TwinsUK: Phenotypic enrichment of the TwinsUK cohort through linkage to electronic health records and other databases.

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

Objective for processing

[2 paragraphs unchanged] In January 2020 the TwinsUK BioBank Consent Form and supporting Volunteer Information [19 words unchanged] given the Data Linkage Information Sheet and the Data Linkage Decision Form. s251 Support under section 251 NHS Act 2006 is in place for those that did not complete the Data Linkage [33 words unchanged] participants are given the opportunity to complete the TwinsUK Biobank consent form. [1 paragraph unchanged] Twins provide a unique natural design with which to study the genetic [97 words unchanged] enhance TwinsUK as a research resource conducting a programme of epidemiological research. To achieve this purpose, the following data is requested: Cancer registration, Mortality Data, Hospital Episodes Statistics (HES), Mental Health Minimum dataset (MHMDS) and Personal Demographics Service (PDS). Cancer registry data will provide details on exact diagnosis and treatment given, which has been and will continue to be invaluable to understanding the biology of cancer, and long-term effects of treatment. The longitudinal nature of the samples data means that the study can look for predictors and effects both before and after diagnosis. The following NHS England Data will be accessed: HES and the MHMDS provide other diagnoses and time stamps which, together with existing data, can be used to study other physical and mental health conditions. Mortality data provides hard endpoints, including lifespan and cause of death. These health record data will augment and complete existing data collected through visits and questionnaires, and significantly extend the ability of the resource to keep providing public benefit through understanding of disease. TwinsUK are also pursuing linkage to GP records for the cohort to further enhance the quality of the data held. Initially, we planned on obtaining GP records via GP software systems. Under this route, we would liaise with the GP software providers, establish GP assent for the data extraction and then extract the primary care data for our participants. This was the most viable option at the time our data linkage programme was first developed. Since then, other options have been assessed. As per other longitudinal cohorts, we are awaiting the Goldacre review focusing on the more efficient and safe use of health data for research and analysis in order to make a firm decision. • Hospital Episode Statistics (HES) o Admitted Patient Care o Accident & Emergency o Critical Care o Outpatients • Mental Health • Civil Registration Mortality • Cancer Registration • Demographics Cancer Registration data will provide details on exact diagnosis and treatment given, which has been and will continue to be invaluable to understanding the biology of cancer, and long-term effects of treatment. The longitudinal nature of the samples data means that the study can look for predictors and effects both before and after diagnosis. HES and the Mental Health data provide other diagnoses and time stamps which, together with existing data, can be used to study other physical and mental health conditions. Mortality data provides hard endpoints, including lifespan and cause of death. These health record data will augment and complete existing data collected through visits and questionnaires, and significantly extend the ability of the resource to keep providing public benefit through understanding of disease. [6 paragraphs unchanged] TwinsUK commits to making every reasonable effort to inform volunteers about the use of their records, respecting objections and continuing to provide fair processing on a regular basis. TwinsUK will mitigate risks and ethical concerns by: only allowing approved researchers to access the data for approved projects; ensuring that researchers only see pseudonymised data and only accessing variables pertinent to their hypothesis (i.e. minimise the data before providing managed access). TwinsUK will mitigate risks and ethical concerns by: only allowing approved King’s College London (KCL) researchers who are substantive employees to access the data for approved project; ensuring that researchers only see pseudonymised data and only accessing variables pertinent to their hypothesis (i.e. minimise the data before providing managed access). [1 paragraph unchanged] Linkage in the context of TwinsUK within the Department of Twin Research at KCL. [10 paragraphs unchanged] Data subjects subjects: [7 paragraphs unchanged] The purpose of the request [8 paragraphs unchanged] This request aims The aim is to gain accurate health and social care data on participants of the [56 words unchanged] order to determine if the findings are robust and can be generalised. The data linkage programme – in this application - seeks approval to link to, extract and use NHS England data (Cancer registration, Mortality Data, Hospital Episode Statistics - HES (Admitted Patient Care , Accident and Emergency, Critical Care, Outpatients), Mental Health Services Data Set- MHMDS and Personal Demographics Service-PDS) for the TwinsUK participants. Only King’s College London approved bona-fide researchers (who are KCL employees, KCL PhD students and PhD Students having signed the IP Assignment Waiver will have the opportunity to use the TwinsUK research database in conjunction with NHS England data under this Data Sharing Agreement (DSA). Only these researchers will have access to the secure platform and analysis tools to work on the linked data and can only take aggregate results out of the system. Only King’s College London approved bona-fide researchers (who are KCL employees, KCL PhD students and PhD Students having signed the IP Assignment Waiver will have the opportunity to use the TwinsUK research database in conjunction with NHS England data. Only these researchers will have access to the secure platform and analysis tools to work on the linked data and can only take aggregate results out of the system. Projects needing access to data for further analysis will require the submission of a Department of Twin Research (DTR) Data Access Request form and the proposal will be reviewed by the TwinsUK Data Access Committee. The linked health record data will solely be used to conduct statistical scientific research in line with the GDPR Articles 6(1)(e) and 9(2)(j) and will not be used for any other purpose. Projects needing access to data for further analysis will require the submission of a Department of Twin Research (DTR) Data Access Request form and the proposal will be reviewed by the TwinsUK Data Access Committee. [1 paragraph unchanged] As per TwinsUK lifetime study-based data, TwinsUK are requesting require access to all the life-course data from birth up to the most [57 words unchanged] and linked records from other providers (e.g. national pupil database records). The data request Data is limited to the TwinsUK cohort participants. TwinsUK participants mainly live in [5 words unchanged] Ireland. Therefore, the present application focuses on the territory covered by England/Wales. A minimal number of identifiable data (personal identifiers) are requested required for each of the datasets to obtain accurate “linkage data”. For historic MRIS reports previously received from NHS England, KCL will undertake a review of the provided data fields and determine whether these are still required or whether it would be appropriate to minimise the data held by destroying some of these fields. Once received and prior to storing the linkage data in the Data [6 words unchanged] will be removed from the research data prior to the pseudonymisation process. [1 paragraph unchanged] The healthcare records requested Data are minimised to TwinsUK participants and those that stay within the campaign [17 words unchanged] redundant variables are excluded. Unless further consent is obtained from participants, the requested data Data will be aligned with HRA CAG minimisation requirements for variables of elevated sensitivity. The level of data that has been requested is paramount to the dataset. [3 paragraphs unchanged] Organisations involved involved: [6 paragraphs unchanged]

