank is simply not able to achieve its aims.

How will the data requested achieve the aim identified?

The addition of the requested data from NHS Digital to the resource enhances the range of information that is available to researchers to enable them to undertake their approved research project. UK Biobank, as a charitable company limited by guarantee, was established to enable researchers to undertake health-related research which is in the public interest and therefore the data are required from NHS Digital to enable UK Biobank to pursue its legitimate interests.

As a result of UK Biobank’s work, patients and the wider public benefit from the advances made in the prevention, diagnosis and treatment of disease. UK Biobank is now one of the largest and most used health research resources in the world. Over 28,000 researchers are registered with and approved by UK Biobank, from 2,300 approved institutions, and 2,900 health-related research applications have been approved to date.

Background to the request / the wider study

UK Biobank is a unique resource due to its very large size and the range and detail of data that it contains on each of its participants, including extensive phenotype and genotype information. Furthermore, rather than recruiting participants with specific diseases or risk factors, UK Biobank is a prospective population-based resource established to support research into the causes of a wide range of diseases affecting people in middle and older age. The ability to link to participant’s health records, so that specific outcomes occurring during long-term participant follow-up can be identified and appropriate disease-based research carried out, is therefore critical.

UK Biobank’s priority is to combine extensive and precise measurement of exposures with detailed and rigorous follow-up for a wide range of health-related outcomes, and to promote innovative science by maximising secure access to de-identified data.

UK Biobank was established in 2003. It is funded by, amongst others, the Wellcome Trust, and the Medical Research Council. No funder is permitted to influence the outcomes nor suppress any outcomes of research.

Access to the resource opened in 2012, since which time over 2,900 applications have been approved and over 28,000 researchers have registered and been approved to use the data. As a result, over 4,700 research papers have been published to date using the UK Biobank resource. The UK Biobank resource is available to all bona fide researchers for all types of health-related research that is in the public interest. All researchers, whether in universities, charities, government agencies or commercial companies, and whether based in the UK or abroad, are subject to the same application process and approval criteria. All approved researchers are required to enter into a standard legally-binding Material Transfer Agreement – on a project-by-project basis – which contains provisions which inter alia require the researchers to comply with the UK’s data protection regime, relating to storage, processing, transfer and deletion of the data.

UK Biobank charges standard fees – which are the same for all applicants - for access to the resource which is calculated on a cost-recovery basis. These fees are based on a cost-recovery model so that UK Biobank can recover its operational access costs. Further information about the fees charged can be found here https://www.ukbiobank.ac.uk/enable-your-research/costs.

The expected and yielded benefits of allowing researchers to access the UK Biobank resource and the linked healthcare data provided by NHSD are described in section 5d.

The data subjects

Between 2006 and 2010, UK Biobank recruited 500,000 people from the general population - known as “participants” - then aged between 40 and 69 to take part. The remaining 466,818 participants are the data subjects for the purpose of this application.

The methods used for the identification, invitation and assessment of participants were developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All UK Biobank participants have given fully informed consent for: access to and long-term storage of information from their medical and other health-related records; use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purposes; and re-contact by UK Biobank to invite them (but without any obligation) to answer further questions or to take part in further assessments.

It is considered very unlikely that participants would object to the processing activities involved in obtaining linked data from NHS Digital. Participants recruited can withdraw from UK Biobank at any time, although since UK Biobank began recruitment in 2006 only a small number have done so.

In terms of participant engagement:

- The lay summary of all approved projects are publicly available on the UK Biobank website as our details of all UK Biobank's major initiatives and developments;

- Participants receive regular newsletters in which use of the resource by international researchers are highlighted;

- Participants are invited to UK Biobank-hosted events at which enhancement projects and the results of research using resource are discussed;

- . Participants can contact the Participant Resource Centre with questions, or to provide feedback. The Centre answers approximately 22,000 calls and 34,000 emails from participants annually. Feedback obtained about enhancement studies, either via the PRC or via other feedback processes is relayed to project teams and appropriate actions taken to address any issues of concern raised;

- Participants who makke enquiries about international research all receive a written response from UK Biobank and are offered the ability to speak directly to UK Biobank staff including the Principal Investigator;

- The UK Biobank Ethics Advisory Committee (EAC), a Committee of the UK Biobank Board whose remit is to provide advice to the Board on ethical issues that arise during the maintenance, development and use of UK Biobank leads a programme of research which includes consulting participant views about relevant issues.

The purpose of the request

Under this Agreement, UK Biobank receives updates on participants’ health and health-related data. These include:

i. Hospital Episode Statistics – details of participants’ attendance at hospital for any reason. 1997/98 to latest available quarterly update (updated quarterly to assist research involving COVID-19).

ii. Mental Health data – record-level data about the care of participants who are in contact with mental health, learning disabilities or autism spectrum disorder services. Mental Health Minimum Data Set 2006/07 to 2014/15, Mental Health and Learning Disabilities Data Set 2014/15 to November 2015/16, Mental Health Services Data Set 2016/17; Improved Access to Psychological Therapies Data Set April 2012 - March 2013 to latest available.

iii. Civil Registration Mortality data – details of participants’ deaths including date and cause. 2011/12 to latest available quarterly update (updated quarterly to assist research involving COVID-19);

iv. Cancer Registration data – details of any cancers participants are registered as having. 2011/12 to latest available with annual updates.

v. National Diabetes Audit data - data collected on diabetes in participants. April 2003 to latest available annual update.

vi. COVID-19 Hospitalization in England Surveillance System (CHESS) / Severe Acute Respiratory Infection (SARI) Watch. UK Biobank need to retain historical data already held, without updates, until value to resource is determined, after which further updates will be requested or data securely destroyed.

vii. COVID-19 Vaccination Status: All historical data to latest available, quarterly update.

The Covid-19 datasets disseminated under this Agreement which are COVID-19 Hospitalization in England Surveillance System (CHESS), and the COVID-19 Vaccination Status, will only be used for Covid-19 research purposes.

Approved projects undertaking COVID-19 related research then have this restricted purpose explicitly added to the terms of the MTA. UK Biobank then reviews the publications to ensure that the research has indeed been COVID-19 related research. If non-compliance was detected, then UK Biobank would require the researcher to instruct withdraw publication.

viii. Personal Demographic Service (PDS) updates – details of changes to participants’ demographics, postal addresses or registered GP practices.

It is very important for UK Biobank to be able to maintain accurate information about participants’ contact details and registered GP practice (which may have changed since participant recruitment), for the following reasons:

- Ongoing approved linkages to primary care data require accurate information about registered GP practice (at practice as well as CCG level) to ensure that data about a particular participant are requested from the correct GP practice.

- Direct contact with GPs is required as part of the UK Biobank imaging sub-study (www.ukbiobank.ac.uk/explore-your-participation/contribute -further/imaging-study) in the event that a participant is identified with a potentially serious incidental finding (as noted by the radiographer and then confirmed by a radiologist). Should such an abnormality be found, UK Biobank will write to the participant and their GP to allow them to further investigate these potentially serious findings.

- Follow-up in a cohort study needs to be as complete as possible, to avoid both unnecessary reduction in cohort numbers (which reduces statistical power of data analyses) and to minimise loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on intermittent re-contact with UK Biobank participants to invite them to complete brief questionnaires about health-related exposures or outcomes, or to other enhancement projects (such as to attend an imaging assessment or to engage in various SARS-CoV-2 studies), for which correct participant contact details are required. Although participants are encouraged to inform UK Biobank when they change address, this is not a failsafe way of ensuring that the most up-to-date contact information is held on all participants.

- As mentioned above, UK Biobank is committed to ensuring that participants are informed of progress and developments. UK Biobank does this in part through its annual newsletter, sent via e-mail and/or postal mail, for which up-to-date contact details are required. UK Biobank prefers to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress.

As a very large cohort, it is not feasible for UK Biobank itself to follow up regularly each of its participants on an individual basis and nor would this generate the necessary clinical detail. Self-reported data do not possess (anything like) the needed coverage, reliability and consistency that health records contain. As such, the data provided via NHS Digital are crucial to the viability of UK Biobank as the primary mechanism for follow-up of the cohort, particularly as they provide information on all 500,000 (now 466,818 remaining) participants in an entirely unobtrusive (i.e. passive) manner. The linked datasets all form part of the participants’ health record and are thus vital to providing the necessary health information. Of key importance are datasets related to death and cancer registry notifications and hospital activity, as they provide crucial information on which participants have developed which diseases and at what time. Without access to these data, UK Biobank could not function as a prospective health resource and therefore research on the associations between social or economic determinations of health and genetic factors and the risk of developing a wide range of diseases over time would not be possible. UK Biobank is a prospective study available to registered researchers worldwide, whose research interests cover a very wide range of health outcomes, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke and mental health. During the pandemic, UK Biobank released data from linked medical records on an approximately monthly basis to the research community in order to facilitate COVID-19 research. As such, access to data on deaths, cancers and hospital admissions and related activities is of vital importance.

In addition to the above-mentioned datasets that UK Biobank holds, other health-related datasets that have now become available are also of significant research interest. The IAPT dataset (requested in 2020 although data has not yet been received) will help identify participants with mental health conditions and enable research into their treatment response and support required. The COVID-19 datasets will provide an objective measure of vaccination status and enable further research into the impacts of SARS-CoV-2 infection (i.e. long-COVID) across the full disease spectrum (including those hospitalised with COVID--with data obtained from the CHESS/SARI Watch dataset) and the longer-term impacts of vaccination on health. The Covid-19 datasets disseminated under this Agreement will only be used for Covid-19 research purposes.

UK Biobank requires access to all retrospective data to enable epidemiological research into antecedent clinical events prior to diagnosis and the effect of multi-morbidity on future health outcomes. Without linked data from NHS Digital, researchers would not be able to ascertain the full burden of disease in the cohort nor the effects of particular treatment and management pathways on health. All of the datasets requested enhance the UK Biobank resource and allow researchers to undertake important health-related research that is in the public interest.

For each of these datasets listed above, UK Biobank requires linkage only for the 500,000 (now 466,818 remaining) UK Biobank participants who joined the study and who were recruited across the UK. It should be noted that UK Biobank is only requesting individual-level identifying data to ensure correct linkage. UK Biobank does not make available to researchers any identifiable data (such as full postcode, date of birth, name, NHS number, etc.); each research project is instead provided with a unique encrypted 9-digit identification number for participants. Hence, it should not be possible for a researcher to identify a UK Biobank participant using the data that UK Biobank releases.

The organisations involved

UK Biobank is the data controller of the data supplied by NHS Digital. Pursuant to a contract between UK Biobank and The Clinical Trials Service Unit (CTSU) at the University of Oxford, CTSU stores the data provided by NHS Digital and therefore acts as a data processor of UK Biobank’s data. CTSU is only able to act on the instructions of UK Biobank and is thus only a data processor. CTSU has no ability to process the data for any other purpose save for those purposes which UK Biobank (alone) determines.

UK Biobank is a company limited by guarantee and registered at Companies House; it therefore has its own legal personality. While it receives funding from the Wellcome Trust and the Medical Research Council amongst others, these organisations have no access to the data and cannot exercise any control in respect of its use.

Once NHS Digital data have been received by UK Biobank (acting through its processor CTSU), they are uploaded to the UK Biobank resource. Registered researchers can then apply to UK Biobank to use these de-identified linked data (alongside the other data in the resource) to undertake health-related research that is in the public interest. Each application to use the resource is considered by UK Biobank pursuant to the UK Biobank access policy and procedures, which were developed through extensive consultation (including public consultation and consultation with UK Biobank participants). Further detail is given in section 5b. UK Biobank has sole control of this process.

De-identified data are made available to researchers worldwide following approval of a research application once they have signed a contract with UK Biobank (known as a Material Transfer Agreements [MTA]) which controls how the data can be used (including ensuring the data are only used for conducting health-related research in the public interest).

The Access Sub-committee, which meets four times annually, has oversight of all applications, after the applications have first been considered by the Access team and Science team, who recommend to the sub-committee whether to approve or discuss further. The Access and Science teams review applications on a rolling basis, aiming to make recommendations within two weeks of receipt of applications. The UK Biobank Access and Science teams review every application to access the UK Biobank resource to ensure that the proposed research is compliant with UK Biobank’s Access Procedures. The access test is to ensure that the researcher is a bona fide researcher and that the proposed research is health-related and in the public interest. The Access and Scientific teams conduct a set of standard checks and seek advice on particular applications as required. A schedule of all applications is provided to each meeting of the Access Sub-Committee along with a recommendation (from the Access and Scientific Teams) that a) an application should be approved (which is the case with the majority of research applications) or b) rejected or c) considered further by the Access Sub-Committee (for example if a research question could be considered contentious). For more information see https://www.ukbiobank.ac.uk/learn-more-about-uk-biobank/governance/access-sub-committee.

All researchers are required to submit Annual Project Reports to UK Biobank. These ask the researchers to make a number of confirmations:

- provide a confirmation that the project remains within the approved remit;

- provide an update of progress and developments in research (and indicate a likely publication date);

- provide a confirmation that the researcher is complying with the terms of the MTA, including its schedules (specifically the provisions relating to data security).

If a researcher fails to complete or to submit an Annual Project Report they are prevented, in the first instance, from accessing or downloading any further data from UK Biobank. A continued failure to complete the Annual Project Report will result in a project and the associated MTA being terminated. As such it provides an annual means by which UK Biobank can be re-assured about progress and the (annual) ability to terminate any project which is not compliant.

The de-identified data are either made available directly to the researcher (for them to download and use in their own environment) or the researcher is granted access to use the data through UK Biobank’s Research Analysis Platform (RAP).

The RAP comprises a web interface provided by DNAnexus and cloud storage provided by Amazon Web Services (AWS) on the London AWS node. The only data that are made available through the RAP (and securely stored by AWS in the cloud) are the de-identified data that have historically been made available directly to researchers for them to download. Once the data are in the RAP, they are only accessible to researchers who have an approved research project and who have entered into an MTA with UK Biobank. DNAnexus is ISO27001 certified by independent third-party auditors and maintains compliance with ISO27001 (alongside compliance with many internationally recognised standards such as NIST, FedRAMP and HITRUST). The platform utilises the AWS infrastructure from the London node - the UK Biobank data are therefore stored in London and when the data are used/analysed on the platform they will remain in London. The dataset is stored using robust security encryption protocols. Although the platform provider (DNAnexus) and the cloud supplier (AWS) only receive de-identified data, the agreement that UKB has with DNAnexus does require DNAnexus to treat the UKB data to the same standard as if they are personal data and thus contains the ICO-recommended GDPR-compliant data processing clauses. UKB has permitted DNAnexus to appoint AWS under the same Agreement as UKB has appointed DNAnexus.

UK Biobank’s main output is enabling access to the resource for the purposes of health-related research, of which linkage to longitudinal health records is key. Data are provided to bona fide researchers worldwide according to the UK Biobank access policy and procedures, as described in section 5b.

During the last five years, the characterisation of the participants has been enhanced substantially (e.g. with whole-exome and whole-genome sequence data, biochemical assay data, and metabolomics data, as well as detailed imaging measures) and much larger numbers of incident cases of many different conditions have been identified from the NHS Digital linked healthcare data. This has resulted in an increasingly large number of researchers conducting and reporting an increasingly extensive range of research, and the range of health conditions that it will be possible to study reliably using even more detailed information about exposures will increase further as the numbers of incident cases of different health outcomes increase during the next 5-10 years.

UK Biobank’s current enhancement projects and specific outputs:

As part of UK Biobank’s mission to enable scientific discoveries to improve human health, UK Biobank conducted a SARS-CoV-2 antibody study in 2021, the aim of which was to collect SARS-CoV-2 antibody status data from all participants who agree to be provided with a home-based lateral flow test. These data, coupled with the existing genetic and social determinations of health data and linkage to future health records for all 500,000 (now 466,818 remaining) participants, have been uploaded to the resource and will create a unique dataset to enable research into the longer-term health effects of SARS-CoV-2, known as ‘long-covid’.

UK Biobank’s imaging study started in 2014 with the aim of collecting multi-modal MRI data on up to 100,000 UK Biobank participants, with repeat scans being performed in up to 60,000 participants a few years later, and which will run for the next 5-6 years. In February 2021, UK Biobank set up a SARS-CoV-2 repeat imaging project to create a unique resource to enable research into the effects of SARS-CoV-2 infection on internal organs and associated pathophysiology and health outcomes. The study recruited about 2,000 participants who had already attended an imaging assessment for a repeat visit, half of whom had evidence of previous SARS-CoV-2 infection (as identified through linkage to health records or an antibody test) and half of whom did not. The associated MRI and health record data have been made available to researchers on an approximately monthly basis to facilitate rapid COVID-19 research. UK Biobank is the only resource globally able to provide this type of large-scale imaging data for participants both before and after infection.

In order to facilitate research into the genetic and environmental determinants of COVID-19, from April 2020 UK Biobank was releasing more frequent updates of linked health record data. Much of these data come from NHS Digital, including death records, hospital inpatients and critical care records. Consequently, it is uniquely placed to allow scientists nationally and internationally to answer questions about the wide-ranging health consequences of SARS-CoV-2 infection. Currently, updated data is released approximately every six months.

Research applications and approved research projects:

To date, over 30,000 researchers have successfully registered to use the UK Biobank resource from across 3,200 research institutions worldwide. The rate of growth of approved research applications to use UK Biobank data has been substantial, growing from 16 applications in 2012 to 530 in 2021 alone, such that there are now 3,500 approved research applications that are using UK Biobank data for health-related research that is in the public interest. This increase has been notable over the last 5 years, with a 4-fold increase in applications to access the resource since 2016, particularly from researchers outside of the UK (over 80% in 2021 compared with 45% in 2016) – predominantly the USA (24%), China (24%) and the rest of Europe (14%) – although UK-based use remains high (18%). The majority of applications are still led by academic researchers, but an increasing number are coming from commercial entities (76 applications in 2021; 10%).

Summaries of research projects that have been approved and are currently underway can be found at https://www.ukbiobank.ac.uk/approved-research/. The nature of the research that is currently being conducted reflects the increasing breadth and depth of data that are now available, with 72% of applications requesting genotype data, 63% requesting imaging data, and nearly all applications requesting health outcome data. Of note, since 2020 over 700 research groups have accessed the health record data for the purposes of performing COVID-19 research.

Publications:

To date, over 4,000 publications have been produced, and the number of papers published each year is increasing rapidly (from 89 in 2016 to over 800 in 2020). In relation to health outcomes, there are over 900 published papers on cardiovascular disease, over 600 on cancer, 250 on mental health and over 400 on obesity. Other broader research projects have published on the identification of novel patterns of association of many different exposures with different health outcomes. The rapid availability of more frequent healthcare records for COVID-19 research has already led to over 250 papers, which have attracted a high level of attention: Altmetric identifies over 20,000 mentions, with the most popular having a score of 2,820 and 89% of the mentions (typically on Twitter) to that paper being from members of the public.

Given the significant increase in the number of approved research applications to use UK Biobank data over the last two years, UK Biobank expects to see a corresponding increase in the number of research publications based upon analysis of UK Biobank data in the next couple of years. A full list of publications using UKB data can be accessed at https://www.ukbiobank.ac.uk/enable-your-research/publications.

The UK Biobank resource enables researchers to make scientific discoveries that provide considerable benefit to human health, not just for people living in the UK but worldwide. Much of this work relies upon data made available by NHS Digital. UK Biobank has approved over 2,900 research applications since it first opened to researchers in 2012. These cover a very broad range of health-related research topics (https://www.ukbiobank.ac.uk/enable-your-research/approved-research). There have been over 4,000 publications using UK Biobank data (https://www.ukbiobank.ac.uk/enable-your-research/publications), the majority since 2019. There are over 2,400 on-going projects.