Processing activities

[1 paragraph unchanged] TwinsUK Administrative Database Database: [2 paragraphs unchanged] Linkage Data Provider (NHS England) England): [1 paragraph unchanged] TwinsUK Data Safe Haven Haven: The linkage linked data, when procured obtained from NHS England, arrive at the TwinsUK Data Safe Haven which provides [7 words unchanged] safe storage and processing of these sensitive data by authorised personnel only. [6 paragraphs unchanged] TwinsUK Data Repository Repository: [3 paragraphs unchanged] KCL Secure Research Platform Platform: [3 paragraphs unchanged] Data Storage UK Longitudinal Linkage Collaboration: TwinsUK is part of the UK Longitudinal Linkage Collaboration (UK LLC). UK LLC is managed and owned by the University of Bristol as data controller. It operates a Trusted Research Environment (TRE) containing de-identified person-level data and encrypted address-level data from contributing longitudinal population studies and linked data sources. The UK LLC aims to facilitate access to longitudinal study data, including TwinsUK data, to authorised researchers via the Secure eResearch Platform (SeRP UK). To support this, TwinsUK needs to share information about its cohort with UK LLC. This Data Sharing Agreement (DSA) authorises TwinsUK to share the minimum necessary subsets of the Data with the UK LLC to enable UK LLC’s data linkages with NHS England and other regional providers as permitted under separate DSAs between the University of Bristol and the respective organisations. Data Storage: [16 paragraphs unchanged] All organisations party to this agreement must comply with the data sharing framework contract requirements, including those regarding the use (and purposes of that use) by “personnel” (as defined within the data sharing framework contract i.e. employees, agents and contractors of the data recipient who may have access to that data). The Department of Twin Research at King’s College London has PhD students and some of these may be users of the enhanced resource. These students’ applications to use the data will be assessed along with all others and must include a qualified supervisor who takes responsibility for the conduct of the research. The PHD students have an employment contract with Kings College London; any individuals without a signed IP Assignment Waiver will not be permitted access to NHS England data. There will be no data linkage undertaken with NHS England data provided under this agreement that is not already noted in the agreement. The Department of Twin Research at King’s College London currently have 21 PhD students and some of these may be users of the enhanced resource. These students’ applications to use the data will be assessed along with all others and must include a qualified supervisor who takes responsibility for the conduct of the research. The PHD students have an employment contract with Kings College London; any individuals without a signed IP Assignment Waiver will not be permitted access to NHS England data. [1 paragraph unchanged]

Benefits reported

[1 paragraph unchanged] Data provided under this agreement have been transferred in summer 2023 while data analysis only started in 2024. Therefore, expected benefits have not been achieved yet. Benefits should be achieved within the term of the new agreement.

Unchanged: Expected output, Expected measurable benefits.

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

Objective for processing

[45 paragraphs unchanged] The data linkage programme – in this application - seeks approval to link to, extract and use NHS Digital England data (Cancer registration, Mortality Data, Hospital Episode Statistics - HES (Admitted Patient [10 words unchanged] Services Data Set- MHMDS and Personal Demographics Service-PDS) for the TwinsUK participants. Only King’s College London approved bona-fide researchers (who are KCL employees, KCL [13 words unchanged] the opportunity to use the TwinsUK research database in conjunction with NHS Digital England data. Only these researchers will have access to the secure platform and [6 words unchanged] linked data and can only take aggregate results out of the system. [2 paragraphs unchanged] As per TwinsUK lifetime study-based data, TwinsUK are requesting access to all [53 words unchanged] alongside study collected information, linked health and social care records from NHS Digital England and linked records from other providers (e.g. national pupil database records). The [19 words unchanged] Ireland. Therefore, the present application focuses on the territory covered by England/Wales. A minimal number of identifiable data (personal identifiers) are requested for each of the datasets to obtain accurate “linkage data”. For historic MRIS reports previously received from NHS Digital, England, KCL will undertake a review of the provided data fields and determine [40 words unchanged] will be removed from the research data prior to the pseudonymisation process. [6 paragraphs unchanged] • Avon Longitudinal Study of Parents and Children (ALSPAC): ALSPAC (University of [55 words unchanged] on the data linkage process but without ALSPAC actually accessing any NHS Digital England data involved in the linkage. Data under this Agreement will not be linked with ALSPAC Data. [4 paragraphs unchanged] None of the organisations listed above are intended to have access to the data provided by NHS Digital, England, but they may provide data to King's College London which could then be linked with the NHS Digital England data.

Processing activities

TwinsUK systems and policies are bound by TwinsUK's commitment to comply with information governance standards included in the NHS Digital England DSP Toolkit 2019. To comply with these, TwinsUK are required to put [24 words unchanged] organisational structure and methodologies that provide a means to meet these requirements. [2 paragraphs unchanged] In order to gain access to health-related data, a very small but identifiable set of data has to be divulged to NHS Digital. England. Participants’ Name, Date of Birth (DOB), Gender and Address are provided to the NHS Digital Engalnd Personal Demographics Service (PDS) and the NHS ID, DOB and General Practices [12 words unchanged] Administrative Database. NHS ID and DOB of participants are provided to NHS Digital England and a number of health-related datasets are received into the TwinsUK Data [39 words unchanged] end-to-end internet connection safeguarding any sensitive data that is passed between systems. Linkage Data Provider (NHS Digital) England) Personal identifiers received by NHS Digital England and the PDS will be limited to the minimum requirements to fulfil the particular request. Whilst the NHS Digital England PDS will be used to confirm or upscale the TwinsUK Administrative Data, NHS Digital England will feed Cancer data, mortality data, Hospital Episodes Statistics and Mental Health [9 words unchanged] These identifiable data are transferred in a secure data tunnel as above. [1 paragraph unchanged] The linkage data, when procured from NHS Digital, England, arrive at the TwinsUK Data Safe Haven which provides a secure and segregated area for the safe storage and processing of these sensitive data by authorised personnel only. [1 paragraph unchanged] The use of identifiable NHS Digital England data in TwinsUK research requires either consent from the cohort participant or [172 words unchanged] activities with identifiable data and staff members undertaking activities with non-identifiable data. [16 paragraphs unchanged] A Data Access Committee (comprised of senior management, data personnel, principal Investigators, [45 words unchanged] KCL researchers. The data access committee will ensure full transparency with NHS Digital England and will not permit any use of data for research that is not in line with the stated purpose of this NHS Digital England Data Sharing Agreement. KCL researchers will need to ensure their outputs are direct and tangible with fulfilling NHS Digital England criteria when requesting for linked data, and these factors will be scrutinized [23 words unchanged] linked data KCL hold, TREC will direct them to apply to NHS Digital England independently. [2 paragraphs unchanged] Name- Linkage purposes: TwinsUK will use “name” for linkage purposes with NHS Digital England Personal Demographic System (flagging and tracing). Date of birth- Linkage and analysis purpose: TwinsUK will use “date of birth” for linkage purposes with NHS Digital England Personal Demographic System (flagging and tracing). Subsequently TwinsUK will use “date of birth” (needed as a check variable) as one of the linkage identifiers for extraction with NHS Digital. England. “Year of birth” will be used to calculate the age of the [25 words unchanged] for many of the researchers in the department who study ageing-related processes. [2 paragraphs unchanged] Gender- Linkage and analysis purpose: TwinsUK will use “gender” for linkage to NHS Digital England Personal Demographic System (flagging and tracing) and the Department for Education. [2 paragraphs unchanged] Address (current/historic)- Linkage purposes: TwinsUK will use “address” (current/historic) for linkage purposes with NHS Digital England Personal Demographic System (flagging and tracing). NHS Number- Linkage purposes: TwinsUK will use “NHS ID” as the main linkage identifiers for data extraction from NHS Digital. England. General Practice Registration Details- Linkage purposes: TwinsUK will ask NHS Digital England to provide the details of the general practice at which the participant [11 words unchanged] seek GP assent to access the primary care records of the participants. [1 paragraph unchanged] There will be no data linkage undertaken with NHS Digital England data provided under this agreement that is not already noted in the agreement. The Department of Twin Research at King’s College London currently have 21 [56 words unchanged] a signed IP Assignment Waiver will not be permitted access to NHS Digital England data. [1 paragraph unchanged]

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

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

Datasets: + Cancer Registration Data; + Civil Registrations of Death; + Demographics; + Emergency Care Data Set (ECDS); + Hospital Episode Statistics Accident and Emergency (HES A and E); + Hospital Episode Statistics Admitted Patient Care (HES APC); + Hospital Episode Statistics Critical Care (HES Critical Care); + Hospital Episode Statistics Outpatients (HES OP); + Mental Health Minimum Data Set (MHMDS)

Objective for processing

TwinsUK is a data registry within King's College London. Longitudinal cohorts like TwinsUK are an important source of information on life course on health and social development. TwinsUK is a very productive resource providing insight into many health questions, with over 1,000 research publications to date, producing insight particularly on the biology of ageing for public benefit. The purpose of the original study was to: Through its Data Linkage programme, TwinsUK will also establish a mechanism whereby future follow-up will be efficient, both in terms of resource efficiencies and also in breadth of data collection, and also where the participant burden of follow-up can be reduced. It is hoped to add general value to epidemiological research through creating methodological solutions that can be shared with other researchers and applied to other studies. Health records linkage will enhance case identification and can track key traits such as body mass index, blood pressure, electronic frailty index, service utilisation and medication. All of this in turn will hopefully benefit the public by enabling better health research that is more representative of the public, for example enabling more complete ascertainment of outcomes, especially for groups less able to visit. For scientific reasons, it is important that TwinsUK collects information on as full and representative a sample originally enrolled as possible. It is hoped to maximise statistical power, minimise participant bias and avoid exclusion of marginalised and underrepresented groups. Thus the research findings from the cohort will have greater accuracy, relevance and impact in improving health of the nation. • To assess the extent to which age-related deterioration is correlated between different organ systems (cardiovascular, muscle, bone, respiratory function, and vision). In January 2020 the TwinsUK BioBank Consent Form and supporting Volunteer Information Sheet were used to recruit participants to be included in the research data base. From 2020 onwards participants were given the Data Linkage Information Sheet and the Data Linkage Decision Form. s251 is in place for those that did not complete the Data Linkage Decision Form and for those recruited prior to January 2020 who have not been seen or completed a questionnaire recently. It is envisaged that the need for s251 support will decline as more participants are given the opportunity to complete the TwinsUK Biobank consent form. • To assess how much of variation in longitudinal rates of physiological deterioration is due to genetic and environmental variation. TwinsUK aims to strike a balance between maximising the opportunity for individuals to participate whilst minimising inappropriate intrusiveness. Following the approval of section 251 support, the described data linkage fair processing campaign offers an opportunity for all TwinsUK participants to object if they so wish. Research use of the routine records will be made clear to participants through communications and face to face engagement in clinic, remote visits (using zoom) or over the phone, and it will be explained that they are free to change their decision or to make their decision now if they have not previously done so, by indicating appropriately on the new TwinsUK Biobank consent form they are required to review and complete every time they take part in the clinic research. • To investigate genetic associations with the five organ systems using candidate gene and genome wide association studies Twins provide a unique natural design with which to study the genetic and environmental factors that influence disease. The primary objective of TwinsUK is to investigate how environmental factors and genetics interact to impact health and disease over the life course. TwinsUK offers a multidimensional approach to the study of human health and individuality, availing health researchers with a portfolio of methods to observe the effects of both genes and environment on development, health and ageing. This is because the unique natural pairing of twins allows key factors to be held stable while the impact of others is investigated. The principal objective of this purpose is to consolidate and enhance TwinsUK as a research resource conducting a programme of epidemiological research. To achieve this purpose, the following data is requested: Cancer registration, Mortality Data, Hospital Episodes Statistics (HES), Mental Health Minimum dataset (MHMDS) and Personal Demographics Service (PDS). • To determine the relative influences on biological ageing (as measured by loss of function/tissue) of environmental factors such as marital status, socio-economic status (income, education, occupation), levels of physical exercise, smoking and alcohol intake, number of children. Cancer registry data will provide details on exact diagnosis and treatment given, which has been and will continue to be invaluable to understanding the biology of cancer, and long-term effects of treatment. The longitudinal nature of the samples data means that the study can look for predictors and effects both before and after diagnosis. • To investigate the value of putative biomarkers of ageing: serum vitamin D, DHEAS, C-reactive protein, creatinine, retinal vascular calibre, and white cell telomere length HES and the MHMDS provide other diagnoses and time stamps which, together with existing data, can be used to study other physical and mental health conditions. Mortality data provides hard endpoints, including lifespan and cause of death. These health record data will augment and complete existing data collected through visits and questionnaires, and significantly extend the ability of the resource to keep providing public benefit through understanding of disease. TwinsUK are also pursuing linkage to GP records for the cohort to further enhance the quality of the data held. Initially, we planned on obtaining GP records via GP software systems. Under this route, we would liaise with the GP software providers, establish GP assent for the data extraction and then extract the primary care data for our participants. This was the most viable option at the time our data linkage programme was first developed. Since then, other options have been assessed. As per other longitudinal cohorts, we are awaiting the Goldacre review focusing on the more efficient and safe use of health data for research and analysis in order to make a firm decision. To achieve this, the following data was requested: Cancer registration, Mortality Data and Demographics data. Cancer registry data provided details on exact diagnosis and treatment given, which has been and will continue to be invaluable to understanding the biology of cancer, and long-term effects of treatment. The longitudinal nature of the study samples data means researchers can look for predictors and effects both before and after diagnosis. TwinsUK operates as a resource for the scientific research community and is continually accepting, subject to review, proposals for the secondary use of the data set. TwinsUK reviews all incoming requests for new research projects that are within its programme of research. This raises the challenge that TwinsUK, in common with all cohort studies, cannot now specify all the precise scientific hypotheses its data may be used to investigate in the future. Due to the very time-consuming processes required to adapt these data to research use it is not practically feasible to extract variables on a case by case basis. Instead TwinsUK propose to collect all health data on participants (within the bounds of the HRA REC ((Health Research Authority - Research Ethics Committee) and CAG (Confidentiality Advisory Group) permissions) and process them ready for research use. Mortality data provided hard endpoints, including lifespan and cause of death. The following points entails the justification and public interest of the proposed programme of research, as data is processed under GDPR (General Data Protection Regulation) Article 6(1)(e) task in the public interest and 9(2)(j) research: Part of the Department of Twin Research, TwinsUK is the largest longitudinal cohort of community dwelling adult twins in the UK, comprising >14,000 volunteer twins now predominantly in midlife (mean age 59) (Verdi S, et al.). Participants have provided self-reported questionnaires, physical/cognitive measures and biological samples through clinical visits approximately every four years, for researchers to investigate how environmental factors and genetics interact to impact health over the life course. Importantly, longitudinal biobanking of samples over the last 28 years enables research on early detection of disease. • Studies such as TwinsUK represent an enormous long-term public investment of effort and resource from scientists and funding agencies and a significant time commitment from the cohort participants. It is essential to find cost-effective ways to maximise scientific return from this investment and to develop solutions to some of the problems that face such studies, in particular the biases associated with self-reported measures (such as errors introduced by recalling events over time) and those that result from participant attrition. Attrition is known to be more prevalent amongst members of the population with certain social or economic determinants