It is therefore relatively early to expect research supported by the UK Biobank resource to translate into widespread policy impact. Nonetheless, UK Biobank is at the heart of the UK Government’s strategy for incorporating genomics in healthcare (HM Government, 2020), and its value was highlighted in the NHS Long-Term Plan as one of the UK’s “outstanding capabilities for research and innovation” (NHS, 2019). As of early 2022, there had been a further 161 documents that refer to UK Biobank or research deriving from it.

The value of the UK Biobank resource for informing a wide range of policy areas is illustrated by the breadth of its utility: for example, there are references to research in many areas as disparate as the health effects of air pollution (PHE. 2020a) to Parliamentary briefing papers on loneliness and isolation (Bellis & Loft, 2020). One study on the association between active commuting and disease helped inform two Government policy documents on the future of urban mobility, cycling and walking (Department for Transport, 2019, 2020), whilst another on risk thresholds for alcohol consumption informed two WHO documents on prevention of alcohol-related harm and labelling policy (WHO. 2019, 2020b). Internationally, UK Biobank is referenced as a best practice model for biobanking policies. For example, WHO, 2020.

As outlined in the NHS Personalised Medicine Strategy, personalised medicine can radically transform medical care and population health through prevention and treatment strategies that are more closely targeted to the individual patient. The Government Healthcare Strategy (HM Government, 2020) recognises UK Biobank’s central role in the development of the polygenic risk score (PRS) concept, which involves using the combined effects of thousands of genetic variants (each of which has only a small effect) to improve risk prediction for common diseases. In terms of utility, UK Biobank data have been used to show that PRS for 16 common conditions (Genomics Plc, 2019) derived from a single low-cost genotype test can identify about one-quarter of the population who might benefit from such precision population health approaches.

Availability of such a wealth of genetic and phenotypic data (including participant characteristics [such as height], but also health outcomes derived from data from NHS Digital) has necessitated methodological innovations, such as fast multi-trait GWASs; these take into account the complex interactions between genetic variants and boost statistical power to detect genetic associations, the results of which are available via searchable online browsers (e.g., https://biobankengine.stanford.edu/ and http://geneatlas.roslin.ed.ac.uk/). The combination of healthcare data from linked health records and genomic data has led to huge advances in the use of Mendelian Randomisation approaches to understand the causality of epidemiological associations and thus contribute to appropriate preventative and therapeutic strategies. For example, such analyses provide strong support for a causal role of body fat mass with aortic valve stenosis and most other cardiovascular conditions (Larsson, et al. 2020), and do not support a protective role of vitamin D supplementation on cardiovascular or cancer outcomes (Meng, et al. 2019).

Other examples of impact on health practice deriving directly from UK Biobank include research showing that the incidence rate of type I diabetes remains about the same up to at least age 60 (Thomas, et al. 2018). As a consequence, clinicians are now alert to considering type I diabetes as a diagnosis in patients who present in older life, particularly when they do not achieve good glycaemic control with standard therapy for type 2 diabetes. Other research has shown that those who have two variant copies of a mutation leading to haemochromatosis are at substantially increased risk of developing liver cancer, rheumatoid arthritis, osteoarthritis and diabetes (Pilling, et al. 2019), leading to calls for the UK National Screening Committee to reconsider routine genetic testing for the condition. Research from UK Biobank has also shown that handgrip strength improves the risk prediction of cardiovascular incidence and mortality over and above that of the existing standard clinical risk score and is as predictive as high-density lipoprotein cholesterol and N-terminal pro B-type natriuretic peptide (Celis-Morales, et al. 2018). Thus, handgrip strength, which is easily measured objectively and is highly reproducible in any clinical setting, may add clinical utility over existing risk prediction scores for cardiovascular disease outcomes.

UK Biobank’s repurposing to support the national and international response to the COVID-19 pandemic (which included the increased frequency of hospitalisation and death data updates from NHS Digital combined with expedited access to the established research community of approved researchers) has already seen benefits. Already, 209 reports on COVID-19, which have been cited ~2,500 times, have been produced using the resource. These include studies of the genetic determinants of severe COVID-19 (Pairo-Castineria, et al. 2021), relationships of pre-existing adverse cardiac phenotypes determined from imaging scans with COVID-19 (Raisi-Estabragh, et al. 2021) and associations of NMR-metabolomic biomarkers for low-grade inflammation and cardiometabolic disease with severe COVID-19 (Julkunen, et al. 2021). Overall, these publications have attracted a high level of attention: To date, Altmetric identifies over 20,000 mentions, with the most popular having a score of 2,820 and 89% of the Twitter mentions to that paper being from members of the public. Some of the findings have translated into policy impact, including the NICE guidelines on vitamin D supplementation, UK Government reporting on the differential impact of COVID-19 among minority ethnic groups and a review of disparities in risks and outcomes (PHE, 2020b). Internationally, UK Biobank-enabled research has been cited by the WHO in their recent COVID-19 clinical management guidance (WHO, 2021), and by an EC-commissioned study on air pollution and COVID-19.

References

Bellis A, Loft P. Tackling loneliness. House of Commons Briefing Paper number 8514. House of Commons Library. 2020

Celis-Morales CA, et al. Associations of grip strength with cardiovascular, respiratory, and cancer outcomes and all-cause mortality: prospective cohort study of half a million UK Biobank participants. BMJ, 2018; 361: k1651

Department for Transport. Future of Mobility: Urban Strategy. 2019

Department for Transport. Gear Change: A bold vision for cycling and walking. 2020

Douaud G, et al. Brain imaging before and after COVID-19 in UK Biobank. medRxiv [Preprint]. 2021 Aug 18:2021.06.11.21258690

European Parliament, Directorate-General for Internal Policies of the Union, Smit L, et al. Air pollution and COVID-19 : including elements of air pollution in rural areas, indoor air pollution, vulnerability and resilience aspects of our society against respiratory disease, social inequality stemming from air pollution, European Parliament, 2021

HM Government. Genome UK: The future of healthcare. 2020

Julkunen H, et al. Metabolic biomarker profiling for identification of susceptibility to severe pneumonia and COVID-19 in the general population. Elife. 2021; 10: e63033

Larsson SC, et al. Body mass index and body composition in relation to 14 cardiovascular conditions in UK Biobank: a Mendelian randomization study. Europ Heart J, 2020; 41: 221-6

Meng X, et al. Phenome-wide Mendelian-randomisation study of genetically determined vitamin D on multiple health outcomes using the UK Biobank study. Int J Epidemiol; 2019; 48: 1425-34

NHS. The NHS Long Term Plan. 2019

NHS England. Improving Outcomes through Personalised Medicine. 2016

Pairo-Castineira E, et al. Genetic mechanisms of critical illness in COVID-19. Nature; 2021; 591, 92–98

Pilling LC, et al. Common conditions associated with hereditary haemochromatosis genetic variants: cohort study in UK Biobank. BMJ. 2019: 364; k5222

PHE. Disparities in the risks and outcomes of COVID-19. 2020b

PHE. Summary of the Committee on the Medical Effects of Air Pollutants COMEAP recommendations for the quantification of health effects associated with air pollutants. 2020a

Raisi-Estabragh Z, et al. Adverse cardiovascular magnetic resonance phenotypes are associated with greater likelihood of incident coronavirus disease 2019: findings from the UK Biobank. Aging Clin Exp Res; 2021; 33: 1133-1144

Thomas NJ, et al. Frequency and phenotype of type 1 diabetes in the first six decades of life: a cross-sectional, genetically stratified survival analysis from UK Biobank. Lancet Diabetes Endocrinol. 2018: 6; 122-129

WHO. Questions and Answers on the prevention and control of alcohol-related harm. 2019

WHO. Regulating the unknown: A guide to regulating genomics for health policy-makers. 2020a

WHO. What is the current alcohol labelling practice in the WHO European Region and what are barriers and facilitators to development and implementation of alcohol labelling policy? 2020b

WHO. COVID-19 Clinical management: living guidance. 2021

**V14 of this agreement is a 2-month SIRO approved Extension at no cost to keep Biobank in agreement. The Extension will permit Biobank to hold and process previously supplied data.**

Aim and purpose of this application

UK Biobank is a major national and international health resource, and a registered charity in its own right, with the aim of improving the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye conditions, depression and forms of dementia. UK Biobank recruited 500,000 people aged between 40-69 years in 2006-2010 from across the UK to take part. They have answered detailed questionnaires, undergone a range of physical measures, provided blood, urine and saliva samples for future analysis, and agreed to have their health followed-up over time through linkage to their health and health-related records. The UK Biobank resource is available to researchers worldwide who can apply to use the de-identified data within the resource to undertake health-related research that is in the public interest.

UK Biobank requires data from NHS Digital to pursue its legitimate interests, which are to set up and manage a major international research resource for health-related research that is in the public interest. UK Biobank’s Article 6 lawful basis for processing the personal data relating to the participants in the UK Biobank study is therefore Article 6 (1) (f) of the GDPR (legitimate interest). UK Biobank processes special category personal data provided by NHS Digital in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. UK Biobank falls clearly within this category as its purpose is to improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses. The data are required for research purposes in the public interest – meeting the conditions on the DPA Schedule 1 Part 1 (4) – which GDPR Recital 52 (2) determines is an appropriate derogation from prohibition on processing special categories of personal data; proportionate to the aim pursued (as above). The ways in which the processing of data will be of benefit to the public – thereby demonstrating that the processing is in the public interest – are described in section ‘5d. ii.

UK Biobank is a prospective study with the specific objective of following up the health of its participants through access to their health records. This resource enables researchers to study the genetic, social or economic determinants of both common and rare conditions. This in turn, will lead to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers (e.g., polygenic risk scores or imaging biomarkers of internal organs) and the development of new and improved treatments for patients with existing conditions, all of which will be of considerable benefit to the public. Without the health record data, UK Biobank is simply not able to achieve its aims.

How will the data requested achieve the aim identified?

The addition of the requested data from NHS Digital to the resource enhances the range of information that is available to researchers to enable them to undertake their approved research project. UK Biobank, as a charitable company limited by guarantee, was established to enable researchers to undertake health-related research which is in the public interest and therefore the data are required from NHS Digital to enable UK Biobank to pursue its legitimate interests.

As a result of UK Biobank’s work, patients and the wider public benefit from the advances made in the prevention, diagnosis and treatment of disease. UK Biobank is now one of the largest and most used health research resources in the world. Over 28,000 researchers are registered with and approved by UK Biobank, from 2,300 approved institutions, and 2,900 health-related research applications have been approved to date.

Background to the request / the wider study

UK Biobank is a unique resource due to its very large size and the range and detail of data that it contains on each of its participants, including extensive phenotype and genotype information. Furthermore, rather than recruiting participants with specific diseases or risk factors, UK Biobank is a prospective population-based resource established to support research into the causes of a wide range of diseases affecting people in middle and older age. The ability to link to participant’s health records, so that specific outcomes occurring during long-term participant follow-up can be identified and appropriate disease-based research carried out, is therefore critical.

UK Biobank’s priority is to combine extensive and precise measurement of exposures with detailed and rigorous follow-up for a wide range of health-related outcomes, and to promote innovative science by maximising secure access to de-identified data.

UK Biobank was established in 2003. It is funded by, amongst others, the Wellcome Trust, and the Medical Research Council. No funder is permitted to influence the outcomes nor suppress any outcomes of research.

Access to the resource opened in 2012, since which time over 2,900 applications have been approved and over 28,000 researchers have registered and been approved to use the data. As a result, over 4,700 research papers have been published to date using the UK Biobank resource. The UK Biobank resource is available to all bona fide researchers for all types of health-related research that is in the public interest. All researchers, whether in universities, charities, government agencies or commercial companies, and whether based in the UK or abroad, are subject to the same application process and approval criteria.

UK Biobank charge a fee for access to the resource which is calculated on a cost-recovery basis. UK Biobank seek to recover only the incremental costs of servicing an application to access the resource. Further information about the fees charged can be found here https://www.ukbiobank.ac.uk/enable-your-research/costs.

As explained above in section 5b, the UK Biobank resource is available to all bona fide researchers for all types of health-related research that is in the public interest. All researchers, whether in universities, charities, government agencies or commercial companies, and whether based in the UK or abroad, are subject to the same application process and approval criteria.

The expected and yielded benefits of allowing researchers to access the UK Biobank resource and the linked healthcare data provided by NHSD are described in section 5d.

The data subjects

Between 2006 and 2010, UK Biobank recruited 500,000 people from the general population - known as “participants” - then aged between 40 and 69 to take part. The participants are the data subjects for the purpose of this application.

The methods used for the identification, invitation and assessment of participants were developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All UK Biobank participants have given fully informed consent for: access to and long-term storage of information from their medical and other health-related records; use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purposes; and re-contact by UK Biobank to invite them (but without any obligation) to answer further questions or to take part in further assessments.

It is considered very unlikely that participants would object to the processing activities involved in obtaining linked data from NHS Digital. Participants recruited can withdraw from UK Biobank at any time, although since UK Biobank began recruitment in 2006 only a small number have done so.

The purpose of the request

Under this Agreement, UK Biobank receives updates on participants’ health and health-related data. These include:

i. Hospital Episode Statistics – details of participants’ attendance at hospital for any reason. 1997/98 to latest available quarterly update (updated quarterly to assist research involving COVID-19).

ii. Mental Health data – record-level data about the care of participants who are in contact with mental health, learning disabilities or autism spectrum disorder services. Mental Health Minimum Data Set 2006/07 to 2014/15, Mental Health and Learning Disabilities Data Set 2014/15 to November 2015/16, Mental Health Services Data Set 2016/17 to latest available annual update; Improved Access to Psychological Therapies Data Set April 2012 - March 2013 to latest available.

iii. Civil Registration Mortality data – details of participants’ deaths including date and cause. 2011/12 to latest available quarterly update (updated quarterly to assist research involving COVID-19);

iv. Cancer Registration data – details of any cancers participants are registered as having. 2011/12 to latest available with quarterly update.

v. National Diabetes Audit data - data collected on diabetes in participants. April 2003 to latest available annual update.

vi. GPES Data for Pandemic Planning and Research - GDPPR (COVID-19) – details of participants’ clinical events and prescriptions. October 2020 to 2021.

vii. COVID-19 Hospitalization in England Surveillance System (CHESS). All historical data to latest available, quarterly update.

viii. COVID-19 Vaccination Status: All historical data to latest available, quarterly update.

ix. Personal Demographic Service (PDS) updates – details of changes to participants’ demographics, postal addresses or registered GP practices.

- Ongoing approved linkages to primary care data require accurate information about registered GP practice (at practice as well as CCG level) to ensure that data about a particular participant are requested from the correct GP practice.

- Direct contact with GPs is required as part of the UK Biobank imaging sub-study (www.ukbiobank.ac.uk/explore-your-participation/contribute -further/imaging-study) in the event that a participant is identified with a potentially serious incidental finding (as noted by the radiographer and then confirmed by a radiologist). Should such an abnormality be found, UK Biobank will write to the participant and their GP to allow them to further investigate these potentially serious findings.

- Follow-up in a cohort study needs to be as complete as possible, to avoid both unnecessary reduction in cohort numbers (which reduces statistical power of data analyses) and to minimise loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on intermittent re-contact with UK Biobank participants to invite them to complete brief questionnaires about health-related exposures or outcomes, or to other enhancement projects (such as to attend an imaging assessment or to engage in various SARS-CoV-2 studies), for which correct participant contact details are required. Although participants are encouraged to inform UK Biobank when they change address, this is not a failsafe way of ensuring that the most up-to-date contact information is held on all participants.

- UK Biobank is committed to ensuring that participants are informed of progress and developments. UK Biobank does this in part through its annual newsletter, sent via e-mail and/or postal mail, for which up-to-date contact details are required. UK Biobank prefers to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress.

- Important potential research questions about the effects of environmental exposures on health require information about current and previous postal address to allow these exposures to be estimated (from postcodes). These include, for example, valuable studies of the health effects of air and noise pollution, living close to power cables, and aspects of the local environment such as nearby local public amenities and green space.

As a large cohort, it is not feasible for UK Biobank itself to follow up regularly each of its participants on an individual basis and nor would this generate the necessary clinical detail. Self-reported data do not possess (anything like) the needed coverage, reliability and consistency that health records contain. As such, the data provided via NHS Digital are crucial to the viability of UK Biobank as the primary mechanism for follow-up of the cohort, particularly as they provide information on all 500,000 participants in an entirely unobtrusive (i.e. passive) manner. The linked datasets all form part of the participants’ health record and are thus vital to providing the necessary health information. Of key importance are datasets related to death and cancer registry notifications and hospital activity, as they provide crucial information on which participants have developed which diseases and at what time. Without access to these data, UK Biobank could not function as a prospective health resource and therefore research on the associations between social or economic determinations of health and genetic factors and the risk of developing a wide range of diseases over time would not be possible. UK Biobank is a prospective study available to registered researchers worldwide, whose research interests cover a very wide range of health outcomes, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke and mental health. During the pandemic, UK Biobank released data from linked medical records on an approximately monthly basis to the research community in order to facilitate COVID-19 research. As such, access to data on deaths, cancers and hospital admissions and related activities is of vital importance.

In addition to the above-mentioned datasets that UK Biobank holds, other health-related datasets that have now become available are also of significant research interest. The IAPT dataset (requested in 2020 although data has not yet been received) will help identify participants with mental health conditions and enable research into their treatment response and support required. The COVID-19 datasets will provide an objective measure of vaccination status and enable further research into the impacts of SARS-CoV-2 infection (i.e. long-COVID) across the full disease spectrum (including those hospitalised with COVID--with data obtained from the CHESS dataset) and the longer-term impacts of vaccination on health.

UK Biobank requires access to all retrospective data to enable epidemiological research into antecedent clinical events prior to diagnosis and the effect of multi-morbidity on future health outcomes. Without linked data from NHS Digital, researchers would not be able to ascertain the full burden of disease in the cohort nor the effects of particular treatment and management pathways on health. All of the datasets requested enhance the UK Biobank resource and allow researchers to undertake important health-related research that is in the public interest.

For each of these datasets listed above, UK Biobank requires linkage only for the 500,000 UK Biobank participants who joined the study and who were recruited across the UK. It should be noted that UK Biobank is only requesting individual-level identifying data to ensure correct linkage. UK Biobank does not make available to researchers any identifiable data (such as full postcode, date of birth, name, NHS number, etc.); each research project is instead provided with a unique encrypted 9-digit identification number for participants. Hence, it should not be possible for a researcher to identify a UK Biobank participant using the data that UK Biobank releases.

The organisations involved

UK Biobank is the data controller of the data supplied by NHS Digital. Pursuant to a contract between UK Biobank and The Clinical Trials Service Unit (CTSU) at the University of Oxford, CTSU stores the data provided by NHS Digital and therefore acts as a data processor of UK Biobank’s data. CTSU is only able to act on the instructions of UK Biobank and is thus only a data processor. CTSU has no ability to process the data for any other purpose save for those purposes which UK Biobank (alone) determines.

UK Biobank is a company limited by guarantee and registered at Companies House; it therefore has its own legal personality. While it receives funding from the Wellcome Trust and the Medical Research Council amongst others, these organisations have no access to the data and cannot exercise any control in respect of its use.

Once NHS Digital data have been received by UK Biobank (acting through its processor CTSU), they are uploaded to the UK Biobank resource. Registered researchers can then apply to UK Biobank to use these de-identified linked data (alongside the other data in the resource) to undertake health-related research that is in the public interest. Each application to use the resource is considered by UK Biobank pursuant to the UK Biobank access policy and procedures, which were developed through extensive consultation (including public consultation and consultation with UK Biobank participants). Further detail is given in section 5b. UK Biobank has sole control of this process.

De-identified data are made available to researchers following approval of a research application once they have signed a contract with UK Biobank (known as a Material Transfer Agreements [MTA]) which controls how the data can be used (including ensuring the data are only used for conducting health-related research in the public interest). The de-identified data are either made available directly to the researcher (for them to download and use in their own environment) or the researcher is granted access to use the data through UK Biobank’s Research Analysis Platform (RAP).

The RAP comprises a web interface provided by DNAnexus and cloud storage provided by Amazon Web Services (AWS). The only data that are made available through the RAP (and securely stored by AWS in the cloud) are the de-identified data that have historically been made available directly to researchers for them to download. Once the data are in the RAP, they are only accessible to researchers who have an approved research project and who have entered into an MTA with UK Biobank. DNAnexus is ISO27001 certified by independent third-party auditors and maintains compliance with ISO27001 (alongside compliance with many internationally recognised standards such as NIST, FedRAMP and HITRUST). The platform utilises the AWS infrastructure from the London node - the UK Biobank data are therefore stored in London and when the data are used/analysed on the platform they will remain in London. The dataset is stored using robust security encryption protocols. Although the platform provider (DNAnexus) and the cloud supplier (AWS) only receive de-identified data, the agreement that UKB has with DNAnexus does require DNAnexus to treat the UKB data to the same standard as if they are personal data and thus contains the ICO-recommended GDPR-compliant data processing clauses. UKB has permitted DNAnexus to appoint AWS on the same basis as UKB has appointed DNAnexus.

UK Biobank’s main output is enabling access to the resource for the purposes of health-related research, of which linkage to longitudinal health records is key. Data are provided to bona fide researchers worldwide according to the UK Biobank access policy and procedures, as described in section 5b.

During the last five years, the characterisation of the participants has been enhanced substantially (e.g. with whole-exome and whole-genome sequence data, biochemical assay data, and metabolomics data, as well as detailed imaging measures) and much larger numbers of incident cases of many different conditions have been identified from the NHS Digital linked healthcare data. This has resulted in an increasingly large number of researchers conducting and reporting an increasingly extensive range of research, and the range of health conditions that it will be possible to study reliably using even more detailed information about exposures will increase further as the numbers of incident cases of different health outcomes increase during the next 5-10 years.

UK Biobank’s current enhancement projects and specific outputs:

As part of UK Biobank’s mission to enable scientific discoveries to improve human health, UK Biobank conducted a SARS-CoV-2 antibody study in 2021, the aim of which was to collect SARS-CoV-2 antibody status data from all participants who agree to be provided with a home-based lateral flow test. These data, coupled with the existing genetic and social determinations of health data and linkage to future health records for all 500,000 participants, have been uploaded to the resource and will create a unique dataset to enable research into the longer-term health effects of SARS-CoV-2, known as ‘long-covid’.

UK Biobank’s imaging study started in 2014 with the aim of collecting multi-modal MRI data on up to 100,000 UK Biobank participants, with repeat scans being performed in up to 60,000 participants a few years later, and which will run for the next 5-6 years. In February 2021, UK Biobank set up a SARS-CoV-2 repeat imaging project to create a unique resource to enable research into the effects of SARS-CoV-2 infection on internal organs and associated pathophysiology and health outcomes. The study recruited about 2,000 participants who had already attended an imaging assessment for a repeat visit, half of whom had evidence of previous SARS-CoV-2 infection (as identified through linkage to health records or an antibody test) and half of whom did not. The associated MRI and health record data have been made available to researchers on an approximately monthly basis to facilitate rapid COVID-19 research. UK Biobank is the only resource globally able to provide this type of large-scale imaging data for participants both before and after infection.

In order to facilitate research into the genetic and environmental determinants of COVID-19, from April 2020 UK Biobank was releasing more frequent updates of linked health record data. Much of these data come from NHS Digital, including death records, hospital inpatients and critical care records. Consequently, it is uniquely placed to allow scientists nationally and internationally to answer questions about the wide-ranging health consequences of SARS-CoV-2 infection. Currently, updated data is released approximately every six months.

Research applications and approved research projects:

To date, over 28,000 researchers have successfully registered to use the UK Biobank resource from across 2,300 research institutions worldwide. The rate of growth of approved research applications to use UK Biobank data has been substantial, growing from 16 applications in 2012 to 530 in 2021 alone, such that there are now 2,900 approved research applications that are using UK Biobank data for health-related research that is in the public interest. This increase has been notable over the last 5 years, with a 4-fold increase in applications to access the resource since 2016, particularly from researchers outside of the UK (over 80% in 2021 compared with 45% in 2016) – predominantly the USA (24%), China (24%) and the rest of Europe (14%) – although UK-based use remains high (18%). The majority of applications are still led by academic researchers, but an increasing number are coming from commercial entities (76 applications in 2021; 10%).

Summaries of research projects that have been approved and are currently underway can be found at https://www.ukbiobank.ac.uk/approved-research/. The nature of the research that is currently being conducted reflects the increasing breadth and depth of data that are now available, with 72% of applications requesting genotype data, 63% requesting imaging data, and nearly all applications requesting health outcome data. Of note, since 2020 over 700 research groups have accessed the health record data for the purposes of performing COVID-19 research.

Publications:

To date, over 4,000 publications have been produced, and the number of papers published each year is increasing rapidly (from 89 in 2016 to over 800 in 2020). In relation to health outcomes, there are over 900 published papers on cardiovascular disease, over 600 on cancer, 250 on mental health and over 400 on obesity. Other broader research projects have published on the identification of novel patterns of association of many different exposures with different health outcomes. The rapid availability of more frequent healthcare records for COVID-19 research has already led to over 250 papers, which have attracted a high level of attention: Altmetric identifies over 20,000 mentions, with the most popular having a score of 2,820 and 89% of the mentions (typically on Twitter) to that paper being from members of the public.

Given the significant increase in the number of approved research applications to use UK Biobank data over the last two years, UK Biobank expects to see a corresponding increase in the number of research publications based upon analysis of UK Biobank data in the next couple of years. A full list of publications using UKB data can be accessed at https://www.ukbiobank.ac.uk/enable-your-research/publications.

The UK Biobank resource enables researchers to make scientific discoveries that provide considerable benefit to human health, not just for people living in the UK but worldwide. Much of this work relies upon data made available by NHS Digital. UK Biobank has approved over 2,900 research applications since it first opened to researchers in 2012. These cover a very broad range of health-related research topics (https://www.ukbiobank.ac.uk/enable-your-research/approved-research). There have been over 4,000 publications using UK Biobank data (https://www.ukbiobank.ac.uk/enable-your-research/publications), the majority within the last 2-3 years. There are over 2,400 on-going projects.

It is therefore relatively early to expect research supported by the UK Biobank resource to translate into widespread policy impact. Nonetheless, UK Biobank is at the heart of the UK Government’s strategy for incorporating genomics in healthcare (HM Government, 2020), and its value was highlighted in the NHS Long-Term Plan as one of the UK’s “outstanding capabilities for research and innovation” (NHS, 2019). As of early 2022, there had been a further 161 documents that refer to UK Biobank or research deriving from it.

The value of the UK Biobank resource for informing a wide range of policy areas is illustrated by the breadth of its utility: for example, there are references to research in many areas as disparate as the health effects of air pollution (PHE. 2020a) to Parliamentary briefing papers on loneliness and isolation (Bellis & Loft, 2020). One study on the association between active commuting and disease helped inform two Government policy documents on the future of urban mobility, cycling and walking (Department for Transport, 2019, 2020), whilst another on risk thresholds for alcohol consumption informed two WHO documents on prevention of alcohol-related harm and labelling policy (WHO. 2019, 2020b). Internationally, UK Biobank is referenced as a best practice model for biobanking policies. For example, WHO, 2020.

As outlined in the NHS Personalised Medicine Strategy, personalised medicine can radically transform medical care and population health through prevention and treatment strategies that are more closely targeted to the individual patient. The Government Healthcare Strategy (HM Government, 2020) recognises UK Biobank’s central role in the development of the polygenic risk score (PRS) concept, which involves using the combined effects of thousands of genetic variants (each of which has only a small effect) to improve risk prediction for common diseases. In terms of utility, UK Biobank data have been used to show that PRS for 16 common conditions (Genomics Plc, 2019) derived from a single low-cost genotype test can identify about one-quarter of the population who might benefit from such precision population health approaches.

Availability of such a wealth of genetic and phenotypic data (including participant characteristics [such as height], but also health outcomes derived from data from NHS Digital) has necessitated methodological innovations, such as fast multi-trait GWASs; these take into account the complex interactions between genetic variants and boost statistical power to detect genetic associations, the results of which are available via searchable online browsers (e.g., https://biobankengine.stanford.edu/ and http://geneatlas.roslin.ed.ac.uk/). The combination of healthcare data from linked health records and genomic data has led to huge advances in the use of Mendelian Randomisation approaches to understand the causality of epidemiological associations and thus contribute to appropriate preventative and therapeutic strategies. For example, such analyses provide strong support for a causal role of body fat mass with aortic valve stenosis and most other cardiovascular conditions (Larsson, et al. 2020), and do not support a protective role of vitamin D supplementation on cardiovascular or cancer outcomes (Meng, et al. 2019).

Other examples of impact on health practice deriving directly from UK Biobank include research showing that the incidence rate of type I diabetes remains about the same up to at least age 60 (Thomas, et al. 2018). As a consequence, clinicians are now alert to considering type I diabetes as a diagnosis in patients who present in older life, particularly when they do not achieve good glycaemic control with standard therapy for type 2 diabetes. Other research has shown that those who have two variant copies of a mutation leading to haemochromatosis are at substantially increased risk of developing liver cancer, rheumatoid arthritis, osteoarthritis and diabetes (Pilling, et al. 2019), leading to calls for the UK National Screening Committee to reconsider routine genetic testing for the condition. Research from UK Biobank has also shown that handgrip strength improves the risk prediction of cardiovascular incidence and mortality over and above that of the existing standard clinical risk score and is as predictive as high-density lipoprotein cholesterol and N-terminal pro B-type natriuretic peptide (Celis-Morales, et al. 2018). Thus, handgrip strength, which is easily measured objectively and is highly reproducible in any clinical setting, may add clinical utility over existing risk prediction scores for cardiovascular disease outcomes.

UK Biobank’s repurposing to support the national and international response to the COVID-19 pandemic (which included the increased frequency of hospitalisation and death data updates from NHS Digital combined with expedited access to the established research community of approved researchers) has already seen benefits. Already, 209 reports on COVID-19, which have been cited ~2,500 times, have been produced using the resource. These include studies of the genetic determinants of severe COVID-19 (Pairo-Castineria, et al. 2021), relationships of pre-existing adverse cardiac phenotypes determined from imaging scans with COVID-19 (Raisi-Estabragh, et al. 2021) and associations of NMR-metabolomic biomarkers for low-grade inflammation and cardiometabolic disease with severe COVID-19 (Julkunen, et al. 2021). Overall, these publications have attracted a high level of attention: To date, Altmetric identifies over 20,000 mentions, with the most popular having a score of 2,820 and 89% of the Twitter mentions to that paper being from members of the public. Some of the findings have translated into policy impact, including the NICE guidelines on vitamin D supplementation, UK Government reporting on the differential impact of COVID-19 among minority ethnic groups and a review of disparities in risks and outcomes (PHE, 2020b). Internationally, UK Biobank-enabled research has been cited by the WHO in their recent COVID-19 clinical management guidance (WHO, 2021), and by an EC-commissioned study on air pollution and COVID-19.

References

Bellis A, Loft P. Tackling loneliness. House of Commons Briefing Paper number 8514. House of Commons Library. 2020

Celis-Morales CA, et al. Associations of grip strength with cardiovascular, respiratory, and cancer outcomes and all-cause mortality: prospective cohort study of half a million UK Biobank participants. BMJ, 2018; 361: k1651

Department for Transport. Future of Mobility: Urban Strategy. 2019

Department for Transport. Gear Change: A bold vision for cycling and walking. 2020

Douaud G, et al. Brain imaging before and after COVID-19 in UK Biobank. medRxiv [Preprint]. 2021 Aug 18:2021.06.11.21258690

European Parliament, Directorate-General for Internal Policies of the Union, Smit L, et al. Air pollution and COVID-19 : including elements of air pollution in rural areas, indoor air pollution, vulnerability and resilience aspects of our society against respiratory disease, social inequality stemming from air pollution, European Parliament, 2021

HM Government. Genome UK: The future of healthcare. 2020

Julkunen H, et al. Metabolic biomarker profiling for identification of susceptibility to severe pneumonia and COVID-19 in the general population. Elife. 2021; 10: e63033

Larsson SC, et al. Body mass index and body composition in relation to 14 cardiovascular conditions in UK Biobank: a Mendelian randomization study. Europ Heart J, 2020; 41: 221-6

Meng X, et al. Phenome-wide Mendelian-randomisation study of genetically determined vitamin D on multiple health outcomes using the UK Biobank study. Int J Epidemiol; 2019; 48: 1425-34

NHS. The NHS Long Term Plan. 2019

NHS England. Improving Outcomes through Personalised Medicine. 2016

Pairo-Castineira E, et al. Genetic mechanisms of critical illness in COVID-19. Nature; 2021; 591, 92–98

Pilling LC, et al. Common conditions associated with hereditary haemochromatosis genetic variants: cohort study in UK Biobank. BMJ. 2019: 364; k5222

PHE. Disparities in the risks and outcomes of COVID-19. 2020b

PHE. Summary of the Committee on the Medical Effects of Air Pollutants COMEAP recommendations for the quantification of health effects associated with air pollutants. 2020a

Raisi-Estabragh Z, et al. Adverse cardiovascular magnetic resonance phenotypes are associated with greater likelihood of incident coronavirus disease 2019: findings from the UK Biobank. Aging Clin Exp Res; 2021; 33: 1133-1144

Thomas NJ, et al. Frequency and phenotype of type 1 diabetes in the first six decades of life: a cross-sectional, genetically stratified survival analysis from UK Biobank. Lancet Diabetes Endocrinol. 2018: 6; 122-129

WHO. Questions and Answers on the prevention and control of alcohol-related harm. 2019

WHO. Regulating the unknown: A guide to regulating genomics for health policy-makers. 2020a

WHO. What is the current alcohol labelling practice in the WHO European Region and what are barriers and facilitators to development and implementation of alcohol labelling policy? 2020b

WHO. COVID-19 Clinical management: living guidance. 2021

Aim and purpose of this application

UK Biobank is a major national and international health resource, and a registered charity in its own right, with the aim of improving the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye conditions, depression and forms of dementia. UK Biobank recruited 500,000 people aged between 40-69 years in 2006-2010 from across the United Kingdom to take part in this project. They have answered detailed questionnaires, undergone a range of physical measures, provided blood, urine and saliva samples for future analysis, and agreed to have their health followed-up over time through linkage to their health and health-related records. The UK Biobank resource is available to researchers worldwide who can apply to use the data (which are de-identified) within the resource to undertake health-related research that is in the public interest.

UK Biobank requires data from NHS Digital to pursue its legitimate interests, which are to set up and manage a major international research resource for health-related research that is in the public interest. UK Biobank’s Article 6 lawful basis for processing the personal data relating to the participants in the UK Biobank study is therefore Article 6 (1) (f) of the GDPR (legitimate interest). UK Biobank’s Article 6 lawful basis for processing the personal data relating to the participants in the UK Biobank study is therefore Article 6 (1) (f) of the GDPR (legitimate interest). UK Biobank processes special category personal data provided by NHS Digital in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. UK Biobank falls clearly within this category as its purpose is to improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses. The data are required for research purposes in the public interest – meeting the conditions on the DPA Schedule 1 Part 1 (4) – which GDPR Recital 52 (2) determines is an appropriate derogation from prohibition on processing special categories of personal data; proportionate to the aim pursued (as above). The ways in which the processing of data will be of benefit to the public – thereby demonstrating that the processing is in the public interest – are described in section ‘5d. ii. Expected Measurable Benefits to Health and/or Social Care Including Target Date’.

UK Biobank is a prospective study with the specific objective of following up the health of its participants through access to their health records. This resource enables researchers to study the genetic, social or economic determinants of both common and rare conditions. This in turn, will lead to the development of public health guidance for the purposes of primary prevention, development of early detection biomarkers (e.g., polygenic risk scores or imaging biomarkers of internal organs) and the development of new and improved treatments for patients with existing conditions, all of which will be of considerable benefit to the public. Without the health record data, UK Biobank is simply not able to achieve its aims.

How will the data requested achieve the aim identified?

The addition of the requested data from NHS Digital to the resource enhances the range of information that is available to researchers to enable them to undertake their approved research project. UK Biobank, as a charitable company limited by guarantee, was established to enable researchers to undertake health-related research which is in the public interest and therefore the data are required from NHS Digital to enable UK Biobank to pursue its legitimate interests.

As a result of UK Biobank’s work, patients and the wider public benefit from the advances made in the prevention, diagnosis and treatment of disease. UK Biobank is now one of the largest and most used health research resources in the world. 21,000 researchers are registered with UK Biobank, from 2,300 approved institutions, and 2,600 health-related research applications have been approved to date.

Background to the request / the wider study

UK Biobank is a unique resource due to the combination of its very large size and the range and detail of data that it contains on its participants, including extensive phenotype and genotype information on each of them. Furthermore, rather than recruiting participants with specific diseases or risk factors, UK Biobank is a prospective population-based resource established to support research into the causes of a wide range of diseases affecting people in middle and older age. The ability to link to participant’s health records, so that specific outcomes occurring during long-term participant follow-up can be identified and appropriate disease-based research carried out, is therefore critical.

UK Biobank’s priority is to combine extensive and precise measurement of exposures with detailed and rigorous follow-up for a wide range of health-related outcomes, and to promote innovative science by maximising secure access to de-identified data.

UK Biobank was established in 2003. It is funded by, amongst others, the Wellcome Trust, and the Medical Research Council. No funder is permitted to influence the outcomes nor suppress any outcomes of research.

Access to the resource opened in 2012, since which time 2,600 applications have been approved and 21,000 researchers have registered to use the data. As a result, over 2,000 research papers have been published to date using the UK Biobank resource. The UK Biobank resource is available to all bona fide researchers for all types of health-related research that is in the public interest. All researchers, whether in universities, charities, government agencies or commercial companies, and whether based in the UK or abroad, are subject to the same application process and approval criteria.

UK Biobank charge a fee for access to the resource which is calculated on a cost-recovery basis. UK Biobank seek to recover only the incremental costs of servicing an application to access the resource. Further information about the fees charged can be found here https://www.ukbiobank.ac.uk/enable-your-research/costs.

As explained above in section 5b, the UK Biobank resource is available to all bona fide researchers for all types of health-related research that is in the public interest. All researchers, whether in universities, charities, government agencies or commercial companies, and whether based in the UK or abroad, are subject to the same application process and approval criteria.

The expected and yielded benefits of allowing researchers to access the UK Biobank resource and the linked healthcare data provided by NHSD are described in section 5d.

The data subjects

Between 2006 and 2010, UK Biobank recruited 500,000 people from the general population - known as “participants” - then aged between 40 and 69 to take part in the project. The participants are the data subjects for the purpose of this application.

The methods used for the identification, invitation and assessment of participants were developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All participants in the UK Biobank study have given fully informed consent for: access to and long-term storage of information from their medical and other health-related records; use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purposes; and re-contact by UK Biobank to invite them (but without any obligation) to answer further questions or to take part in further assessments.

It is considered very unlikely that participants would object to the processing activities involved in obtaining linked data from NHS Digital. Participants recruited can withdraw from UK Biobank at any time, although since UK Biobank began recruitment in 2006 only a small number have done so.