of health. If left unchecked this can lead to unrepresentative findings that are difficult to generalise to the wider population. As a longitudinal population study and an established research database, TwinsUK secured funding up to December 2022 by public funding from the Wellcome Trust (212904/Z/18/Z). The proposal funded by the Wellcome Trust has been approved following an open competitive funding process including scientific peer review. This project is part of the portfolio of research funded by the Wellcome Trust, Medical Research Council, European Union, the National Institute for Health Research (NIHR)-funded BioResource, Clinical Research Facility and Biomedical Research Centre based at Guy’s and St Thomas’ NHS Foundation Trust in partnership with King’s College London. • Linkage is a non-intrusive way for researchers to follow up participants confidentially over time. Over the years, some participants may find it difficult to continue to take part in projects at TwinsUK as actively as before. Therefore, data linkage would allow the study to minimise bias in the research and mitigate the effects of participant attrition. From feedback participants really value being able to take part in health research at TwinsUK. Linkage is an easy, effortless way for the participants to continue taking part in the research, and the individuals can continue to feel connected through regular newsletters and other participant engagements including social media channels and the study website. In addition, it is hoped that data linkage will reduce burden on participants when it comes to recalling medical events or prescriptions as part of the research e.g. if there is a study that requires knowing participants’ antibiotic use throughout their life, linkage is a burden-free and more accurate method of data collection than a questionnaire administered to twins that relies on recall. Linkage will ensure that research findings are more representative of a wider population and so are more generalisable to the wider public. This project will establish a mechanism whereby future follow-up costs could be reduced. It is hoped to add general value to epidemiological research through creating methodological solutions that can be shared with other researchers and applied to other studies • Health records linkage will enhance case identification and track key traits such as body mass index, blood pressure, electronic frailty index, healthcare and medication. All of these in turn is hoped to benefit the public by enabling better health research that is more representative of the population. TwinsUK commits to making every reasonable effort to inform volunteers about the use of their records, respecting objections and continuing to provide fair processing on a regular basis. TwinsUK will mitigate risks and ethical concerns by: only allowing approved King’s College London (KCL) researchers who are substantive employees to access the data for approved project; ensuring that researchers only see pseudonymised data and only accessing variables pertinent to their hypothesis (i.e. minimise the data before providing managed access). When participants sign up to TwinsUK, the Volunteer Information Sheet they are provided with describes how the main benefit of taking part is the contribution towards the advancement of health research. For some of the TwinsUK studies involving a clinical visit, participants will receive a set of clinically useful results for them and their GP, which may help identify health conditions. Linkage in the context of TwinsUK within the Department of Twin Research at KCL. Part of the Department of Twin Research, TwinsUK is the largest longitudinal cohort of community dwelling adult twins in the UK, comprising >14,000 volunteer twins now predominantly in midlife (mean age 59) (Verdi S, et al.). Participants have provided self-reported questionnaires, physical/cognitive measures and biological samples through clinical visits approximately every four years, for researchers to investigate how environmental factors and genetics interact to impact health over the life course. Importantly, longitudinal biobanking of samples over the last 30 years enables research on early detection of disease. TwinsUK participants have, from enrolment, given their informed consent for these data to be used for health research. Linkage to electronic health records will ensure that outcome data is robustly established. As a longitudinal population study and an established research database, TwinsUK secured funding up to December 2022 by public funding from the Wellcome Trust (212904/Z/18/Z). This project is part of the portfolio of research funded by the Wellcome Trust, Medical Research Council, European Union, the National Institute for Health Research (NIHR)-funded BioResource, Clinical Research Facility and Biomedical Research Centre based at Guy’s and St Thomas’ NHS Foundation Trust in partnership with King’s College London. These funding bodies do not have any decision making responsibilities for how the data will be processed or the purposes for which data is used. [1 paragraph unchanged] Data will solely be used to conduct statistical scientific research in line with the GDPR Articles 6 and 9 and will not be used for any other purpose. The linked NHS records will be incorporated into the TwinsUK research database and therefore linked to the other data within the resource. This includes data collected directly from participants (via questionnaires, study administered assessments or assayed from biological samples) and also data linked from other sources. This latter category includes participants education records (sourced from the Department of Education’s National Pupil Database) and spatial datasets recording aspects of the built (e.g. neighbourhood measures of deprivation, service availability) and natural environment (e.g. air pollution, climate records, land use and land cover). Aside from scientific and public good reasons, researchers consider this request reasonable as they will restrict the use of identifiable information to ensure accurate linkage and de-identify data prior to research use. Stringent data processing, analysis and security procedures alongside binding confidentiality agreements ensure patient confidentiality is protected. The Department of Twin Research is committed to support research projects with the aim of: The following shows the datasets requested and the rational for requesting each of them. • Facilitating multidisciplinary research and general access to both phenotypic and genotypic data for a diverse set of clinical traits, through: Cancer registration o Harmonising with other longitudinal population studies with similar purpose Help researchers establish what changes in the serum (metabolomics, proteomics) predicts cancer development 5 years prior to onset. This can be examined within twin pairs to identify factors which are independent of genetics. o Enhancing discoverability of the datasets through NIHR ARCs (National Institute for Health Research - Applied Research Collaboration) Mortality Data • Publishing in peer reviewed journals as well as creating an open forum to interact with the scientific community. Aided the investigation of the biological and epidemiological predictors of death, and cause of death, and the predictors of death given the same genetic background. • Where appropriate, feeding directly to Public Health England, NHSX and the Chief Scientific and Medical Offices and National Statistician on matters of public health. Demographic Data subjects Data was used to maintain the TwinsUK study database in order to update contact details where necessary and to flag deaths to the study (in order to send a condolence letter and to block further insensitive contacts). TwinsUK propose including all twin participants who have not explicitly withdrawn from the study. This will provide an eligible pool of >14,000 individuals. The group includes both male and female participants. The current group of participants are all adults (aged 18 +), have an average age of 59 and the eldest participants are in their 90s. TwinsUK has been set up for more than 25 years and previous linkage to health records datasets has been implemented 10 years ago. The TwinsUK administrative/data linkage team maintain an administrative database of participants who are alive but wish to stay inactive (no involvement in any new studies) or those who have withdrawn of consent, in order to make informed decisions about their involvement in the study. Participants who are alive but inactive (not participated in the studies or questionnaires recently but is still a part of TwinsUK) will be included in the data linkage programme. In line with the principals set out in the Mental Capacity Act 2009 TwinsUK will assume that an individual has the capacity to make a consent decision unless there is information to the contrary. Researchers aim to strike a balance between maximising the opportunity for individuals to participate whilst minimising inappropriate intrusiveness. Following the approval of section 251 support, the fair processing campaign will offer an opportunity for all TwinsUK participants to object if they so wish. TwinsUK have designed a legal basis mechanism with the CAG to run a data linkage fair processing campaign amongst twin participants to collect initial objection in taking part in the data linkage programme. As an exit route, consent will be sought where practicable. The Data Linkage Information Sheet and the Data Linkage Decision Form will be sent to twins to notify them about the data linkage programme and give them an opportunity to opt out. In the future, as twins visit clinic and as they take part in studies, Twins UK will seek consent for data linkage directly and thus decrease reliance on CAG and s251 over time. Everyone for whom Twins UK hold an email or mailing