The purpose of the request

Under this Agreement, UK Biobank receives updates on participants’ health and health-related data. These include:

i. Hospital Episode Statistics – details of participants’ attendance at hospital for any reason. 1997/98 to latest available annual update (updated to latest available on a monthly basis to assist research involving COVID-19). HES Outpatients required up to 2021;

ii. Mental Health data – record-level data about the care of participants who are in contact with mental health, learning disabilities or autism spectrum disorder services. Mental Health Minimum Data Set 2006/07 to 2014/15, Mental Health and Learning Disabilities Data Set 2014/15 to November 2015/16, Mental Health Services Data Set 2016/17 to latest available annual update;

iii. Diagnostic Imaging data – information about the imaging tests and scans carried out on participants such as x-rays, MRI scans or ultrasound. 2012/13 to 2020;

iv. Civil Registration Mortality data – details of participants’ deaths including date and cause. 2011/12 to latest available annual update (updated to latest available on a monthly basis to assist research involving COVID-19);

v. Cancer Registration data – details of any cancers participants are registered as having. 2011/12 to latest available with an annual update;

vi. Emergency Care Data Set data - details of participant attendance in A&E. October 2017 to latest available annual update (updated to latest available on a monthly basis to assist research involving COVID-19);

vii. National Diabetes Audit data - data collected on diabetes in participants. April 2003 to latest available annual update ;

viii. GPES Data for Pandemic Planning and Research - GDPPR (COVID-19) – details of participants’ clinical events and prescriptions. October 2020 to 2021.

UK Biobank is also requesting COVID-19 Hospitalization in England Surveillance System (CHESS) and COVID-19 Vaccination Status. These data will enable comprehensive data about COVID-19 infection in persons requiring hospitalisation to be made available to approved UK Biobank researchers who, in combination with other UK Biobank data, wish to undertake research into the determinants of severe COVID-19 infection and the impact of vaccination on disease outcomes.

For clarity, some of the data required monthly (i.e. death, hospitalisations) are not restricted to COVID-19 related research. However the CV19 specific datasets (vaccinations, CHESS, GPDRR are all restricted in terms of use for CV19 related research only.

UK Biobank is also not currently receiving data from the GP systems (i.e. EMIS and TPP) for non-COVID-19 research.**

ix. Personal Demographic Service (PDS) updates – details of changes to participants’ demographics, postal addresses or registered GP practices. It is very important for UK Biobank to be able to maintain accurate information about participants’ contact details (including postal address) and registered GP practice (which may have changed since participant recruitment), for the following reasons:

- Ongoing approved linkages to primary care data require accurate information about registered GP practice (at practice as well as CCG level) to ensure that data about a particular participant are requested from the correct GP practice.

- Direct contact with GPs is required as part of the UK Biobank imaging sub-study (www.ukbiobank.ac.uk/explore-your-participation/contribute -further/imaging-study) in the event that a participant is identified with a potentially serious incidental finding (as noted by the radiographer and then confirmed by a radiologist). Should such an abnormality be found, UK Biobank will write to the participant and their GP to allow them to further investigate these potentially serious findings.

- Follow-up in a cohort study needs to be as complete as possible, to avoid both unnecessary reduction in cohort numbers (which reduces statistical power of data analyses) and to minimise loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on intermittent re-contact with UK Biobank participants to invite them to complete brief questionnaires about health-related exposures or outcomes, or to other enhancement projects (such as the longitudinal and cross-sectional coronavirus antibody studies conducted in 2020/21), for which correct participant contact details are required. Although participants are encouraged to inform UK Biobank when they change address, this is not a comprehensive method of ensuring that the most up-to-date contact information is held on all participants.

- UK Biobank is committed to ensuring that participants are informed of progress and developments in UK Biobank, to obtaining participants’ opinions on proposed enhancements to the study, and to providing them with information on how to contact UK Biobank should they have any questions. UK Biobank does this in part through its annual newsletter, sent via e-mail and/or postal mail, for which up-to-date contact details are required. UK Biobank prefers to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress.

- Important potential research questions about the effects of environmental exposures on health require information about current and previous postal address to allow these exposures to be estimated (from postcodes). These include, for example, valuable studies of the health effects of air and noise pollution, living close to power cables, and aspects of the local environment such as nearby local public amenities and green space.

As a large cohort, it is not feasible for UK Biobank itself to follow up regularly each of its participants on an individual basis and nor would this generate the necessary clinical detail. Self-reported data do not possess (anything like) the needed coverage, reliability and consistency that health records contain. As such, the data provided via NHS Digital are crucial to the viability of UK Biobank as the primary mechanism for follow-up of the cohort, particularly as they provide information on all 500,000 participants in an entirely unobtrusive (i.e. passive) manner. The linked datasets all form part of the participants’ health record and are thus vital to providing the necessary health information. Of key importance are datasets related to death and cancer registry notifications and hospital activity, as they provide crucial information on which participants have developed which diseases and at what time. Without access to these data, UK Biobank could not function as a prospective health resource and therefore research on the associations between social or economic determinations of health and genetic factors and the risk of developing a wide range of diseases over time would not be possible. UK Biobank is a prospective study available to registered researchers worldwide, whose research interests cover a very wide range of health outcomes, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke and mental health. UK Biobank currently releases data from linked medical records on an approximately monthly basis to the research community in order to facilitate COVID-19 research. As such, access to data on deaths, cancers and hospital admissions and related activities is of vital importance.

In addition to the above-mentioned datasets that UK Biobank holds, other health-related datasets that have now become available are also of significant research interest. The IAPT was requested in the 2020 application but current data has not yet been received. This dataset will help identify participants with mental health conditions and enable research into their treatment response and support required. The COVID-19 datasets will provide an objective measure of vaccination status and enable further research into the impacts of SARS-CoV-2 infection (i.e. long-COVID) across the full disease spectrum and the longer-term impacts of vaccination.

- UK Biobank requires access to all retrospective data to enable epidemiological research into antecedent clinical events prior to diagnosis and the effect of multi-morbidity on future health outcomes. Without these linked data provided by NHS Digital, researchers would not be able to ascertain the full burden of disease in the cohort nor the effects of particular treatment and management pathways on health. All of the datasets requested enhance the UK Biobank resource and allow researchers to undertake important health-related research that is in the public interest.

For each of these datasets listed above, UK Biobank requires linkage only for the 500,000 UK Biobank participants who joined the study and who were recruited across the UK. It should be noted that UK Biobank is only requesting individual-level identifying data to ensure correct linkage. UK Biobank does not make available to researchers any identifiable data (such as full postcode, date of birth, name, NHS number, etc.); each research project is instead provided with a unique encrypted 9-digit identification number for participants. Hence, it should not be possible for a researcher to identify a UK Biobank participant using the data that UK Biobank releases.

The organisations involved

UK Biobank is the data controller of the data supplied by NHS Digital. Pursuant to a contract between UK Biobank and The Clinical Trials Service Unit (CTSU) at the University of Oxford, CTSU acts as a data processor of UK Biobank’s data. CTSU is only able to act on the instructions of UK Biobank and is thus only a data processor. CTSU has no ability to process the data for any other purpose save for those purposes which UK Biobank (alone) determines.

The data supplied by NHS Digital are stored and processed at the CTSU under the sole direction of UK Biobank and pursuant to the contract between UK Biobank and CTSU (which confirms these data controller / data processor arrangements).

UK Biobank is a company limited by guarantee and registered at Companies House; it therefore has its own legal personality. While it receives funding from the Wellcome Trust and the Medical Research Council amongst others, these organisations have no access to the data and cannot exercise any control in respect of its use.

Once NHS Digital data have been received by UK Biobank (acting through its processor CTSU), they are uploaded to the UK Biobank resource. Registered researchers can then apply to UK Biobank to use these de-identified linked data (alongside the other data in the resource) to undertake health-related research that is in the public interest. Each application to use the resource is considered by UK Biobank pursuant to UK Biobank access policy and procedures, which were developed through extensive consultation (including public consultation and consultation with UK Biobank participants). Further detail is given section 5b. UK Biobank has sole control of this process.

De-identified data are made available to researchers following approval of a research application once they have signed a contract with UK Biobank (known as a Material Transfer Agreements [MTA]) which controls how the data can be used (including ensuring the data are only used for conducting health-related research in the public interest). The de-identified data are either made available directly to the researcher (for them to download and use in their own environment) or the researcher is granted access to use the data through UK Biobank’s Research Analysis Platform (RAP).

The RAP comprises a web interface provided by DNAnexus and cloud storage provided by Amazon Web Services (AWS). The only data that are made available through the RAP (and securely stored by AWS in the cloud) are the de-identified data that have historically been made available directly to researchers for them to download. Once the data are in the RAP, they are only accessible to researchers who have an approved research project and who have entered into an MTA with UK Biobank. DNAnexus is ISO27001 certified by independent third-party auditors and maintains compliance with ISO27001 (alongside compliance with many internationally recognised standards such as NIST, FedRAMP and HITRUST). The platform utilises the AWS infrastructure from the London node - the UK Biobank data are therefore stored in London and when the data are used/analysed on the platform they will remain in London. The dataset is stored using robust security encryption protocols. Although the platform provider (DNAnexus) and the cloud supplier (AWS) only receive de-identified data, the agreement that UKB has with DNAnexus does require DNAnexus to treat the UKB data to the same standard as if they are personal data and thus contains the ICO-recommended GDPR-compliant data processing clauses. UKB has permitted DNAnexus to appoint AWS on the same basis as UKB has appointed DNAnexus.

UK Biobank’s main output is enabling access to the resource for the purposes of health-related research, of which linkage to longitudinal health records is key. Data are provided to bona fide researchers worldwide according to the UK Biobank access policy and procedures, as described in section 5b.

During the last five years, the characterisation of the participants has been enhanced substantially (e.g. with genetic sequence and biochemical assay data, as well as detailed imaging measures) and much larger numbers of incident cases of many different conditions have been identified from the NHSD linked healthcare data. This has resulted in an increasingly large number of researchers conducting and reporting an increasingly extensive range of research, and the range of health conditions that it will be possible to study reliably using even more detailed information about exposures will increase further as the numbers of incident cases of different health outcomes increase during the next 5-10 years.

UK Biobank’s current enhancement projects and specific outputs:

As part of UK Biobank’s mission to enable scientific discoveries to improve human health, UK Biobank started a SARS-CoV-2 antibody study in Feb 2021, which aims to collect SARS-CoV-2 antibody status data from all participants who agree to be provided with a home-based lateral flow test. These data, coupled with the existing genetic and social determinations of health data and linkage to future health records for all 500,000 participants, will be uploaded to the resource in June/July 2021 and will create a unique dataset to enable research into the longer-term health effects of SARS-CoV-2, known as ‘long-covid’.

UK Biobank’s imaging study started in 2014 with the aim of collecting multi-modal MRI data on 100,000 UK Biobank participants, with repeat scans being performed in up to 70,000 participants a few years later, and which will run for the next 5-6 years. In February 2021, UK Biobank set up a SARS-CoV-2 repeat imaging project to create a unique resource to enable research into the effects of SARS-CoV-2 infection on internal organs and associated pathophysiology and health outcomes. The intention is to invite up to 3,000 participants who have already attended an imaging assessment for a repeat visit, half of whom have evidence of previous SARS-CoV-2 infection (as identified through linkage to health records or an antibody test) and half of whom do not. This project will run until Summer 2021 and the associated MRI and health record data are being made available to researchers on an approx monthly basis to facilitate rapid COVID-19 research. UK Biobank is the only resource globally able to provide this type of large-scale imaging data for participants both before and after infection.

In order to facilitate research into the genetic and environmental determinants of COVID-19, since April 2020 UK Biobank has been releasing more regular (approx. monthly) updates of linked health record data. Much of these data come from NHSD, including death records, hospital inpatients and critical care records. Consequently, it is uniquely placed to allow scientists nationally and internationally to answer questions about the wide-ranging health consequences of SARS-CoV-2 infection.

Research applications and approved research projects:

To date, 20,000 researchers have successfully registered to use the UK Biobank resource from across 2,300 research institutions worldwide. The rate of growth of approved research applications to use UK Biobank data has been substantial, growing from 16 applications in 2012 to 655 in 2020 alone, such that there are now 2,600 approved research applications that are using UK Biobank data for health-related research that is in the public interest. This increase has been notable over the last 5 years, with a 4-fold increase in applications to access the resource since 2016, particularly from researchers outside of the UK (80% in 2020 compared with 45% in 2016) – predominantly the USA (31%), the rest of Europe (16%) and China (15%) – although UK-based use remains high (20%). The majority of applications are still led by academic researchers, but an increasing number are coming from commercial entities (78 applications in 2020; 9%).

Summaries of research projects that have been approved and are currently underway can be found at https://www.ukbiobank.ac.uk/approved-research/. The nature of the research that is currently being conducted reflects the increasing breadth and depth of data that are now available, with 77% of applications requesting genotype data, 60% requesting imaging data, and 92% requesting health outcome data. Of note, over 700 research groups have accessed the health record data for the purposes of performing COVID-19 research in the last 12 months.

Publications:

To date, 2,000 publications covering over forty areas of interest have been produced, and the number of papers published each year is increasing rapidly (from 89 in 2016 to over 800 in 2020). In relation to health outcomes, there are about 300 published papers on cardiovascular disease, 250 on cancer, 150 on mental health and 100 on obesity, while other, broader, research projects have published on the identification of novel patterns of association of many different exposures with different health outcomes. The rapid availability of more frequent healthcare records for COVID-19 research has already led to 150 papers, which have attracted a high level of attention: Altmetric identifies over 12,800 mentions, with the most popular having a score of 1,181 (99th percentile) and three quarter of the mentions (typically on Twitter) to that paper being from members of the public.

Given the significant increase in the number of approved research applications to use UK Biobank data over the last two years, UK Biobank expects to see a corresponding increase in the number of research publications based upon analysis of UK Biobank data in the next couple of years. A full list of publications using UKB data can be accessed at https://www.ukbiobank.ac.uk/enable-your-research/publications.

The UK Biobank resource enables researchers to make scientific discoveries that provide considerable benefit to human health, not just for people living in the UK but worldwide. Much of this work relies upon data made available by NHS Digital. UK Biobank has approved 2,600 research applications since it first opened to researchers in 2012. These cover a very broad range of health-related research topics that are in the public interest (https://www.ukbiobank.ac.uk/enable-your-research/approved-research). There have been over 2,000 publications by researchers using UK Biobank data (https://www.ukbiobank.ac.uk/enable-your-research/publications), most of which have been published in the last 2-3 years.

It is therefore relatively early to expect research supported by the UK Biobank resource to translate into widespread policy impact. Nonetheless, UK Biobank is at the heart of the UK Government’s strategy for incorporating genomics in healthcare (HM Government. 2020), and its value was highlighted in the NHS Long-Term Plan as one of the UK’s “outstanding capabilities for research and innovation” (NHS. 2019). There have been a further 128 documents that refer to UK Biobank or research deriving from it.

The value of the UK Biobank resource for informing a wide range of policy areas is illustrated by the breadth of its utility: for example, there are references to research in many areas as disparate as the health effects of air pollution (Public Health England. 2020a) to Parliamentary briefing papers on loneliness and isolation (Bellis & Loft. 2020). One study on the association between active commuting and disease helped inform two Government policy documents on the future of urban mobility, cycling and walking (Department for Transport. 2019, 2020), whilst another on risk thresholds for alcohol consumption informed two WHO documents on prevention of alcohol-related harm and labelling policy (WHO. 2019, 2020b). Internationally, UK Biobank is referenced as a best practice model for biobanking policies. For example, WHO’s recent policy briefing on regulating genomics espouses UK Biobank’s approach to consent as a best-practice exemplar (WHO. 2020).

As outlined in the NHS Personalised Medicine Strategy, personalised medicine can radically transform medical care and population health through prevention and treatment strategies that are more closely targeted to the individual patient. The Government Healthcare Strategy (HM Government. 2020) recognises UK Biobank’s central role in the development of the polygenic risk score (PRS) concept, which involves using the combined effects of thousands of genetic variants (each of which has only a small effect) to improve risk prediction for common diseases. In terms of utility, UK Biobank data have been used to show that PRS for 16 common conditions (https://www.genomicsplc.com/wp-content/uploads/2020/11/Genomics-plc-PRS-details_White-Paper-April-2019.pdf) derived from a single low-cost genotype test can identify about one-quarter of the population who might benefit from such precision population health approaches.

Availability of such a wealth of genetic and phenotypic data (including participant characteristics [such as height], but also health outcomes derived from data from NHS Digital) has necessitated methodological innovations, such as fast multi-trait GWASs; these take into account the complex interactions between genetic variants and boost statistical power to detect genetic associations, the results of which are available via searchable online browsers (e.g., https://biobankengine.stanford.edu/ and http://geneatlas.roslin.ed.ac.uk/). The combination of healthcare data from linked health records and genomic data has led to huge advances in the use of Mendelian Randomisation approaches to understand the causality of epidemiological associations and thus contribute to appropriate preventative and therapeutic strategies. For example, such analyses provide strong support for a causal role of body fat mass with aortic valve stenosis and most other cardiovascular conditions (Larsson, et al. 2020), and do not support a protective role of vitamin D supplementation on cardiovascular or cancer outcomes (Meng, et al. 2019).

Other examples of impact on health practice deriving directly from UK Biobank include research showing (based on known genetic susceptibility variants) that the incidence rate of type I diabetes remains about the same up to at least age 60 years (Thomas, et al. 2018). As a consequence, clinicians are now alert to considering type I diabetes as a diagnosis in patients who present in older life, particularly when they do not achieve good glycaemic control with standard therapy for type 2 diabetes. Other research has shown that those who have two variant copies of a mutation leading to haemochromatosis are at substantially increased risk of developing liver cancer, rheumatoid arthritis, osteoarthritis and diabetes (Pilling, et al. 2019), leading to calls for the UK National Screening Committee to reconsider routine genetic testing for the condition. Research from UK Biobank has also shown that handgrip strength, which was measured at the recruitment assessment clinics for the full cohort, improves the risk prediction of cardiovascular incidence and mortality over and above that of the existing standard clinical risk score and is as predictive as high-density lipoprotein cholesterol and N-terminal pro B-type natriuretic peptide (Celis-Morales, et al. 2018). Thus, handgrip strength, which is easily measured objectively and is highly reproducible in clinical settings – including in low- and middle-income countries – may add clinical utility over existing risk prediction scores for cardiovascular disease outcomes.

UK Biobank’s repurposing to support the national and international response to the COVID-19 pandemic (which included the increased frequency of hospitalisation and death data updates from NHS Digital combined with expedited access to the established research community of approved researchers) has already seen benefits in the form of policy impact. Already, 225 reports on COVID-19, which have been cited 2,500 times, have been produced based on the UK Biobank resource. These include studies of the genetic determinants of severe COVID-19 (Pairo-Castineria, et al. 2021), relationships of pre-existing adverse cardiac phenotypes determined from imaging scans with COVID-19 (Raisi-Estabragh, et al. 2021) and associations of NMR-metabolomic biomarkers for low-grade inflammation and cardiometabolic disease with severe COVID-19 (Julkunen, et al. 2021). Overall, these publications have attracted a high level of attention: Altmetric identifies over 17,400 mentions, with the most popular having a score of 1,587 (99th percentile) and over 90% of the Twitter mentions to that paper being from members of the public. Some of the findings have translated into policy impact during the pandemic, including the NICE guidelines on vitamin D supplementation, UK Government reporting on the differential impact of COVID-19 among minority ethnic groups and a review of disparities in risks and outcomes (Public Health England, 2020b). Internationally, UK Biobank-enabled research has been cited by the WHO in their recent COVID-19 clinical management guidance (WHO. 2021), and by an EC-commissioned study on air pollution and COVID-19. Given UK Biobank’s success in pivoting so effectively in response to SARS-CoV-2, Eric Lander (Presidential Science Adviser to President Biden, and Director of the US Office of Science and Technology Policy) suggested it should be a key part of the UK’s pandemic preparedness plans.

References

Bellis A, Loft P. Tackling loneliness. House of Commons Briefing Paper number 8514. House of Commons Library. 2020. Available at: https://commonslibrary.parliament.uk/

Celis-Morales CA, et al. Associations of grip strength with cardiovascular, respiratory, and cancer outcomes and all-cause mortality: prospective cohort study of half a million UK Biobank participants. BMJ, 2018; 361: k1651.

Department for Transport. Future of Mobility: Urban Strategy. 2019. Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/846593/future-of-mobility-strategy.pdf

Department for Transport. Gear Change: A bold vision for cycling and walking. 2020. Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/904146/gear-change-a-bold-vision-for-cycling-and-walking.pdf

HM Government. Genome UK: The future of healthcare. 2020. Available at:

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/920378/Genome_UK_-_the_future_of_healthcare.pdf’

Julkunen et al. Metabolic biomarker profiling for identification of susceptibility to severe pneumonia and COVID-19 in the general population. 2020 https://doi.org/10.7554/eLife.63033.sa2

Larsson SC, et al. Body mass index and body composition in relation to 14 cardiovascular conditions in UK Biobank: a Mendelian randomization study. Europ Heart J, 2020; 41: 221-6.

Meng X, et al. Phenome-wide Mendelian-randomisation study of genetically determined vitamin D on multiple health outcomes using the UK Biobank study. Int J Epidemiol; 2019; 48: 1425-34.

NHS. The NHS Long Term Plan. 2019. Available at: www.longtermplan.nhs.uk

Pairo-Castineira, E., et al. Genetic mechanisms of critical illness in COVID-19. Nature; 2021; 591, 92–98.

Pilling LC, et al. Common conditions associated with hereditary haemochromatosis genetic variants: cohort study in UK Biobank. BMJ. 2019: 364; k5222.

Public Health England. Summary of the Committee on the Medical Effects of Air Pollutants COMEAP recommendations for the quantification of health effects associated with air pollutants. 2020a. Available at: https://www.gov.uk/government/publications/air-pollutants-quantification-of-associated-health-effects

UK Biobank is a major national and international health resource, and a registered charity in its own right, with the aim of improving the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye disorders, depression and forms of dementia. UK Biobank recruited 500,000 people aged between 40-69 years in 2006-2010 from across the country to take part in this project. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed via linkage to their health and health related records. Over many years this will continue to build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

UK Biobank requires data from NHS Digital to pursue its legitimate interests which are to set up and manage a major international research resource for health-related research that is in the public interest. UK Biobank is a prospective resource with the specific objective of following up its participants through the health records. This enables researchers to study the actual causes of disease and in turn develop treatments (for common and rare disease) which will be of considerable benefit to the public. Without the health record data the UK Biobank resource is simply not able to achieve its aims. This would, in UK Biobank’s view, constitute a (very)substantial public interest.

The UK Biobank resource is open to researchers who can apply to use the data within the resource (which data is de-identified) to undertake health related research which is in the public interest. The addition of the requested data from NHS Digital to the resource enhances the range of information which is available to researchers to enable them to undertake their approved research project. UK Biobank, as a charitable company limited by guarantee, was established to enable researchers to undertake health related research which is in the public interest and therefore the data is required from NHS Digital to enable UK Biobank to pursue its legitimate interests.

As a result of UK Biobank’s work, patients and the wider public benefit from the advances made in the prevention, diagnosis and treatment of disease. UK Biobank is now one of the largest and most used health research resources in the world. Over 13,000 researchers are registered with UK Biobank and over 1,600 health-related research applications have been approved.

Participants have signed consent forms permitting access to their data for this purpose. It is considered very unlikely that participants would object to the processing activities involved in obtaining linked data from NHS Digital. Participants recruited have the ability to withdraw at any point. Since UK Biobank started to recruit participants in 2006 a small number of those initially recruited have asked to withdraw and asked that their information be deleted. For clarity, UK Biobank relies on GDPR 6 (1) (a) - explicit consent for the collection and storage of bio samples and the permission to make available for health related research. The data from NHS Digital is secondary health care data and therefore UK Biobank relies on Legitimate Interests (GDPR Article 6(1)(f) ) in regards to the data they obtain from NHS Digital.

UK Biobank processes the special category personal data provided by NHS Digital in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. UK Biobank falls clearly within this category as its purpose is to improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye disorders, depression and forms of dementia.

Lifestyle and environmental information, medical history, physical measurements, and biological samples are to be collected from about 500,000 people aged 40 to 69 at presentation and then, with consent, their health will be followed for many years through medical and other health related records. The biological samples will be stored so that they can be used for a wide range of biochemical and genetic analyses in the future.

The data are stored in a way that makes it is extremely difficult even for UK Biobank and their data processors (University of Oxford) to re-identify participants. Only a very few individuals within the University of Oxford are allowed to do so (and they are strictly monitored) in order that further information about participants can be added. Data provided to researchers have personal identifiers removed so that an individual participant cannot be identified. There are no circumstances in which data can be processed in a manner that could have an adverse impact on the data subject.

The overall purpose of UK Biobank is to create a prospective epidemiological resource of 500,000 people aged 45 - 69 at the time of recruitment from around the UK. The methods to be used for the identification, invitation and assessment of participants have been developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All participants in the UK Biobank study have given fully informed consent for access to and long-term storage of information from their medical and other health-related records, and use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purpose; and recontact by UK Biobank to invite them (but without any obligation) to answer further questions or take part in further assessments.

UK Biobank is a unique resource, because of the combination of its very large size and the range and detail of data that it contains on its participants, including extensive phenotype and genotype information on each of them. Further, rather than recruiting participants with specific diseases or risk factors, UK Biobank is a prospective population-based resource established to support research into the causes of a wide range of diseases affecting people in middle and older age. Thus, the ability to link to participant’s health records, so that specific outcomes occurring during long-term follow-up of the participants can be identified and appropriate disease-based research carried out, is critical.

UK Biobank’s priority is to combine extensive and precise measurement of exposures, along with the detailed and rigorous follow-up for a wide range of health related outcomes, and to promote innovative science by maximising secure access to deidentified data.

UK Biobank was established in 2003. It is funded by the Wellcome Trust, the Department of Health and the Medical Research Council. No funder is permitted to influence the outcomes nor suppress any outcomes of research.

The recruitment of its 500,000 participants took place between 2007 and 2010, whose health is now being followed long term primarily through linking to NHS health data. Access to the resource opened in 2012, since which time there have been over 1,600 applications submitted and over 13,000 researchers have registered to use the resource, resulting in 1,000 research papers published using the UK Biobank resource.

At baseline (i.e. at time of recruitment), UK Biobank obtained and holds identifiable information on its participants. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed by linking to their health and health related records. Over many years this will build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

Under this Agreement, UK Biobank receives updates on participants’ health and health-related data. This includes:

i. Hospital Episode Statistics – details of participants attendances at hospital for any reason. 1997/98 to latest available annual year (updated to latest available on a monthly basis to assist research involving COVID-19);

ii. Mental Health data – contains record-level data about the care of participants who are in contact with mental health, learning disabilities or autism spectrum disorder services. Mental Health Minimum Data Set 2006/07 to 2014/15, Mental Health and Learning Disabilities Data Set 2014/15 to November 2015/16, Mental Health Services Data Set 2016/17 to latest available annual year;

iii. Diagnostic Imaging data – collects information about the imaging tests and scans carried out on participants such as x-rays, MRI scans or ultrasound. 2012/13 to latest available annual year;

iv. Civil Registration Mortality data – details of participants’ deaths including date and cause. 2011/12 to latest available annual update (updated to latest available on a monthly basis to assist research involving COVID-19);

v. Cancer Registration data – details of any cancers participants’ are registered as having. 2011/12 to latest available with a bi-annual update;

vi. Emergency Care Data Set data - details around participant attendance in A&E. October 2017 to latest available annual update. (updated to latest available on a monthly basis to assist research involving COVID-19);

vii National Diabetes Audit data - data collected on diabetes in participants. April 2003 to latest available annual year;

viii. Improving Access to Psychological Therapies (IAPT)- measure of anxiety and depression in participants. April 2012 to latest available annual year;

** AMENDMENT REQUEST AUGUST 2020

To add GDPPR data so that they can be made available to UKB approved researchers for the purposes of COVID-19 research.

It is UK BioBanks intention that the GDPPR data would replace the existing data flows rather than supplement it. The reasons for this are three fold:

• The GDPPR data would ensure a more complete coverage of GP Practices than is the case with the current data flows from TPP and EMIS.

• The GDPPR data extract is a single standardised dataset which UK BioBank would be able to process more efficiently and rapidly than is the case with the separate datasets that are currently obtain from TPP & EMIS (that require bespoke processing and validation checks).

• UK BioBank currently release data from linked medical records (hospitalisation, deaths, diagnostic test results) on an approximately monthly basis to the research community to facilitate COVID-19 research. UK BioBank would aim to release GDPPR data on a two-monthly basis, owing to the volume of data processing required. This will help researchers to ascertain less severe cases of COVID-19 and to better understand the role of underlying health conditions in the presentation and progression of the disease.

For clarity, the data that are now required on a monthly basis (i.e. death, hospitalisations) is not restricted to COVID-19 related research. UK Biobank are also not currently receiving data from the GP system suppliers (i.e. EMIS and TPP) for non-COVID-19 research.**

As a large cohort it is not feasible for UK Biobank itself to follow up each of its participants on an individual basis regularly and nor would this generate the necessary clinical detail. Self-reported data does not possess (anything like) the needed coverage, reliability and consistency, that the health records contain. As such, the data provided via NHS Digital is crucial to the viability of UK Biobank as the primary mechanism for follow up of the cohort, particularly as it provides coverage for all 500,000 participants in an entirely unobtrusive (i.e. passive) manner. The linked datasets all form part of the participants’ health record and are thus all vital to providing the necessary health record information. Of key importance are datasets related to death and cancer registry notifications and hospital activity. Without access to these data, UK Biobank would not be a prospective resource and research on the associations of lifestyle and genetic factors on the risk of developing a wide range of diseases over time would simply not be possible. This is because these datasets provide vital information on which participants have developed which diseases and at what time. Given UK Biobank is a prospective study available to registered researchers worldwide, research interest covers a very wide range of health outcomes, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke, mental health, and many others. As such, access to data on deaths, cancers and hospital admissions and related activities is of vital importance.

In addition to the above-mentioned datasets that UK Biobank hold, other health-related datasets that have now become available are also of high research interest. For example, the IAPT and mental health datasets will help to identify participants with mental health disorders and enable research into their treatment response and support required. The diagnostic imaging dataset will enable research into the treatment pathways for a wide range of conditions, while the emergency department data will enable research into the reasons for and consequences of acute admissions. Finally, the national diabetes audit will help to identify participants diagnosed with diabetes and their management pathways (as these participants are often under-represented in the HES data). UK Biobank require access to all retrospective data to enable epidemiological research into antecedent clinical events prior to diagnosis and is crucial for researchers assessing the effect of multi-morbidity on future health outcomes. Without these linked data provided by NHS Digital, researchers would not be able to fully ascertain the full burden of disease in the cohort or effects of particular treatment and management pathways on health. All of the datasets requested enhance the UK Biobank resource and allow researchers to undertake health-related research that is in the public interest.

Other datasets include:

ix. Personal Demographic Service (PDS) updates – details of changes to participants’ demographics, postal addresses or registered GP practices.

It is very important for UK Biobank to be able to maintain accurate information about participants’ contact details (including postal address) and registered GP practice (which were up to date for the whole cohort at the time of recruitment but there will be undoubtedly have been changes since then). There are several reasons for this requirement:

- Ongoing approved linkages to primary care data require accurate information about registered GP practice (at practice as well as CCG level) to ensure that data about a particular participant are requested from the correct GP practice.

- Direct contact with GPs is required as part of the UK Biobank imaging sub study (https://imaging.ukbiobank.ac.uk/) in the event that a participant is identified with a potentially serious incidental finding (as noted by the radiographer and then confirmed by a specialist doctor (radiologist)). Should such an abnormality be found, UK Biobank will write to the participant and their GP to allow them to further investigate these potentially serious findings.

- To enable GPs to be confident about extraction of primary care data from GP practice system suppliers to UK Biobank, UK Biobank will need to be able to provide GP practices with up-to-date listings of UK Biobank participants registered at their practices, and – where requested – evidence of their consent.

- Follow-up in a cohort study needs to be as complete as possible. This is to avoid both unnecessary reduction in size of the cohort (which reduces statistical power of data analyses), and loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on intermittent re-contact with UK Biobank’s participants to invite them to complete brief questionnaires about health related exposures or outcomes. It is important that UK Biobank have correct and up-to-date participant contact details for this. Although UK Biobank encourage participants to inform UK Biobank when they move, this is not a failsafe method of ensuring the most accurate information.

- UK Biobank are committed to ensuring participants continue to be kept aware of progress and developments in UK Biobank, to obtaining their opinions on proposed enhancements to the study, and to providing them with information on how to contact UK Biobank should they have any questions. UK Biobank do this in part through UK Biobank’s annual newsletter, sent via e-mail and/or postal mail. Clearly, UK Biobank need up-to-date contact details (especially postal address) to do this. UK Biobank would like to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress.

- Important potential research questions about the effects of environmental exposures on health require information about current and previous postal address since this allows these exposures to be estimated (from postcodes). These include, for example, valuable studies of the health effects of air and noise pollution, living close to power cables and aspects of the local environment, such as nearby local public amenities and green space.

For each of these datasets listed above, UK Biobank require linkage only for the 500,000 UK Biobank participants who joined the study and who were recruited across the UK. It should be noted that the UK Biobank are only requesting individual-level identifying data to ensure correct linkage. UK Biobank do not make available to researchers any identifiable data (such as postcode, date of birth, name, NHS number, etc.) and instead, each research project is provided with a unique encrypted 9-digit identification number. Hence, it is not possible for a researcher to identify a UK Biobank participant using the data that UK Biobank release.

UK Biobank is the data controller of the data supplied by NHS Digital. Pursuant to a contract between UK Biobank and The Clinical Trials Service Unit (CTSU) at the University of Oxford, CTSU act as a data processor of UK Biobank’s data. CTSU is only able to act on the instructions of UK Biobank and is thus only a data processor. CTSU has no ability to process the data for any other purpose save for those purposes which UK Biobank (alone) determines.

The data supplied by NHS Digital are stored and processed at the CTSU under the sole direction of UK Biobank and pursuant to the contract between UK Biobank and CTSU (which confirms these data controller / data processor arrangements).

UK Biobank is a company limited by guarantee and registered at Companies House – it therefore has its own legal personality. Whist it receives funding from, amongst others the Wellcome Trust and the Medical Research Council, these organisations have no access to the data and cannot exercise any control in respect of its use. Once NHS Digital data has been received by UK Biobank (acting through its processor CTSU), it is uploaded to the UK Biobank resource, at which point registered researchers can apply to UK Biobank to use these linked data (alongside the other data in the resource) to undertake health related research which is in the public interest. Each application to use the resource is considered by UK Biobank pursuant to UK Biobank access policy and procedures, which were developed through extensive consultation, including public consultation and consultation with UK Biobank participants. Further detail is given within the Outputs section. UK Biobank has sole control of this process. De-identified data is made available to researchers following approval of a research application once they have signed a contract with UK Biobank (known as a Material Transfer Agreements (MTA)) which controls how the data can be used (including ensuring the data is only used for conducting health related research which is in the public interest).

Access to the resource:

UK Biobank’s access policy is set out in the Ethics and Governance Framework (https://www.ukbiobank.ac.uk/wp-content/uploads/2011/05/EGF20082.pdf). Its Access Procedures (http://www.ukbiobank.ac.uk/wp-content/uploads/2012/09/Access-Procedures-2011.pdf) were prepared during 2010 to 2011 and were the subject of extensive discussion and dialogue with its funders, the REC and the Ethics and Governance Committee. The Access Procedures were also put out to public consultation, including consultation with UK Biobank participants.

In summary:

- Every researcher has to register and apply to UK Biobank. If successful, they have to execute UK Biobank's Material Transfer Agreement (MTA). The application process is the same irrespective of the nature of the applicant and the applicant's location. The MTA governs the legal relationship between UK Biobank and each researcher and, inter alia, ensures that the researcher complies with all the applicable data protection requirements relating to all participant data that they receive. In addition to audit provisions (i.e. UK Biobank is entitled to audit any applicant if it transpires that a transgression has or may have taken place) the MTA contains explicit provisions stating that UK Biobank will bar a transgressing institution from any further use of the resource as well as publicly informing its governing bodies, funders et al;

- The criteria are that the applicant must be a bona fide researcher conducting health-related research in the public interest. Note that the researcher may be based internationally, but that the terms under which that international researcher may access the data is set out within the MTA and are the same as those for a UK researcher. The adjudication of each application is made by UK Biobank’s Access and Scientific teams, with independent ethical advice (from the University of Oxford’s multidisciplinary bioethics research centre, Ethox: http://www.ndph.ox.ac.uk/research/ethox-centre) as well as input from the UK Biobank Ethics Advisory Committee (http://www.ukbiobank.ac.uk/wp-content/uploads/2018/10/ukb-EACToR-march-2018-item-9a-2.pdf), and the whole process is overseen by UK Biobank’s Access Sub Committee (http://www.ukbiobank.ac.uk/access-to-the-resource/), which contains both scientific and legal / ethical expertise;

- Any data released to researchers is pseudonymised and encrypted. The researcher is only permitted to use the data for a particular (approved) research project for a particular (approved) time frame after which the researcher must securely destroy the data. The results of the research have to come back to UK Biobank – where they are then accessible by other researchers – and the researcher is obliged to publish their work;

- In other words, any data supplied by UK Biobank to a researcher can only be used for the research project in question and for a limited time through UK Biobank’s access procedures. Further, any such data cannot be sub-licensed or made available by researchers to any other party under any circumstances. This is all covered in UK Biobank’s MTA, which is non-negotiable. Access to the resource can be made by any bona fide researcher in the world. This agreement places no restriction on the geographical location of the researchers but any sharing must meet GDPR requirements.

UK Biobank has approved over 1,600 applications since it first opened to researches in 2012 and covers a very broad area of research interests. List of the approved applications can be found here:

http://www.ukbiobank.ac.uk/approved-research/.

There have been over 1000 publications from researchers using the UK Biobank data (http://www.ukbiobank.ac.uk/published-papers/), most of which have been in the last 2-3 years. As such, it is still a little premature to see tangible public health benefits as a direct result of these findings; Further, any public health benefits (e.g., changes in guidelines, clinical practice) are based on an accumulation of evidence from many sources (i.e. not just UK Biobank). Nonetheless, the NHS Digital data included in the resource has led to many findings that have obvious implications for public health. These include findings related to the benefit of even a small amount of physical activity on the risk of death and the benefit of increasing fibre intake and reducing meat intake on the risk of colorectal cancer. Much research has focused on how polygenic risk scores could help to predict an individual’s risk of developing a specific disease, thereby helping to target preventative action (e.g. lifestyle advice, screening programmes) to this subgroup of the population. Other examples include the finding that an active lifestyle can help offset the genetic risk of developing dementia and that statins may help reduce the risk of developing motor neurone disease.

UK Biobank is a major national and international health resource, and a registered charity in its own right, with the aim of improving the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye disorders, depression and forms of dementia. UK Biobank recruited 500,000 people aged between 40-69 years in 2006-2010 from across the country to take part in this project. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed via linkage to their health and health related records. Over many years this will continue to build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

UK Biobank requires data from NHS Digital to pursue its legitimate interests which are to set up and manage a major international research resource for health-related research that is in the public interest. UK Biobank is a prospective resource with the specific objective of following up its participants through the health records. This enables researchers to study the actual causes of disease and in turn develop treatments (for common and rare disease) which will be of considerable benefit to the public. Without the health record data the UK Biobank resource is simply not able to achieve its aims. This would, in UK Biobank’s view, constitute a (very)substantial public interest.

The UK Biobank resource is open to researchers who can apply to use the data within the resource (which data is de-identified) to undertake health related research which is in the public interest. The addition of the requested data from NHS Digital to the resource enhances the range of information which is available to researchers to enable them to undertake their approved research project. UK Biobank, as a charitable company limited by guarantee, was established to enable researchers to undertake health related research which is in the public interest and therefore the data is required from NHS Digital to enable UK Biobank to pursue its legitimate interests.