address will be directly contacted, whether or not they are active or inactive. They received slightly different cover letters/emails if they were active or inactive. The Data Linkage Information Leaflet is also publicly available on our website. TwinsUK also featured the data linkage fair processing campaign in an e-newsletter, which is also publicly available on the website. Participation in data linkage is entirely optional, and twins may opt out of any or all data linkage without affecting their membership and participation with TwinsUK. Twins UK also included on the Data Linkage Decision Form the following narrative , to reassure twins: 'Your decision will not affect your membership status with TwinsUK.' Those that will be excluded from the data linkage programme are individuals who have withdrawn consent, those who are not part of the consented cohort those who will opt-out from the fair processing campaign launched with the section 251 support and those who do not consent to have their health records use for health research (via direct consent used as an exit route to section 251 support). The purpose of the request The purpose of this is to conduct research within the programmes of activity in line with the fair processing information to be provided to participants: • Investigate how environmental factors and genetics interact to impact health and disease over the life course. o Environmental factors on which TwinsUK have extensive longitudinal data stem from proximal factors such as diet, to more distal factors such as pollution (directly and indirectly measured), water quality and the built environment. • Identify predictors of physiology and disease, especially using new technologies; o for example, immunophenotyping, metabolomics, epigenomics, proteomics and genomics from >25 years of stored data and samples. • Identify new targets for intervention strategies, o gene targets, nutritional or wider environmental interventions • Test non-pharmacological personal interventions on selected twins. This request aims to gain accurate health and social care data on participants of the TwinsUK in order to understand ill health and disease, and contribute to the development of prevention and management strategies. In addition to being used to identify new associations, researchers will use the TwinsUK research database to replicate findings from other longitudinal studies or use the data in conjunction with similar data from other longitudinal studies, in order to determine if the findings are robust and can be generalised. The data linkage programme – in this application - seeks approval to link to, extract and use NHS Digital data (Cancer registration, Mortality Data, Hospital Episode Statistics - HES (Admitted Patient Care , Accident and Emergency, Critical Care, Outpatients), Mental Health Services Data Set- MHMDS and Personal Demographics Service-PDS) for the TwinsUK participants. Only King’s College London approved bona-fide researchers (who are KCL employees, KCL PhD students and PhD Students having signed the IP Assignment Waiver will have the opportunity to use the TwinsUK research database in conjunction with NHS Digital data. Only these researchers will have access to the secure platform and analysis tools to work on the linked data and can only take aggregate results out of the system. The linked health record data will solely be used to conduct statistical scientific research in line with the GDPR Articles 6(1)(e) and 9(2)(j) and will not be used for any other purpose. Projects needing access to data for further analysis will require the submission of a Department of Twin Research (DTR) Data Access Request form and the proposal will be reviewed by the TwinsUK Data Access Committee. The information from the flagging and tracing reports will be used to maintain the TwinsUK study database in order to update contact details where necessary and to flag deaths to the study. Access to contact information is restricted to study staff although derived data (e.g. air pollution estimates based on residential address) are released to researchers once they are processed to remove direct identifiers. As per TwinsUK lifetime study-based data, TwinsUK are requesting access to all the life-course data from birth up to the most recent finalised datasets. It is hoped to allow the assessment of changing health status over time, changing severity of health status and the precursor health events leading up to health outcomes. TwinsUK require detailed information about these in order to build event sequence records alongside study collected information, linked health and social care records from NHS Digital and linked records from other providers (e.g. national pupil database records). The data request is limited to the TwinsUK cohort participants. TwinsUK participants mainly live in England, Wales, Scotland and Northern Ireland. Therefore, the present application focuses on the territory covered by England/Wales. A minimal number of identifiable data (personal identifiers) are requested for each of the datasets to obtain accurate “linkage data”. For historic MRIS reports previously received from NHS Digital, KCL will undertake a review of the provided data fields and determine whether these are still required or whether it would be appropriate to minimise the data held by destroying some of these fields. Once received and prior to storing the linkage data in the Data Safe Haven, personal identifiers, if included, will be removed from the research data prior to the pseudonymisation process. TwinsUK has been set up for 30 years and previous linkage to health records datasets has been implemented 10 years ago. Considering that TwinsUK will keep recruiting new participants every year and that resources will be in place to maintain the resource. The healthcare records requested are minimised to TwinsUK participants and those that stay within the campaign after fair processing. They are also minimised to health datasets relevant to the research programme objectives and redundant variables are excluded. Unless further consent is obtained from participants, the requested data will be aligned with HRA CAG minimisation requirements for variables of elevated sensitivity. The level of data that has been requested is paramount to the dataset. The data made available to individual researchers are minimised further through the removal of all personal identifiers and the deployment of sufficient technological and information security controls to render the data effectively pseudonymised. TwinsUK will exclude the records of individuals who object to the research use of NHS records or those who have withdrawn from the study. The TwinsUK research data made available to researchers does not contain any personal identifiers, as identifiers are held within a separate secure database within KCL. A minimal number of personal identifiers are initially used to obtain accurate “linkage data” from providers. Once received and prior to finally storing the linkage data in the Data Safe Haven, personal identifiers, if included, are removed as part of the pseudonymisation process. KCL approved researchers are only allowed to use the data items pertinent to the specific hypotheses they are investigating using the TwinsUK Data Access approval process- only variables relevant to the specific research project will be given. This will all occur within the Data Safe Haven where specific IDs/ credentials are necessary to prevent user linking and or re-purposing data. TwinsUK’s data collection strategy is to collect data on the participants from a range of sources, including self-reported, clinic visits and from official records. By triangulating this blend of self-reported and objective measures from diverse sources TwinsUK can improve the quality of the resource and ensure TwinsUK provides its researchers with the data quality and validated information they require for rigorous peer reviewed science. [1 paragraph unchanged] The sole Data Controller is King’s College London who also process the data. • Avon Longitudinal Study of Parents and Children (ALSPAC): ALSPAC (University of Bristol) is an established population cohort based in Bristol. Over the past 10 years, ALSPAC have pioneered the development of data linkage activities in England through PEARL – The Project to Enhance ALSPAC through Record Linkage. TwinsUK is partnering up with ALSAPC to facilitate and harmonise data linkage activities, with ALSPAC providing guidance to KCL on the data linkage process but without ALSPAC actually accessing any NHS Digital data involved in the linkage. Data under this Agreement will not be linked with ALSPAC Data. • Department for Education (DfE): A data request application will be made to the Office of National Statistic (ONS), the official data processor of the dataset in order to include the National Pupil Database (NPD) in the data linkage programme. • NHS Wales Informatics Service: A data application will be made to this organisation in order to include the Patient Episodes Datasets Wales (PEDW) in the data linkage programme. • Egton Medical Information System Limited (EMIS): EMIS is a GP software provider and therefore a data processor of the primary health records controlled by General practitioners (GP). GP and software companies will be approached in order to include the GP primary health records in the data linkage programme. • The Phoenix Partnership (TPP): TPP is a GP software provider and therefore a data processor of the primary health records controlled by General practitioners. GP and software companies will be approached in order to include the GP primary health records in the data linkage programme. None of the organisations listed above are intended to have access to the data provided by NHS Digital, but they may provide data to King's College London which could then be linked with the NHS Digital data.