As a result of UK Biobank’s work, patients and the wider public benefit from the advances made in the prevention, diagnosis and treatment of disease. UK Biobank is now one of the largest and most used health research resources in the world. Over 13,000 researchers are registered with UK Biobank and over 1,600 health-related research applications have been approved.

Participants have signed consent forms permitting access to their data for this purpose. It is considered very unlikely that participants would object to the processing activities involved in obtaining linked data from NHS Digital. During the past 10 years since participants joined UK Biobank during 2006-10, fewer than 800 of the 500,000 participants have withdrawn from the study and asked that their information be deleted. For clarity, UK Biobank relies on GDPR 6 (1) (a) - explicit consent for the collection and storage of bio samples and the permission to make available for health related research. The data from NHS Digital is secondary health care data and therefore UK Biobank relies on Legitimate Interests (GDPR Article 6(1)(f) ) in regards to the data they obtain from NHS Digital.

UK Biobank processes the special category personal data provided by NHS Digital in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. UK Biobank falls clearly within this category as its purpose is to improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye disorders, depression and forms of dementia.

Lifestyle and environmental information, medical history, physical measurements, and biological samples are to be collected from about 500,000 people aged 40 to 69 at presentation and then, with consent, their health will be followed for many years through medical and other health related records. The biological samples will be stored so that they can be used for a wide range of biochemical and genetic analyses in the future.

The data are stored in a way that makes it is extremely difficult even for UK Biobank and their data processors (University of Oxford) to re-identify participants. Only a very few individuals within the University of Oxford are allowed to do so (and they are strictly monitored) in order that further information about participants can be added. Data provided to researchers have personal identifiers removed so that an individual participant cannot be identified. There are no circumstances in which data can be processed in a manner that could have an adverse impact on the data subject.

The overall purpose of UK Biobank is to create a prospective epidemiological resource of 500,000 people aged 45 - 69 at the time of recruitment from around the UK. The methods to be used for the identification, invitation and assessment of participants have been developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All participants in the UK Biobank study have given fully informed consent for access to and long-term storage of information from their medical and other health-related records, and use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purpose; and recontact by UK Biobank to invite them (but without any obligation) to answer further questions or take part in further assessments.

UK Biobank is a unique resource, because of the combination of its very large size and the range and detail of data that it contains on its participants, including extensive phenotype and genotype information on each of them. Further, rather than recruiting participants with specific diseases or risk factors, UK Biobank is a prospective population-based resource established to support research into the causes of a wide range of diseases affecting people in middle and older age. Thus, the ability to link to participant’s health records, so that specific outcomes occurring during long-term follow-up of the participants can be identified and appropriate disease-based research carried out, is critical.

UK Biobank’s priority is to combine extensive and precise measurement of exposures, along with the detailed and rigorous follow-up for a wide range of health related outcomes, and to promote innovative science by maximising secure access to deidentified data.

UK Biobank was established in 2003. It is funded by the Wellcome Trust, the Department of Health and the Medical Research Council. No funder is permitted to influence the outcomes nor suppress any outcomes of research.

The recruitment of its 500,000 participants took place between 2007 and 2010, whose health is now being followed long term primarily through linking to NHS health data. Access to the resource opened in 2012, since which time there have been over 1,600 applications submitted and over 13,000 researchers have registered to use the resource, resulting in 1,000 research papers published using the UK Biobank resource.

At baseline (i.e. at time of recruitment), UK Biobank obtained and holds identifiable information on its participants. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed by linking to their health and health related records. Over many years this will build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

Under this Agreement, UK Biobank receives updates on participants’ health and health-related data. This includes:

i. Hospital Episode Statistics – details of participants attendances at hospital for any reason. 1997/98 to latest available annual year (updated to latest available on a monthly basis to assist research involving COVID-19);

ii. Mental Health data – contains record-level data about the care of participants who are in contact with mental health, learning disabilities or autism spectrum disorder services. Mental Health Minimum Data Set 2006/07 to 2014/15, Mental Health and Learning Disabilities Data Set 2014/15 to November 2015/16, Mental Health Services Data Set 2016/17 to latest available annual year;

iii. Diagnostic Imaging data – collects information about the imaging tests and scans carried out on participants such as x-rays, MRI scans or ultrasound. 2012/13 to latest available annual year;

iv. Civil Registration Mortality data – details of participants’ deaths including date and cause. 2011/12 to latest available annual update (updated to latest available on a monthly basis to assist research involving COVID-19);

v. Cancer Registration data – details of any cancers participants’ are registered as having. 2011/12 to latest available with a bi-annual update;

vi. Emergency Care Data Set data - details around participant attendance in A&E. October 2017 to latest available annual update. (updated to latest available on a monthly basis to assist research involving COVID-19);

vii National Diabetes Audit data - data collected on diabetes in participants. April 2003 to latest available annual year;

viii. Improving Access to Psychological Therapies (IAPT)- measure of anxiety and depression in participants. April 2012 to latest available annual year;

For clarity, data now required on a monthly basis is not restricted to COVID-19 related research.

As a large cohort it is not feasible for UK Biobank itself to follow up each of its participants on an individual basis regularly and nor would this generate the necessary clinical detail. Self-reported data does not possess (anything like) the needed coverage, reliability and consistency, that the health records contain. As such, the data provided via NHS Digital is crucial to the viability of UK Biobank as the primary mechanism for follow up of the cohort, particularly as it provides coverage for all 500,000 participants in an entirely unobtrusive (i.e. passive) manner. The linked datasets all form part of the participants’ health record and are thus all vital to providing the necessary health record information. Of key importance are datasets related to death and cancer registry notifications and hospital activity. Without access to these data, UK Biobank would not be a prospective resource and research on the associations of lifestyle and genetic factors on the risk of developing a wide range of diseases over time would simply not be possible. This is because these datasets provide vital information on which participants have developed which diseases and at what time. Given UK Biobank is a prospective study available to registered researchers worldwide, research interest covers a very wide range of health outcomes, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke, mental health, and many others. As such, access to data on deaths, cancers and hospital admissions and related activities is of vital importance.

In addition to the above-mentioned datasets that UK Biobank hold, other health-related datasets that have now become available are also of high research interest. For example, the IAPT and mental health datasets will help to identify participants with mental health disorders and enable research into their treatment response and support required. The diagnostic imaging dataset will enable research into the treatment pathways for a wide range of conditions, while the emergency department data will enable research into the reasons for and consequences of acute admissions. Finally, the national diabetes audit will help to identify participants diagnosed with diabetes and their management pathways (as these participants are often under-represented in the HES data). UK Biobank require access to all retrospective data to enable epidemiological research into antecedent clinical events prior to diagnosis and is crucial for researchers assessing the effect of multi-morbidity on future health outcomes. Without these linked data provided by NHS Digital, researchers would not be able to fully ascertain the full burden of disease in the cohort or effects of particular treatment and management pathways on health. All of the datasets requested enhance the UK Biobank resource and allow researchers to undertake health-related research that is in the public interest.

Other datasets include:

ix. Personal Demographic Service (PDS) updates – details of changes to participants’ demographics, postal addresses or registered GP practices.

It is very important for UK Biobank to be able to maintain accurate information about participants’ contact details (including postal address) and registered GP practice (which were up to date for the whole cohort at the time of recruitment but there will be undoubtedly have been changes since then). There are several reasons for this requirement:

- Ongoing approved linkages to primary care data require accurate information about registered GP practice (at practice as well as CCG level) to ensure that data about a particular participant are requested from the correct GP practice.

- Direct contact with GPs is required as part of the UK Biobank imaging sub study (https://imaging.ukbiobank.ac.uk/) in the event that a participant is identified with a potentially serious incidental finding (as noted by the radiographer and then confirmed by a specialist doctor (radiologist)). Should such an abnormality be found, UK Biobank will write to the participant and their GP to allow them to further investigate these potentially serious findings.

- To enable GPs to be confident about extraction of primary care data from GP practice system suppliers to UK Biobank, UK Biobank will need to be able to provide GP practices with up-to-date listings of UK Biobank participants registered at their practices, and – where requested – evidence of their consent.

- Follow-up in a cohort study needs to be as complete as possible. This is to avoid both unnecessary reduction in size of the cohort (which reduces statistical power of data analyses), and loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on intermittent re-contact with UK Biobank’s participants to invite them to complete brief questionnaires about health related exposures or outcomes. It is important that UK Biobank have correct and up-to-date participant contact details for this. Although UK Biobank encourage participants to inform UK Biobank when they move, this is not a failsafe method of ensuring the most accurate information.

- UK Biobank are committed to ensuring participants continue to be kept aware of progress and developments in UK Biobank, to obtaining their opinions on proposed enhancements to the study, and to providing them with information on how to contact UK Biobank should they have any questions. UK Biobank do this in part through UK Biobank’s annual newsletter, sent via e-mail and/or postal mail. Clearly, UK Biobank need up-to-date contact details (especially postal address) to do this. UK Biobank would like to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress.

- Important potential research questions about the effects of environmental exposures on health require information about current and previous postal address since this allows these exposures to be estimated (from postcodes). These include, for example, valuable studies of the health effects of air and noise pollution, living close to power cables and aspects of the local environment, such as nearby local public amenities and green space.

For each of these datasets listed above, UK Biobank require linkage only for the 500,000 UK Biobank participants who joined the study and who were recruited across the UK. It should be noted that the UK Biobank are only requesting individual-level identifying data to ensure correct linkage. UK Biobank do not make available to researchers any identifiable data (such as postcode, date of birth, name, NHS number, etc.) and instead, each research project is provided with a unique encrypted 9-digit identification number. Hence, it is not possible for a researcher to identify a UK Biobank participant using the data that UK Biobank release.

UK Biobank is the data controller of the data supplied by NHS Digital. Pursuant to a contract between UK Biobank and The Clinical Trials Service Unit (CTSU) at the University of Oxford, CTSU act as a data processor of UK Biobank’s data. CTSU is only able to act on the instructions of UK Biobank and is thus only a data processor. CTSU has no ability to process the data for any other purpose save for those purposes which UK Biobank (alone) determines.

The data supplied by NHS Digital are stored and processed at the CTSU under the sole direction of UK Biobank and pursuant to the contract between UK Biobank and CTSU (which confirms these data controller / data processor arrangements).

UK Biobank is a company limited by guarantee and registered at Companies House – it therefore has its own legal personality. Whist it receives funding from, amongst others the Wellcome Trust and the Medical Research Council, these organisations have no access to the data and cannot exercise any control in respect of its use. Once NHS Digital data has been received by UK Biobank (acting through its processor CTSU), it is uploaded to the UK Biobank resource, at which point registered researchers can apply to UK Biobank to use these linked data (alongside the other data in the resource) to undertake health related research which is in the public interest. Each application to use the resource is considered by UK Biobank pursuant to UK Biobank access policy and procedures, which were developed through extensive consultation, including public consultation and consultation with UK Biobank participants. Further detail is given within the Outputs section. UK Biobank has sole control of this process. De-identified data is made available to researchers following approval of a research application once they have signed a contract with UK Biobank (known as a Material Transfer Agreements (MTA)) which controls how the data can be used (including ensuring the data is only used for conducting health related research which is in the public interest).

Access to the resource:

UK Biobank’s access policy is set out in the Ethics and Governance Framework (https://www.ukbiobank.ac.uk/wp-content/uploads/2011/05/EGF20082.pdf). Its Access Procedures (http://www.ukbiobank.ac.uk/wp-content/uploads/2012/09/Access-Procedures-2011.pdf) were prepared during 2010 to 2011 and were the subject of extensive discussion and dialogue with its funders, the REC and the Ethics and Governance Committee. The Access Procedures were also put out to public consultation, including consultation with UK Biobank participants.

In summary:

- Every researcher has to register and apply to UK Biobank. If successful, they have to execute UK Biobank's Material Transfer Agreement (MTA). The application process is the same irrespective of the nature of the applicant and the applicant's location. The MTA governs the legal relationship between UK Biobank and each researcher and, inter alia, ensures that the researcher complies with all the applicable data protection requirements relating to all participant data that they receive. In addition to audit provisions (i.e. UK Biobank is entitled to audit any applicant if it transpires that a transgression has or may have taken place) the MTA contains explicit provisions stating that UK Biobank will bar a transgressing institution from any further use of the resource as well as publicly informing its governing bodies, funders et al;

- The criteria are that the applicant must be a bona fide researcher conducting health-related research in the public interest. Note that the researcher may be based internationally, but that the terms under which that international researcher may access the data is set out within the MTA and are the same as those for a UK researcher. The adjudication of each application is made by UK Biobank’s Access and Scientific teams, with independent ethical advice (from the University of Oxford’s multidisciplinary bioethics research centre, Ethox: http://www.ndph.ox.ac.uk/research/ethox-centre) as well as input from the UK Biobank Ethics Advisory Committee (http://www.ukbiobank.ac.uk/wp-content/uploads/2018/10/ukb-EACToR-march-2018-item-9a-2.pdf), and the whole process is overseen by UK Biobank’s Access Sub Committee (http://www.ukbiobank.ac.uk/access-to-the-resource/), which contains both scientific and legal / ethical expertise;

- Any data released to researchers is pseudonymised and encrypted. The researcher is only permitted to use the data for a particular (approved) research project for a particular (approved) time frame after which the researcher must securely destroy the data. The results of the research have to come back to UK Biobank – where they are then accessible by other researchers – and the researcher is obliged to publish their work;

- In other words, any data supplied by UK Biobank to a researcher can only be used for the research project in question and for a limited time through UK Biobank’s access procedures. Further, any such data cannot be sub-licensed or made available by researchers to any other party under any circumstances. This is all covered in UK Biobank’s MTA, which is non-negotiable. Access to the resource can be made by any bona fide researcher in the world. There is no restriction on the location of the researcher.

UK Biobank has approved over 1,600 applications since it first opened to researches in 2012. List of the approved applications can be found here:

http://www.ukbiobank.ac.uk/approved-research/

In addition there have been over 1000 publications from researchers using the UK Biobank resources on a wide variety of topics including genetics, diet, exercise, mental health, respiratory disease and many more topics. The full list of publications based on the resource here: http://www.ukbiobank.ac.uk/published-papers/

UK Biobank is a major national and international health resource, and a registered charity in its own right, with the aim of improving the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye disorders, depression and forms of dementia. UK Biobank recruited 500,000 people aged between 40-69 years in 2006-2010 from across the country to take part in this project. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed via linkage to their health and health related records. Over many years this will continue to build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

UK Biobank requires data from NHS Digital to pursue its legitimate interests which are to set up and manage a major international research resource for health-related research that is in the public interest. UK Biobank is a prospective resource with the specific objective of following up its participants through the health records. This enables researchers to study the actual causes of disease and in turn develop treatments (for common and rare disease) which will be of considerable benefit to the public. Without the health record data the UK Biobank resource is simply not able to achieve its aims. This would, in UK Biobank’s view, constitute a (very)substantial public interest.

The UK Biobank resource is open to researchers who can apply to use the data within the resource (which data is de-identified) to undertake health related research which is in the public interest. The addition of the requested data from NHS Digital to the resource enhances the range of information which is available to researchers to enable them to undertake their approved research project. UK Biobank, as a charitable company limited by guarantee, was established to enable researchers to undertake health related research which is in the public interest and therefore the data is required from NHS Digital to enable UK Biobank to pursue its legitimate interests.

As a result of UK Biobank’s work, patients and the wider public benefit from the advances made in the prevention, diagnosis and treatment of disease. UK Biobank is now one of the largest and most used health research resources in the world. Over 13,000 researchers are registered with UK Biobank and over 1,600 health-related research applications have been approved.

Participants have signed consent forms permitting access to their data for this purpose. It is considered very unlikely that participants would object to the processing activities involved in obtaining linked data from NHS Digital. During the past 10 years since participants joined UK Biobank during 2006-10, fewer than 800 of the 500,000 participants have withdrawn from the study and asked that their information be deleted. For clarity, UK Biobank relies on GDPR 6 (1) (a) - explicit consent for the collection and storage of bio samples and the permission to make available for health related research. The data from NHS Digital is secondary health care data and therefore UK Biobank relies on Legitimate Interests (GDPR Article 6(1)(f) ) in regards to the data they obtain from NHS Digital.

UK Biobank processes the special category personal data provided by NHS Digital in accordance with the provisions of Article 9(2) (j) of the GDPR, namely that the processing is necessary for reasons of public interest in the area of scientific research. UK Biobank falls clearly within this category as its purpose is to improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye disorders, depression and forms of dementia.

Lifestyle and environmental information, medical history, physical measurements, and biological samples are to be collected from about 500,000 people aged 40 to 69 at presentation and then, with consent, their health will be followed for many years through medical and other health related records. The biological samples will be stored so that they can be used for a wide range of biochemical and genetic analyses in the future.

The data are stored in a way that makes it is extremely difficult even for UK Biobank and their data processors (University of Oxford) to re-identify participants. Only a very few individuals within the University of Oxford are allowed to do so (and they are strictly monitored) in order that further information about participants can be added. Data provided to researchers have personal identifiers removed so that an individual participant cannot be identified. There are no circumstances in which data can be processed in a manner that could have an adverse impact on the data subject.

The overall purpose of UK Biobank is to create a prospective epidemiological resource of 500,000 people aged 45 - 69 at the time of recruitment from around the UK. The methods to be used for the identification, invitation and assessment of participants have been developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All participants in the UK Biobank study have given fully informed consent for access to and long-term storage of information from their medical and other health-related records, and use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purpose; and recontact by UK Biobank to invite them (but without any obligation) to answer further questions or take part in further assessments.

UK Biobank is a unique resource, because of the combination of its very large size and the range and detail of data that it contains on its participants, including extensive phenotype and genotype information on each of them. Further, rather than recruiting participants with specific diseases or risk factors, UK Biobank is a prospective population-based resource established to support research into the causes of a wide range of diseases affecting people in middle and older age. Thus, the ability to link to participant’s health records, so that specific outcomes occurring during long-term follow-up of the participants can be identified and appropriate disease-based research carried out, is critical.

UK Biobank’s priority is to combine extensive and precise measurement of exposures, along with the detailed and rigorous follow-up for a wide range of health related outcomes, and to promote innovative science by maximising secure access to deidentified data.

UK Biobank was established in 2003. It is funded by the Wellcome Trust, the Department of Health and the Medical Research Council. No funder is permitted to influence the outcomes nor suppress any outcomes of research.

The recruitment of its 500,000 participants took place between 2007 and 2010, whose health is now being followed long term primarily through linking to NHS health data. Access to the resource opened in 2012, since which time there have been over 1,600 applications submitted and over 13,000 researchers have registered to use the resource, resulting in 1,000 research papers published using the UK Biobank resource.

At baseline (i.e. at time of recruitment), UK Biobank obtained and holds identifiable information on its participants. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed by linking to their health and health related records. Over many years this will build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

Under this Agreement, UK Biobank receives updates on participants’ health and health-related data. This includes:

i. Hospital Episode Statistics – details of participants attendances at hospital for any reason. 1997/98 to latest available annual year (updated to latest available on a monthly basis to assist research involving COVID-19);

ii. Mental Health data – contains record-level data about the care of participants who are in contact with mental health, learning disabilities or autism spectrum disorder services. Mental Health Minimum Data Set 2006/07 to 2014/15, Mental Health and Learning Disabilities Data Set 2014/15 to November 2015/16, Mental Health Services Data Set 2016/17 to latest available annual year;

iii. Diagnostic Imaging data – collects information about the imaging tests and scans carried out on participants such as x-rays, MRI scans or ultrasound. 2012/13 to latest available annual year;

iv. Civil Registration Mortality data – details of participants’ deaths including date and cause. 2011/12 to latest available annual update (updated to latest available on a monthly basis to assist research involving COVID-19);

v. Cancer Registration data – details of any cancers participants’ are registered as having. 2011/12 to latest available with a bi-annual update;

vi. Emergency Care Data Set data - details around participant attendance in A&E. October 2017 to latest available annual update. (updated to latest available on a monthly basis to assist research involving COVID-19);

vii National Diabetes Audit data - data collected on diabetes in participants. April 2003 to latest available annual year;

viii. Improving Access to Psychological Therapies (IAPT)- measure of anxiety and depression in participants. April 2012 to latest available annual year;

For clarity, data now required on a monthly basis is not restricted to COVID-19 related research.