Processing activities

TwinsUK systems and policies are bound by the study's TwinsUK's commitment to comply with information governance standards included in the NHS Digital [36 words unchanged] organisational structure and methodologies that provide a means to meet these requirements. [1 paragraph unchanged] The Administrative database contains identifiable information about the participants and is kept segregated from any other database within the KCL firewall. Logged and role-based access is allowed for authorised personnel only. In order to gain access to health-related data, a very small but identifiable set of data has to be divulged to NHS Digital. Participants’ Name, Date of Birth (DOB), Gender and Address are provided to the NHS Digital Personal Demographics Service (PDS) and the NHS ID, DOB and General Practices registration details are received to upscale the data held within the TwinsUK Administrative Database. NHS ID and DOB of participants are provided to NHS Digital and a number of health-related datasets are received into the TwinsUK Data Safe Haven. Typically, these are yearly tasks and the encrypted transfer of these personal identifiers uses high standard security protocols secure sockets layer (SSL)/Transport layer security (TLS) and AES 256 encryption. Using these protocols ensures the security of the end-to-end internet connection safeguarding any sensitive data that is passed between systems. The Administrative database contains identifiable information about the participants and is kept segregated from any other database within the KCL firewall. Logged and role-based access is allowed for authorised personnel only. In order to gain access to health-related data, a very small but identifiable set of data has to be divulged to NHS Digital. Participants’ Name, Date of Birth (DOB), Gender and Address are provided to the NHS Digital Personal Demographics Service (PDS) and the NHS ID, DOB and General Practices registration details are received to upscale the data held within the TwinsUK Administrative Database. NHS ID and DOB of participants are provided to NHS Digital and a number of health-related datasets are received into the TwinsUK Data Safe Haven. Typically, these are yearly tasks and the encrypted transfer of these personal identifiers uses high standard security protocols secure sockets layer (SSL)/Transport layer security (TLS) and AES 256 encryption. Using these protocols ensures the security of the end-to-end internet connection safeguarding any sensitive data that is passed between systems. [1 paragraph unchanged] Personal identifiers received by NHS Digital and the PDS will be limited [18 words unchanged] confirm or upscale the TwinsUK Administrative Data, NHS Digital will feed Cancer data data, mortality data, Hospital Episodes Statistics and mortality data Mental Health Services Data Set into the TwinsUK Data Safe Haven. These identifiable data are transferred in a secure data tunnel as above. [3 paragraphs unchanged] The use of identifiable NHS Digital data in TwinsUK research requires either [12 words unchanged] of the NHS 2006 Act. The consent sought by TwinsUK under which researchers collect administrative and research data is collected are based on the assurance that data are de-identified as soon as [18 words unchanged] Data Safe Haven provides the business processes required to meet these commitments. TwinsUK conform to the relevant best practice guidelines and are compliant with the NHS Data Security and Protection Toolkit (DSPT) standard. Access to the Safe Haven and the data (identifiable data and the linkage) within it is restricted to registered, authorised users. Access to the Safe Haven is password controlled by individual user accounts and passwords. Accounts and passwords are not shared. Password management conforms to best practice, including password strength and refresh frequency. Paper-based data and information handling and paper storage are operated in secured areas with limited access. Where possible the work is organised so that there is clear segregation between staff members undertaking activities with identifiable data and staff members undertaking activities with non-identifiable data. Once the data is received in the Safe Haven, it will go through the processes of quality control, pseudonymisation and harmonisation with the rest of the longitudinal data. In order to keep the data clean, accurate and relevant, there is an element of data feedback in the Safe Haven process. The Linkage Team will manage the acquisition of the aforementioned data and maintain these for research. They will collect, store, process and describe these data in line with any relevant legislation, ethical requirements, and Data Sharing Agreements. A crucial function of these criteria is to subject these data to a robust pseudonymisation process prior to their processing and usage by TwinsUK and KCL approved researchers. Once the data is received in the Safe Haven, it will go through the processes of quality control, pseudonymisation and harmonisation with the rest of the longitudinal data. Following the de-identification process using TwinsUK pseudo identifiers, each set of linked data received will be held in an individual silo called a “Data Linkage File”. These files will be held in the TwinsUK Data Safe Haven and whilst there, they will not be linked together. Over time, as TwinsUK receive more datasets, the number of Data Linkage Files for each category will grow. An overarching database will be used for the purpose of locating, versioning, usage tracking and the storage of various metadata associated with each dataset. This central reference database will not store actual data. Rather it is there to track the Data Linkage Files to ensure their validity, integrity, relevance and security. TwinsUK Data Repository TwinsUK collects the bulk of its data from a range of sources including, directly from study participants (clinical test, biological samples, and questionnaire data) and indirectly via analysed datasets through collaborations with other groups. In line with the security and confidentiality commitments given to cohort participants and King’s College London (KCL), the collection and use of these records meet the requirements of legislation (UK & EU), ethical oversight (KCL and those specified by the data source organisations), Data Sharing Agreements (between TwinsUK and the data source organisations) and the confidentiality safeguard commitments that TwinsUK have given the study participants via informed consent. These phenotypic, genetic or omics (i.e. metabolomics, proteomics and genomics) data are held within the TwinsUK data repository on encrypted servers, in password protected databases and data files accessible by authorised TwinsUK data personnel only. KCL Secure Research Platform Once all data cleaning processes have been completed, the “linkage” data files are ready to be combined with TwinsUK data from the repository upon request by approved KCL researchers with a clear and justifiable hypothesis. The approval of the data (and linkage) requests will be sought at this stage from the TwinsUK Data Access Committee, TREC. Upon approval of a request, the TwinsUK data team is authorised to use the Linkage Controller Process to link all the requested data and further de-identify them with the use of Public IDs generating pseudonymised project specific datasets to be used in the KCL Secure Research Platform by individual researchers. This process is described in the TwinsUK Biobank Volunteer Information Sheet and the TwinsUK Data Linkage Information Leaflet. Formally agreeing to all the terms and conditions of the data request form by signing it ensures that the project specific data are not shared by the researcher and are destroyed upon completion of the project. Moreover, the work will be carried out within the KCL platform and data will not be taken out. In effect, each dataset is “effectively anonymised”. The Linkage Controller Process takes de-identified data and creates pseudonymised data for each unique project using “public ID” generated by TwinsUK. - The KCL Secure Research Platform is for use by KCL approved researchers only. It sits within the KCL firewall and is accessible by username and password and logged usage. It provides a “data safe” environment for researchers to analyse the datasets shared with them. Keeping each project’s dataset unique and segregated, the data within them is effectively anonymised. All processing of linkage data is only carried out by employees of TwinsUK at King's College London. At the Department of Twin Research all staff are aware of TwinsUK information security policy, the Data Protection Act 2018 and have annual training in information governance and GDPR provided by King’s College London. Those concerned with data processing also undergo training and certification for Data Security Awareness by the NHS. Systems and processes/protocols follow the NHS Data Security and Protection Toolkit Standards and are subject to annual independent audit. [1 paragraph unchanged] The data is housed in various secure databases on servers that sit within the King’s College London (KCL) KCL network, protected from the outside world by a firewall. TwinsUK never stores [86 words unchanged] All desktops, laptops and servers used in the KCL network are encrypted. Alongside the technical processes TwinsUK have developed rigorous governance procedures, including: • All staff handling any data being aware of their responsibilities, being trained in data protection regulations and regularly undertaking certification tests A Data Access Committee (comprised of senior management, data personnel, principal Investigators, a TwinsUK participant representative, an NIHR (National Institute for Health Research) Biomedical Research centre of Guys & St Thomas NHS Foundation Trust and King’s College London representative) meets weekly to deal with any data issues and risks and to approve any data sharing with other KCL researchers. The data access committee will ensure full transparency with NHS Digital and will not permit any use of data for research that is not in line with the stated purpose of this NHS Digital Data Sharing Agreement. KCL researchers will need to ensure their outputs are direct and tangible with fulfilling NHS Digital criteria when requesting for linked data, and these factors will be scrutinized by TREC in decision making. No onward sharing between differing organisations will be permitted, should such organisations request access for the NHS health linked data KCL hold, TREC will direct them to apply to NHS Digital independently. • A Volunteer Advisory Panel (VAP) (comprised of representative members of the TwinsUK cohort) that meets quarterly, whether face to face or online to discuss and comment on any ethical issues that have been escalated to them. During the ongoing Pandemic VAP meetings are being held using Zoom. The following shows the confidential patient information which will be utilised in the described data linkage activities and the justification for using them is also provided. Name- Linkage purposes: TwinsUK will use “name” for linkage purposes with NHS Digital Personal Demographic System (flagging and tracing). Date of birth- Linkage and analysis purpose: TwinsUK will use “date of birth” for linkage purposes with NHS Digital Personal Demographic System (flagging and tracing). Subsequently TwinsUK will use “date of birth” (needed as a check variable) as one of the linkage identifiers for extraction with NHS Digital. “Year of birth” will be used to calculate the age of the participants for a specific event or sample when analysing data. It can also be used as a confounder in the analysis. Age is particularly relevant for many of the researchers in the department who study ageing-related processes. Date of death- Linkage and analysis purpose: TwinsUK will use “date of death” in order to update the study administrative database and block future contacts to deceased participants, minimising any undue stress to family members. Note that the cause of death isn’t used in the administrative database. “Date of death” will also be used to study mortality. Gender- Linkage and analysis purpose: TwinsUK will use “gender” for linkage to NHS Digital Personal Demographic System (flagging and tracing) and the Department for Education. “Gender” will be used as confounder in data analysis. “Gender” can also be used to study gender specific traits such as menopause. Ethnicity- Analysis purposes: TwinsUK will use “ethnicity” as confounders in data analysis. Address (current/historic)- Linkage purposes: TwinsUK will use “address” (current/historic) for linkage purposes with NHS Digital Personal Demographic System (flagging and tracing). NHS Number- Linkage purposes: TwinsUK will use “NHS ID” as the main linkage identifiers for data extraction from NHS Digital. General Practice Registration Details- Linkage purposes: TwinsUK will ask NHS Digital to provide the details of the general practice at which the participant is registered. TwinsUK will then use “General Practice Registration Details” to seek GP assent to access the primary care records of the participants. All organisations party to this agreement must comply with the data sharing framework contract requirements, including those regarding the use (and purposes of that use) by “personnel” (as defined within the data sharing framework contract i.e. employees, agents and contractors of the data recipient who may have access to that data). There will be no data linkage undertaken with NHS Digital data provided under this agreement that is not already noted in the agreement. The Department of Twin Research at King’s College London currently have 21 PhD students and some of these may be users of the enhanced resource. These students’ applications to use the data will be assessed along with all others and must include a qualified supervisor who takes responsibility for the conduct of the research. The PHD students have an employment contract with Kings College London; any individuals without a signed IP Assignment Waiver will not be permitted access to NHS Digital data. Data will only be accessed by KCL employees, KCL PhD students and Honorary Contract Holders subject to the PhD Students having signed the IP Assignment Waiver and the Honorary Contract Holders having a contract in place. Data will only be processed by KCL employee at TwinsUK Data will not be accessed or processed by any other third parties not mentioned in this agreement.