As a large cohort it is not feasible for UK Biobank itself to follow up each of its participants on an individual basis regularly and nor would this generate the necessary clinical detail. Self-reported data does not possess (anything like) the needed coverage, reliability and consistency, that the health records contain. As such, the data provided via NHS Digital is crucial to the viability of UK Biobank as the primary mechanism for follow up of the cohort, particularly as it provides coverage for all 500,000 participants in an entirely unobtrusive (i.e. passive) manner. The linked datasets all form part of the participants’ health record and are thus all vital to providing the necessary health record information. Of key importance are datasets related to death and cancer registry notifications and hospital activity. Without access to these data, UK Biobank would not be a prospective resource and research on the associations of lifestyle and genetic factors on the risk of developing a wide range of diseases over time would simply not be possible. This is because these datasets provide vital information on which participants have developed which diseases and at what time. Given UK Biobank is a prospective study available to registered researchers worldwide, research interest covers a very wide range of health outcomes, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke, mental health, and many others. As such, access to data on deaths, cancers and hospital admissions and related activities is of vital importance.

In addition to the above-mentioned datasets that UK Biobank hold, other health-related datasets that have now become available are also of high research interest. For example, the IAPT and mental health datasets will help to identify participants with mental health disorders and enable research into their treatment response and support required. The diagnostic imaging dataset will enable research into the treatment pathways for a wide range of conditions, while the emergency department data will enable research into the reasons for and consequences of acute admissions. Finally, the national diabetes audit will help to identify participants diagnosed with diabetes and their management pathways (as these participants are often under-represented in the HES data). UK Biobank require access to all retrospective data to enable epidemiological research into antecedent clinical events prior to diagnosis and is crucial for researchers assessing the effect of multi-morbidity on future health outcomes. Without these linked data provided by NHS Digital, researchers would not be able to fully ascertain the full burden of disease in the cohort or effects of particular treatment and management pathways on health. All of the datasets requested enhance the UK Biobank resource and allow researchers to undertake health-related research that is in the public interest.

Other datasets include:

ix. Personal Demographic Service (PDS) updates – details of changes to participants’ demographics, postal addresses or registered GP practices.

It is very important for UK Biobank to be able to maintain accurate information about participants’ contact details (including postal address) and registered GP practice (which were up to date for the whole cohort at the time of recruitment but there will be undoubtedly have been changes since then). There are several reasons for this requirement:

- Ongoing approved linkages to primary care data require accurate information about registered GP practice (at practice as well as CCG level) to ensure that data about a particular participant are requested from the correct GP practice.

- Direct contact with GPs is required as part of the UK Biobank imaging sub study (https://imaging.ukbiobank.ac.uk/) in the event that a participant is identified with a potentially serious incidental finding (as noted by the radiographer and then confirmed by a specialist doctor (radiologist)). Should such an abnormality be found, UK Biobank will write to the participant and their GP to allow them to further investigate these potentially serious findings.

- To enable GPs to be confident about extraction of primary care data from GP practice system suppliers to UK Biobank, UK Biobank will need to be able to provide GP practices with up-to-date listings of UK Biobank participants registered at their practices, and – where requested – evidence of their consent.

- Follow-up in a cohort study needs to be as complete as possible. This is to avoid both unnecessary reduction in size of the cohort (which reduces statistical power of data analyses), and loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on intermittent re-contact with UK Biobank’s participants to invite them to complete brief questionnaires about health related exposures or outcomes. It is important that UK Biobank have correct and up-to-date participant contact details for this. Although UK Biobank encourage participants to inform UK Biobank when they move, this is not a failsafe method of ensuring the most accurate information.

- UK Biobank are committed to ensuring participants continue to be kept aware of progress and developments in UK Biobank, to obtaining their opinions on proposed enhancements to the study, and to providing them with information on how to contact UK Biobank should they have any questions. UK Biobank do this in part through UK Biobank’s annual newsletter, sent via e-mail and/or postal mail. Clearly, UK Biobank need up-to-date contact details (especially postal address) to do this. UK Biobank would like to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress.

- Important potential research questions about the effects of environmental exposures on health require information about current and previous postal address since this allows these exposures to be estimated (from postcodes). These include, for example, valuable studies of the health effects of air and noise pollution, living close to power cables and aspects of the local environment, such as nearby local public amenities and green space.

For each of these datasets listed above, UK Biobank require linkage only for the 500,000 UK Biobank participants who joined the study and who were recruited across the UK. It should be noted that the UK Biobank are only requesting individual-level identifying data to ensure correct linkage. UK Biobank do not make available to researchers any identifiable data (such as postcode, date of birth, name, NHS number, etc.) and instead, each research project is provided with a unique encrypted 9-digit identification number. Hence, it is not possible for a researcher to identify a UK Biobank participant using the data that UK Biobank release.

UK Biobank is the data controller of the data supplied by NHS Digital. Pursuant to a contract between UK Biobank and The Clinical Trials Service Unit (CTSU) at the University of Oxford, CTSU act as a data processor of UK Biobank’s data. CTSU is only able to act on the instructions of UK Biobank and is thus only a data processor. CTSU has no ability to process the data for any other purpose save for those purposes which UK Biobank (alone) determines.

The data supplied by NHS Digital are stored and processed at the CTSU under the sole direction of UK Biobank and pursuant to the contract between UK Biobank and CTSU (which confirms these data controller / data processor arrangements).

UK Biobank is a company limited by guarantee and registered at Companies House – it therefore has its own legal personality. Whist it receives funding from, amongst others the Wellcome Trust and the Medical Research Council, these organisations have no access to the data and cannot exercise any control in respect of its use. Once NHS Digital data has been received by UK Biobank (acting through its processor CTSU), it is uploaded to the UK Biobank resource, at which point registered researchers can apply to UK Biobank to use these linked data (alongside the other data in the resource) to undertake health related research which is in the public interest. Each application to use the resource is considered by UK Biobank pursuant to UK Biobank access policy and procedures, which were developed through extensive consultation, including public consultation and consultation with UK Biobank participants. Further detail is given within the Outputs section. UK Biobank has sole control of this process. De-identified data is made available to researchers following approval of a research application once they have signed a contract with UK Biobank (known as a Material Transfer Agreements (MTA)) which controls how the data can be used (including ensuring the data is only used for conducting health related research which is in the public interest).

Access to the resource:

UK Biobank’s access policy is set out in the Ethics and Governance Framework (https://www.ukbiobank.ac.uk/wp-content/uploads/2011/05/EGF20082.pdf). Its Access Procedures (http://www.ukbiobank.ac.uk/wp-content/uploads/2012/09/Access-Procedures-2011.pdf) were prepared during 2010 to 2011 and were the subject of extensive discussion and dialogue with its funders, the REC and the Ethics and Governance Committee. The Access Procedures were also put out to public consultation, including consultation with UK Biobank participants.

In summary:

- Every researcher has to register and apply to UK Biobank. If successful, they have to execute UK Biobank's Material Transfer Agreement (MTA). The application process is the same irrespective of the nature of the applicant and the applicant's location. The MTA governs the legal relationship between UK Biobank and each researcher and, inter alia, ensures that the researcher complies with all the applicable data protection requirements relating to all participant data that they receive. In addition to audit provisions (i.e. UK Biobank is entitled to audit any applicant if it transpires that a transgression has or may have taken place) the MTA contains explicit provisions stating that UK Biobank will bar a transgressing institution from any further use of the resource as well as publicly informing its governing bodies, funders et al;

- The criteria are that the applicant must be a bona fide researcher conducting health-related research in the public interest. Note that the researcher may be based internationally, but that the terms under which that international researcher may access the data is set out within the MTA and are the same as those for a UK researcher. The adjudication of each application is made by UK Biobank’s Access and Scientific teams, with independent ethical advice (from the University of Oxford’s multidisciplinary bioethics research centre, Ethox: http://www.ndph.ox.ac.uk/research/ethox-centre) as well as input from the UK Biobank Ethics Advisory Committee (http://www.ukbiobank.ac.uk/wp-content/uploads/2018/10/ukb-EACToR-march-2018-item-9a-2.pdf), and the whole process is overseen by UK Biobank’s Access Sub Committee (http://www.ukbiobank.ac.uk/access-to-the-resource/), which contains both scientific and legal / ethical expertise;

- Any data released to researchers is pseudonymised and encrypted. The researcher is only permitted to use the data for a particular (approved) research project for a particular (approved) time frame after which the researcher must securely destroy the data. The results of the research have to come back to UK Biobank – where they are then accessible by other researchers – and the researcher is obliged to publish their work;

- In other words, any data supplied by UK Biobank to a researcher can only be used for the research project in question and for a limited time through UK Biobank’s access procedures. Further, any such data cannot be sub-licensed or made available by researchers to any other party under any circumstances. This is all covered in UK Biobank’s MTA, which is non-negotiable. Access to the resource can be made by any bona fide researcher in the world. There is no restriction on the location of the researcher.

UK Biobank has approved over 1,600 applications since it first opened to researches in 2012. List of the approved applications can be found here:

http://www.ukbiobank.ac.uk/approved-research/

In addition there have been over 1000 publications from researchers using the UK Biobank resources on a wide variety of topics including genetics, diet, exercise, mental health, respiratory disease and many more topics. The full list of publications based on the resource here: http://www.ukbiobank.ac.uk/published-papers/

UK Biobank is a major national and international health resource, and a registered charity in its own right, with the aim of improving the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye disorders, depression and forms of dementia. UK Biobank recruited 500,000 people aged between 40-69 years in 2006-2010 from across the country to take part in this project. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed via linkage to their health and health related records. Over many years this will continue to build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

UK Biobank requires data from NHS Digital to pursue its legitimate interests which are to set up and manage a major international research resource for health-related research that is in the public interest. UK Biobank is a prospective resource with the specific objective of following up its participants through the health records. This enables researchers to study the actual causes of disease and in turn develop treatments (for common and rare disease) which will be of considerable benefit to the public. Without the health record data the UK Biobank resource is simply not able to achieve its aims. This would, in UK Biobank’s view, constitute a (very)substantial public interest.

The UK Biobank resource is open to researchers who can apply to use the data within the resource (which data is de-identified) to undertake health related research which is in the public interest. The addition of the requested data from NHS Digital to the resource enhances the range of information which is available to researchers to enable them to undertake their approved research project. UK Biobank, as a charitable company limited by guarantee, was established to enable researchers to undertake health related research which is in the public interest and therefore the data is required from NHS Digital to enable UK Biobank to pursue its legitimate interests.

As a result of UK Biobank’s work, patients and the wider public benefit from the advances made in the prevention, diagnosis and treatment of disease. UK Biobank is now one of the largest and most used health research resources in the world. Over 13,000 researchers are registered with UK Biobank and over 1,600 health-related research applications have been approved.

Participants have signed consent forms permitting access to their data for this purpose. It is considered very unlikely that participants would object to the processing activities involved in obtaining linked data from NHS Digital. During the past 10 years since participants joined UK Biobank during 2006-10, fewer than 800 of the 500,000 participants have withdrawn from the study and asked that their information be deleted. For clarity, UK Biobank relies on GDPR 6 (1) (a) - explicit consent for the collection and storage of bio samples and the permission to make available for health related research. The data from NHS Digital is secondary health care data and therefore UK Biobank relies on Legitimate Interests (GDPR Article 6(1)(f) ) in regards to the data they obtain from NHS Digital.

Lifestyle and environmental information, medical history, physical measurements, and biological samples are to be collected from about 500,000 people aged 40 to 69 at presentation and then, with consent, their health will be followed for many years through medical and other health related records. The biological samples will be stored so that they can be used for a wide range of biochemical and genetic analyses in the future.

The data are stored in a way that makes it is extremely difficult even for UK Biobank and their data processors (University of Oxford) to re-identify participants. Only a very few individuals within the University of Oxford are allowed to do so (and they are strictly monitored) in order that further information about participants can be added. Data provided to researchers have personal identifiers removed so that an individual participant cannot be identified. There are no circumstances in which data can be processed in a manner that could have an adverse impact on the data subject.

The overall purpose of UK Biobank is to create a prospective epidemiological resource of 500,000 people aged 45 - 69 at the time of recruitment from around the UK. The methods to be used for the identification, invitation and assessment of participants have been developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All participants in the UK Biobank study have given fully informed consent for access to and long-term storage of information from their medical and other health-related records, and use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purpose; and recontact by UK Biobank to invite them (but without any obligation) to answer further questions or take part in further assessments.

UK Biobank is a unique resource, because of the combination of its very large size and the range and detail of data that it contains on its participants, including extensive phenotype and genotype information on each of them. Further, rather than recruiting participants with specific diseases or risk factors, UK Biobank is a prospective population-based resource established to support research into the causes of a wide range of diseases affecting people in middle and older age. Thus, the ability to link to participant’s health records, so that specific outcomes occurring during long-term follow-up of the participants can be identified and appropriate disease-based research carried out, is critical.

UK Biobank’s priority is to combine extensive and precise measurement of exposures, along with the detailed and rigorous follow-up for a wide range of health related outcomes, and to promote innovative science by maximising secure access to deidentified data.

UK Biobank was established in 2003. It is funded by the Wellcome Trust, the Department of Health and the Medical Research Council. No funder is permitted to influence the outcomes nor suppress any outcomes of research.

The recruitment of its 500,000 participants took place between 2007 and 2010, whose health is now being followed long term primarily through linking to NHS health data. Access to the resource opened in 2012, since which time there have been over 1,600 applications submitted and over 13,000 researchers have registered to use the resource, resulting in 1,000 research papers published using the UK Biobank resource.

At baseline (i.e. at time of recruitment), UK Biobank obtained and holds identifiable information on its participants. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed by linking to their health and health related records. Over many years this will build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

Under this Agreement, UK Biobank receives updates on participants’ health and health-related data. This includes:

i. Hospital Episode Statistics – details of participants attendances at hospital for any reason. 1997/98 to latest available annual year;

ii. Mental Health data – contains record-level data about the care of

participants who are in contact with mental health, learning disabilities or autism spectrum disorder services. Mental Health Minimum Data Set 2006/07 to 2014/15, Mental Health and Learning Disabilities Data Set 2014/15 to November 2015/16, Mental Health Services Data Set 2016/17 to latest available annual year;

iii. Diagnostic Imaging data – collects information about the imaging tests and scans carried out on participants such as x-rays, MRI scans or ultrasound. 2012/13 to latest available annual year;

iv. Civil Registration Mortality data – details of participants’ deaths including date and cause. 2011/12 to latest available with a bi-annual update;

v. Cancer Registration data – details of any cancers participants’ are registered as having. 2011/12 to latest available with a bi-annual update;

vi. Emergency Care Data Set data - details around participant attendance in A&E. October 2017 to latest available annual year;

vii National Diabetes Audit data - data collected on diabetes in participants. April 2003 to latest available annual year;

viii. Improving Access to Psychological Therapies (IAPT)- measure of anxiety and depression in participants. April 2012 to latest available annual year;

As a large cohort it is not feasible for UK Biobank itself to follow up each of its participants on an individual basis regularly and nor would this generate the necessary clinical detail. Self-reported data does not possess (anything like) the needed coverage, reliability and consistency, that the health records contain. As such, the data provided via NHS Digital is crucial to the viability of UK Biobank as the primary mechanism for follow up of the cohort, particularly as it provides coverage for all 500,000 participants in an entirely unobtrusive (i.e. passive) manner. The linked datasets all form part of the participants’ health record and are thus all vital to providing the necessary health record information. Of key importance are datasets related to death and cancer registry notifications and hospital activity. Without access to these data, UK Biobank would not be a prospective resource and research on the associations of lifestyle and genetic factors on the risk of developing a wide range of diseases over time would simply not be possible. This is because these datasets provide vital information on which participants have developed which diseases and at what time. Given UK Biobank is a prospective study available to registered researchers worldwide, research interest covers a very wide range of health outcomes, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke, mental health, and many others. As such, access to data on deaths, cancers and hospital admissions and related activities is of vital importance.

In addition to the above-mentioned datasets that UK Biobank hold, other health-related datasets that have now become available are also of high research interest. For example, the IAPT and mental health datasets will help to identify participants with mental health disorders and enable research into their treatment response and support required. The diagnostic imaging dataset will enable research into the treatment pathways for a wide range of conditions, while the emergency department data will enable research into the reasons for and consequences of acute admissions. Finally, the national diabetes audit will help to identify participants diagnosed with diabetes and their management pathways (as these participants are often under-represented in the HES data). UK Biobank require access to all retrospective data to enable epidemiological research into antecedent clinical events prior to diagnosis and is crucial for researchers assessing the effect of multi-morbidity on future health outcomes. Without these linked data provided by NHS Digital, researchers would not be able to fully ascertain the full burden of disease in the cohort or effects of particular treatment and management pathways on health. All of the datasets requested enhance the UK Biobank resource and allow researchers to undertake health-related research that is in the public interest.

Other datasets include:

ix. Personal Demographic Service (PDS) updates – details of changes to participants’ demographics, postal addresses or registered GP practices.

It is very important for UK Biobank to be able to maintain accurate information about participants’ contact details (including postal address) and registered GP practice (which were up to date for the whole cohort at the time of recruitment but there will be undoubtedly have been changes since then). There are several reasons for this requirement:

- Ongoing approved linkages to primary care data require accurate information about registered GP practice (at practice as well as CCG level) to ensure that data about a particular participant are requested from the correct GP practice.

- Direct contact with GPs is required as part of the UK Biobank imaging sub study (https://imaging.ukbiobank.ac.uk/) in the event that a participant is identified with a potentially serious incidental finding (as noted by the radiographer and then confirmed by a specialist doctor (radiologist)). Should such an abnormality be found, UK Biobank will write to the participant and their GP to allow them to further investigate these potentially serious findings.

- To enable GPs to be confident about extraction of primary care data from GP practice system suppliers to UK Biobank, UK Biobank will need to be able to provide GP practices with up-to-date listings of UK Biobank participants registered at their practices, and – where requested – evidence of their consent.

- Follow-up in a cohort study needs to be as complete as possible. This is to avoid both unnecessary reduction in size of the cohort (which reduces statistical power of data analyses), and loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on intermittent re-contact with UK Biobank’s participants to invite them to complete brief questionnaires about health related exposures or outcomes. It is important that UK Biobank have correct and up-to-date participant contact details for this. Although UK Biobank encourage participants to inform UK Biobank when they move, this is not a failsafe method of ensuring the most accurate information.

- UK Biobank are committed to ensuring participants continue to be kept aware of progress and developments in UK Biobank, to obtaining their opinions on proposed enhancements to the study, and to providing them with information on how to contact UK Biobank should they have any questions. UK Biobank do this in part through UK Biobank’s annual newsletter, sent via e-mail and/or postal mail. Clearly, UK Biobank need up-to-date contact details (especially postal address) to do this. UK Biobank would like to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress.

- Important potential research questions about the effects of environmental exposures on health require information about current and previous postal address since this allows these exposures to be estimated (from postcodes). These include, for example, valuable studies of the health effects of air and noise pollution, living close to power cables and aspects of the local environment, such as nearby local public amenities and green space.

For each of these datasets listed above, UK Biobank require linkage only for the 500,000 UK Biobank participants who joined the study and who were recruited across the UK. It should be noted that the UK Biobank are only requesting individual-level identifying data to ensure correct linkage. UK Biobank do not make available to researchers any identifiable data (such as postcode, date of birth, name, NHS number, etc.) and instead, each research project is provided with a unique encrypted 9-digit identification number. Hence, it is not possible for a researcher to identify a UK Biobank participant using the data that UK Biobank release.

UK Biobank is the data controller of the data supplied by NHS Digital. Pursuant to a contract between UK Biobank and The Clinical Trials Service Unit (CTSU) at the University of Oxford, CTSU act as a data processor of UK Biobank’s data. CTSU is only able to act on the instructions of UK Biobank and is thus only a data processor. CTSU has no ability to process the data for any other purpose save for those purposes which UK Biobank (alone) determines.

The data supplied by NHS Digital are stored and processed at the CTSU under the sole direction of UK Biobank and pursuant to the contract between UK Biobank and CTSU (which confirms these data controller / data processor arrangements).

UK Biobank is a company limited by guarantee and registered at Companies House – it therefore has its own legal personality. Whist it receives funding from, amongst others the Wellcome Trust and the Medical Research Council, these organisations have no access to the data and cannot exercise any control in respect of its use. Once NHS Digital data has been received by UK Biobank (acting through its processor CTSU), it is uploaded to the UK Biobank resource, at which point registered researchers can apply to UK Biobank to use these linked data (alongside the other data in the resource) to undertake health related research which is in the public interest. Each application to use the resource is considered by UK Biobank pursuant to UK Biobank access policy and procedures, which were developed through extensive consultation, including public consultation and consultation with UK Biobank participants. Further detail is given within the Outputs section. UK Biobank has sole control of this process. De-identified data is made available to researchers following approval of a research application once they have signed a contract with UK Biobank (known as a Material Transfer Agreements (MTA)) which controls how the data can be used (including ensuring the data is only used for conducting health related research which is in the public interest).

Access to the resource:

UK Biobank’s access policy is set out in the Ethics and Governance Framework (https://www.ukbiobank.ac.uk/wp-content/uploads/2011/05/EGF20082.pdf). Its Access Procedures (http://www.ukbiobank.ac.uk/wp-content/uploads/2012/09/Access-Procedures-2011.pdf) were prepared during 2010 to 2011 and were the subject of extensive discussion and dialogue with its funders, the REC and the Ethics and Governance Committee. The Access Procedures were also put out to public consultation, including consultation with UK Biobank participants.

In summary:

- Every researcher has to register and apply to UK Biobank. If successful, they have to execute UK Biobank's Material Transfer Agreement (MTA). The application process is the same irrespective of the nature of the applicant and the applicant's location. The MTA governs the legal relationship between UK Biobank and each researcher and, inter alia, ensures that the researcher complies with all the applicable data protection requirements relating to all participant data that they receive. In addition to audit provisions (i.e. UK Biobank is entitled to audit any applicant if it transpires that a transgression has or may have taken place) the MTA contains explicit provisions stating that UK Biobank will bar a transgressing institution from any further use of the resource as well as publicly informing its governing bodies, funders et al;

- The criteria are that the applicant must be a bona fide researcher conducting health-related research in the public interest. Note that the researcher may be based internationally, but that the terms under which that international researcher may access the data is set out within the MTA and are the same as those for a UK researcher. The adjudication of each application is made by UK Biobank’s Access and Scientific teams, with independent ethical advice (from the University of Oxford’s multidisciplinary bioethics research centre, Ethox: http://www.ndph.ox.ac.uk/research/ethox-centre) as well as input from the UK Biobank Ethics Advisory Committee (http://www.ukbiobank.ac.uk/wp-content/uploads/2018/10/ukb-EACToR-march-2018-item-9a-2.pdf), and the whole process is overseen by UK Biobank’s Access Sub Committee (http://www.ukbiobank.ac.uk/access-to-the-resource/), which contains both scientific and legal / ethical expertise;

- Any data released to researchers is pseudonymised and encrypted. The researcher is only permitted to use the data for a particular (approved) research project for a particular (approved) time frame after which the researcher must securely destroy the data. The results of the research have to come back to UK Biobank – where they are then accessible by other researchers – and the researcher is obliged to publish their work;

- In other words, any data supplied by UK Biobank to a researcher can only be used for the research project in question and for a limited time through UK Biobank’s access procedures. Further, any such data cannot be sub-licensed or made available by researchers to any other party under any circumstances. This is all covered in UK Biobank’s MTA, which is non-negotiable. Access to the resource can be made by any bona fide researcher in the world. There is no restriction on the location of the researcher.

UK Biobank has approved over 1,600 applications since it first opened to researches in 2012. List of the approved applications can be found here:

http://www.ukbiobank.ac.uk/approved-research/

In addition there have been over 1000 publications from researchers using the UK Biobank resources on a wide variety of topics including genetics, diet, exercise, mental health, respiratory disease and many more topics. The full list of publications based on the resource here: http://www.ukbiobank.ac.uk/published-papers/

UK Biobank is a major national and international health resource, and a registered charity in its own right, with the aim of improving the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses – including cancer, heart diseases, stroke, diabetes, arthritis, osteoporosis, eye disorders, depression and forms of dementia. UK Biobank recruited 500,000 people aged between 40-69 years in 2006-2010 from across the country to take part in this project. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed via linkage to their health and health related records. Over many years this will continue to build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

UK Biobank requires data from NHS Digital to pursue its legitimate interests which are to set up and manage a major international research resource for health-related research that is in the public interest. UK Biobank is a prospective resource with the specific objective of following up its participants through the health records. This enables researchers to study the actual causes of disease and in turn develop treatments (for common and rare disease) which will be of considerable benefit to the public. Without the health record data the UK Biobank resource is simply not able to achieve its aims. This would, in UK Biobank’s view, constitute a (very)substantial public interest.

The UK Biobank resource is open to researchers who can apply to use the data within the resource (which data is de-identified) to undertake health related research which is in the public interest. The addition of the requested data from NHS Digital to the resource enhances the range of information which is available to researchers to enable them to undertake their approved research project. UK Biobank, as a charitable company limited by guarantee, was established to enable researchers to undertake health related research which is in the public interest and therefore the data is required from NHS Digital to enable UK Biobank to pursue its legitimate interests.

As a result of UK Biobank’s work, patients and the wider public benefit from the advances made in the prevention, diagnosis and treatment of disease. UK Biobank is now one of the largest and most used health research resources in the world. Over 13,000 researchers are registered with UK Biobank and over 1,600 health-related research applications have been approved.

Participants have signed consent forms permitting access to their data for this purpose. It is considered very unlikely that participants would object to the processing activities involved in obtaining linked data from NHS Digital. During the past 10 years since participants joined UK Biobank during 2006-10, fewer than 800 of the 500,000 participants have withdrawn from the study and asked that their information be deleted. For clarity, UK Biobank relies on GDPR 6 (1) (a) - explicit consent for the collection and storage of bio samples and the permission to make available for health related research. The data from NHS Digital is secondary health care data and therefore UK Biobank relies on Legitimate Interests (GDPR Article 6(1)(f) ) in regards to the data they obtain from NHS Digital.

The data are stored in a way that makes it is extremely difficult even for UK Biobank and their data processors (University of Oxford) to re-identify participants. Only a very few individuals within the University of Oxford are allowed to do so (and they are strictly monitored) in order that further information about participants can be added. Data provided to researchers have personal identifiers removed so that an individual participant cannot be identified. There are no circumstances in which data can be processed in a manner that could have an adverse impact on the data subject.

The overall purpose of UK Biobank is to create a prospective epidemiological resource of 500,000 people aged 45 - 69 at the time of recruitment from around the UK. The methods to be used for the identification, invitation and assessment of participants have been developed following extensive consultation with leading research groups in the UK and internationally, and with research participants and representatives of the general population. All participants in the UK Biobank study have given fully informed consent for access to and long-term storage of information from their medical and other health-related records, and use of this and other information for health-related research purposes; long-term storage and use of their blood and urine samples for health-related research purpose; and recontact by UK Biobank to invite them (but without any obligation) to answer further questions or take part in further assessments.

UK Biobank is a unique resource, because of the combination of its very large size and the range and detail of data that it contains on its participants, including extensive phenotype and genotype information on each of them. Further, rather than recruiting participants with specific diseases or risk factors, UK Biobank is a prospective population-based resource established to support research into the causes of a wide range of diseases affecting people in middle and older age. Thus, the ability to link to participant’s health records, so that specific outcomes occurring during long-term follow-up of the participants can be identified and appropriate disease-based research carried out, is critical.

UK Biobank’s priority is to combine extensive and precise measurement of exposures, along with the detailed and rigorous follow-up for a wide range of health related outcomes, and to promote innovative science by maximising secure access to deidentified data.

UK Biobank was established in 2003. It is funded by the Wellcome Trust, the Department of Health and the Medical Research Council. No funder is permitted to influence the outcomes nor suppress any outcomes of research.

The recruitment of its 500,000 participants took place between 2007 and 2010, whose health is now being followed long term primarily through linking to NHS health data. Access to the resource opened in 2012, since which time there have been over 1,600 applications submitted and over 13,000 researchers have registered to use the resource, resulting in 1,000 research papers published using the UK Biobank resource.

At baseline (i.e. at time of recruitment), UK Biobank obtained and holds identifiable information on its participants. They have undergone measures, provided blood, urine and saliva samples for future analysis, detailed information about themselves and agreed to have their health followed by linking to their health and health related records. Over many years this will build into a powerful resource to help scientists discover why some people develop particular diseases and others do not.

Under this Agreement, UK Biobank receives updates on participants’ health and health-related data. This includes:

i. Hospital Episode Statistics – details of participants attendances at hospital for any reason. 1997/98 to latest available annual year;

ii. Mental Health data – contains record-level data about the care of

participants who are in contact with mental health, learning disabilities or autism spectrum disorder services. Mental Health Minimum Data Set 2006/07 to 2014/15, Mental Health and Learning Disabilities Data Set 2014/15 to November 2015/16, Mental Health Services Data Set 2016/17 to latest available annual year;

iii. Diagnostic Imaging data – collects information about the imaging tests and scans carried out on participants such as x-rays, MRI scans or ultrasound. 2012/13 to latest available annual year;

iv. Civil Registration Mortality data – details of participants’ deaths including date and cause. 2011/12 to latest available with a bi-annual update;

v. Cancer Registration data – details of any cancers participants’ are registered as having. 2011/12 to latest available with a bi-annual update;

vi. Emergency Care Data Set data - details around participant attendance in A&E. October 2017 to latest available annual year;

vii National Diabetes Audit data - data collected on diabetes in participants. April 2003 to latest available annual year;

viii. Improving Access to Psychological Therapies (IAPT)- measure of anxiety and depression in participants. April 2012 to latest available annual year;

As a large cohort it is not feasible for UK Biobank itself to follow up each of its participants on an individual basis regularly and nor would this generate the necessary clinical detail. Self-reported data does not possess (anything like) the needed coverage, reliability and consistency, that the health records contain. As such, the data provided via NHS Digital is crucial to the viability of UK Biobank as the primary mechanism for follow up of the cohort, particularly as it provides coverage for all 500,000 participants in an entirely unobtrusive (i.e. passive) manner. The linked datasets all form part of the participants’ health record and are thus all vital to providing the necessary health record information. Of key importance are datasets related to death and cancer registry notifications and hospital activity. Without access to these data, UK Biobank would not be a prospective resource and research on the associations of lifestyle and genetic factors on the risk of developing a wide range of diseases over time would simply not be possible. This is because these datasets provide vital information on which participants have developed which diseases and at what time. Given UK Biobank is a prospective study available to registered researchers worldwide, research interest covers a very wide range of health outcomes, including cardiovascular, metabolic, endocrine, neurodegenerative, cancers, stroke, mental health, and many others. As such, access to data on deaths, cancers and hospital admissions and related activities is of vital importance.

In addition to the above-mentioned datasets that UK Biobank hold, other health-related datasets that have now become available are also of high research interest. For example, the IAPT and mental health datasets will help to identify participants with mental health disorders and enable research into their treatment response and support required. The diagnostic imaging dataset will enable research into the treatment pathways for a wide range of conditions, while the emergency department data will enable research into the reasons for and consequences of acute admissions. Finally, the national diabetes audit will help to identify participants diagnosed with diabetes and their management pathways (as these participants are often under-represented in the HES data). UK Biobank require access to all retrospective data to enable epidemiological research into antecedent clinical events prior to diagnosis and is crucial for researchers assessing the effect of multi-morbidity on future health outcomes. Without these linked data provided by NHS Digital, researchers would not be able to fully ascertain the full burden of disease in the cohort or effects of particular treatment and management pathways on health. All of the datasets requested enhance the UK Biobank resource and allow researchers to undertake health-related research that is in the public interest.

Other datasets include:

ix. Personal Demographic Service (PDS) updates – details of changes to participants’ demographics, postal addresses or registered GP practices.

It is very important for UK Biobank to be able to maintain accurate information about participants’ contact details (including postal address) and registered GP practice (which were up to date for the whole cohort at the time of recruitment but there will be undoubtedly have been changes since then). There are several reasons for this requirement:

- Ongoing approved linkages to primary care data require accurate information about registered GP practice (at practice as well as CCG level) to ensure that data about a particular participant are requested from the correct GP practice.

- Direct contact with GPs is required as part of the UK Biobank imaging sub study (https://imaging.ukbiobank.ac.uk/) in the event that a participant is identified with a potentially serious incidental finding (as noted by the radiographer and then confirmed by a specialist doctor (radiologist)). Should such an abnormality be found, UK Biobank will write to the participant and their GP to allow them to further investigate these potentially serious findings.

- To enable GPs to be confident about extraction of primary care data from GP practice system suppliers to UK Biobank, UK Biobank will need to be able to provide GP practices with up-to-date listings of UK Biobank participants registered at their practices, and – where requested – evidence of their consent.

- Follow-up in a cohort study needs to be as complete as possible. This is to avoid both unnecessary reduction in size of the cohort (which reduces statistical power of data analyses), and loss-to-follow-up bias (which can cause misleading research results). Part of UK Biobank’s follow-up strategy relies on intermittent re-contact with UK Biobank’s participants to invite them to complete brief questionnaires about health related exposures or outcomes. It is important that UK Biobank have correct and up-to-date participant contact details for this. Although UK Biobank encourage participants to inform UK Biobank when they move, this is not a failsafe method of ensuring the most accurate information.

- UK Biobank are committed to ensuring participants continue to be kept aware of progress and developments in UK Biobank, to obtaining their opinions on proposed enhancements to the study, and to providing them with information on how to contact UK Biobank should they have any questions. UK Biobank do this in part through UK Biobank’s annual newsletter, sent via e-mail and/or postal mail. Clearly, UK Biobank need up-to-date contact details (especially postal address) to do this. UK Biobank would like to avoid contacting participants at the incorrect address, as this may cause confusion or even occasional distress.

- Important potential research questions about the effects of environmental exposures on health require information about current and previous postal address since this allows these exposures to be estimated (from postcodes). These include, for example, valuable studies of the health effects of air and noise pollution, living close to power cables and aspects of the local environment, such as nearby local public amenities and green space.

For each of these datasets listed above, UK Biobank require linkage only for the 500,000 UK Biobank participants who joined the study and who were recruited across the UK. It should be noted that the UK Biobank are only requesting individual-level identifying data to ensure correct linkage. UK Biobank do not make available to researchers any identifiable data (such as postcode, date of birth, name, NHS number, etc.) and instead, each research project is provided with a unique encrypted 9-digit identification number. Hence, it is not possible for a researcher to identify a UK Biobank participant using the data that UK Biobank release.

UK Biobank is the data controller of the data supplied by NHS Digital. Pursuant to a contract between UK Biobank and The Clinical Trials Service Unit (CTSU) at the University of Oxford, CTSU act as a data processor of UK Biobank’s data. CTSU is only able to act on the instructions of UK Biobank and is thus only a data processor. CTSU has no ability to process the data for any other purpose save for those purposes which UK Biobank (alone) determines.

The data supplied by NHS Digital are stored and processed at the CTSU under the sole direction of UK Biobank and pursuant to the contract between UK Biobank and CTSU (which confirms these data controller / data processor arrangements).

UK Biobank is a company limited by guarantee and registered at Companies House – it therefore has its own legal personality. Whist it receives funding from, amongst others the Wellcome Trust and the Medical Research Council, these organisations have no access to the data and cannot exercise any control in respect of its use. Once NHS Digital data has been received by UK Biobank (acting through its processor CTSU), it is uploaded to the UK Biobank resource, at which point registered researchers can apply to UK Biobank to use these linked data (alongside the other data in the resource) to undertake health related research which is in the public interest. Each application to use the resource is considered by UK Biobank pursuant to UK Biobank access policy and procedures, which were developed through extensive consultation, including public consultation and consultation with UK Biobank participants. Further detail is given within the Outputs section. UK Biobank has sole control of this process. De-identified data is made available to researchers following approval of a research application once they have signed a contract with UK Biobank (known as a Material Transfer Agreements (MTA)) which controls how the data can be used (including ensuring the data is only used for conducting health related research which is in the public interest).

Access to the resource:

UK Biobank’s access policy is set out in the Ethics and Governance Framework (https://www.ukbiobank.ac.uk/wp-content/uploads/2011/05/EGF20082.pdf). Its Access Procedures (http://www.ukbiobank.ac.uk/wp-content/uploads/2012/09/Access-Procedures-2011.pdf) were prepared during 2010 to 2011 and were the subject of extensive discussion and dialogue with its funders, the REC and the Ethics and Governance Committee. The Access Procedures were also put out to public consultation, including consultation with UK Biobank participants.

In summary:

- Every researcher has to register and apply to UK Biobank. If successful, they have to execute UK Biobank's Material Transfer Agreement (MTA). The application process is the same irrespective of the nature of the applicant and the applicant's location. The MTA governs the legal relationship between UK Biobank and each researcher and, inter alia, ensures that the researcher complies with all the applicable data protection requirements relating to all participant data that they receive. In addition to audit provisions (i.e. UK Biobank is entitled to audit any applicant if it transpires that a transgression has or may have taken place) the MTA contains explicit provisions stating that UK Biobank will bar a transgressing institution from any further use of the resource as well as publicly informing its governing bodies, funders et al;

- The criteria are that the applicant must be a bona fide researcher conducting health-related research in the public interest. Note that the researcher may be based internationally, but that the terms under which that international researcher may access the data is set out within the MTA and are the same as those for a UK researcher. The adjudication of each application is made by UK Biobank’s Access and Scientific teams, with independent ethical advice (from the University of Oxford’s multidisciplinary bioethics research centre, Ethox: http://www.ndph.ox.ac.uk/research/ethox-centre) as well as input from the UK Biobank Ethics Advisory Committee (http://www.ukbiobank.ac.uk/wp-content/uploads/2018/10/ukb-EACToR-march-2018-item-9a-2.pdf), and the whole process is overseen by UK Biobank’s Access Sub Committee (http://www.ukbiobank.ac.uk/access-to-the-resource/), which contains both scientific and legal / ethical expertise;

- Any data released to researchers is pseudonymised and encrypted. The researcher is only permitted to use the data for a particular (approved) research project for a particular (approved) time frame after which the researcher must securely destroy the data. The results of the research have to come back to UK Biobank – where they are then accessible by other researchers – and the researcher is obliged to publish their work;

- In other words, any data supplied by UK Biobank to a researcher can only be used for the research project in question and for a limited time through UK Biobank’s access procedures. Further, any such data cannot be sub-licensed or made available by researchers to any other party under any circumstances. This is all covered in UK Biobank’s MTA, which is non-negotiable. Access to the resource can be made by any bona fide researcher in the world. There is no restriction on the location of the researcher.

UK Biobank has approved over 1,600 applications since it first opened to researches in 2012. List of the approved applications can be found here:

http://www.ukbiobank.ac.uk/approved-research/

In addition there have been over 1000 publications from researchers using the UK Biobank resources on a wide variety of topics including genetics, diet, exercise, mental health, respiratory disease and many more topics. The full list of publications based on the resource here: http://www.ukbiobank.ac.uk/published-papers/