Expected output

Over the course of the research programme, TwinsUK will continue to engage with the scientific community, the community of twins who participate in [8 words unchanged] has a Communications and Engagement Manager who is dedicated to these responsibilities. Scientific community: Researchers TwinsUK will continue to engage with the scientific community through participation in international [24 words unchanged] media, which further aids dissemination and communication of novel research and findings. As a result of the data processing, TwinsUK will also engage with other longitudinal population cohorts such as Avon Longitudinal Study of Parents and Children (ALSPAC), Twins Early Development Study (TEDS) and Cohort and Longitudinal Studies Enhancement Resources (CLOSER) to harmonise and streamline methodologies to access and process health records. TwinsUK participants: TwinsUK has maintained and encouraged engagement of cohort participants in various ways. Researchers have There is a Communications and Engagement Manager who provides the main link between the TwinsUK research team and cohort participants. TwinsUK have There is a regular newsletter which that is used to share research updates, and updates. TwinsUK also actively disseminate study findings through local and national media. TwinsUK are active on social media and engage the twins particularly through the Facebook page, and also Instagram. TwinsUK have There is a Volunteer Advisory Panel (VAP) made up of 12 TwinsUK participants and an eVAP made up of 17 participants. TwinsUK The VAP meet at least twice a year in person with the VAP and [34 words unchanged] data linkage research programme, and the eVAP too to a lesser extent. Wider public: Staff TwinsUK staff members regularly take part in public engagement activities, including interactive stands at events, collaborations with artists, and talks e.g. in schools. These opportunities allow us TwinsUK to communicate the research and findings to date with a wider audience. These activities are coordinated and [7 words unchanged] TwinsUK also seek national media interest for research papers and findings, and the study is are regularly featured in the news and on TV. Policymakers: TwinsUK already work with policymakers and influence policy through participation in All-Party Parliamentary Groups (APPGs) relevant to the research. research study. For example, TwinsUK are members of the Human Microbiome APPG and linked to the APPG on Longevity. During the COVID-19 pandemic, the work on COVID Symptoms Study, has established close links with Public Health England and NHSX (SPI-M) and have directly fed through findings to the Chief Scientific Officer, Chief Medical Officer, National Statistician and the Scientific Advisory Group for Emergencies (SAGE). These links will remain in place and can be utilised to channel outputs from scientific findings to reach maximum benefit for the public good. The outputs will be an ongoing process, where each year different papers, presentations will be made. TwinsUK are part of Health Data Research UK (HDRUK) which helps to promote findings to the health informatics and clinical community. TwinsUK will contribute case studies of the successful use of linked health records with study data to the Understanding Patient Data taskforce run by the Wellcome Trust and the National Data Guardian. All outputs will be restricted to aggregate data with small numbers suppressed in line with the HES analysis guide. TwinsUK will ensure that the anonymity of research participants will be maintained when disseminating research findings or engaging with different level of stakeholders via different mediums. All peer-reviewed publications will be submitted to TREC for approval prior to dissemination. This approval involves further assessment of disclosure risk (for example risk related to small cell counts of sensitive variables) and where necessary an obligation for researchers to revise their outputs to remove such risk prior to dissemination. No target dates or deliverables were stated in the original Data Sharing Agreement valid up to the 20th July 2020. Dissemination of data in the public domain will usually involve aggregated data from analysis results. There have been >1,000 articles published using TwinsUK data; details of these can be found on the TwinsUK study website (https://twinsuk.ac.uk/our-research/publications/). Cancer and mortality data provided under the Data Sharing Agreement in place for the past 10 years have been used in a wide range of studies and publications in line with the original purpose. The purpose was to: • To assess the extent to which age-related deterioration is correlated between different organ systems (cardiovascular, muscle, bone, respiratory function, and vision). • To assess how much of variation in longitudinal rates of physiological deterioration is due to genetic and environmental variation. • To investigate genetic associations with the five organ systems using candidate gene and genome wide association studies • To determine the relative influences on biological ageing (as measured by loss of function/tissue) of environmental factors such as marital status, socio-economic status (income, education, occupation), levels of physical exercise, smoking and alcohol intake, number of children. • To investigate the value of putative biomarkers of ageing: serum vitamin D, DHEAS, C-reactive protein, creatinine, retinal vascular calibre, and white cell telomere length Details on death certification and cancer registration have been utilised in 24 original scientific publications by members of the department. For example, these data were central to understanding the role of specific genes and genomic alterations (such as epigenetic changes and telomere length) in the development of many cancer types; identifying specific metabolite patterns associating with longevity and mortality (including biomarkers of ageing); identifying for the first time, changes in the gut microbiome associated with common cancers; and validating the prediction power of the models of frailty, which have been used in many follow-on studies highly relevant to TwinsUK ageing population. Much of the research is early in the translational pathway, contributing to understanding of mechanisms of disease which will ultimately influence the NHS by enabling earlier detection of biological ageing and disease, and intervention strategies. This will benefit twins and their healthcare in the years to come. As aforementioned it is also beneficial to the twins as linkage is a non-intrusive way for twins to remain a part of the cohort. Linkage with health records data in the future will augment active COVID-19 research, and help patients and participants through enabling Twins UK to accurately identify biological markers which predict COVID-19 outcomes. These could be used for preventative strategies in the future. The outputs will be an ongoing process, where each year different papers, presentations will be made. Twins UK is an ongoing programme and new funding will be sought to continue the cohort past 2023. All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide.

Expected measurable benefits

An example of benefits accrued from the existing linkage program has included identification of clinical and genomic risk factors for the development of cancer, including breast, prostate and skin cancer. The study-based TwinsUK data has produced significant impact for the NHS through contribution to identification of genes, and genomic alterations in almost every health trait, as well as identifying proteins, metabolites and microbes influencing health. More recently the team have been able to identify personalised nutritional targets which are key in the development of metabolic syndrome. TwinsUK have continually collected extensive physiological and multi-omic data on participants over the last 30 years. Available multi-omic data includes genome-wide profiles of genetic variants, gene expression and epigenetics in multiple disease-relevant tissues, as well as large-scale profiling of the immune system, metabolites, protein levels and the microbiome. The TwinsUK records linkage programme will benefit the Health and Social Care system by linking health records data with this extensive longitudinal data in TwinsUK, thereby enabling researchers to identify changes in the TwinsUK longitudinal data that precede or follow development of disease. It is hoped to enhance researchers’ ability to predict or diagnose illness and to identify which physiological systems influence disease development, leading to new ways to improve care. The information can be used to help understand more about disease risks and causes, improve diagnoses, develop new treatments and prevention strategies. In the future the future, it is hoped that NHS will benefit from this our research through a wide range of routes, including early identification of potential [40 words unchanged] outcomes through triangulation of Health Records data with direct clinical measures and self-reported self- reported data. Examples of research projects this programme will enable are: • Explore how medication history relates to omics profiles, including blood-based biomarkers of disease (e.g., epigenetics) • Explore specifically antibiotic usage to microbiome markers of antibiotic resistance • Identification of novel genomic and metabolic biomarkers for early-stage breast cancer detection using 20-year biobank in matched twin pairs. • Identification of Immune-metabolic signatures of early inflammation as biomarkers and therapeutic targets for inflammatory disease and cancer outcome. • Pharmacometabolomics study of drug concentration and detection in different fluids to determine individual response to prescription medication and impact on treatment outcomes. • Immune response and clinical outcomes to viral infections and vaccinations • Understanding the interaction between multi-morbidity and dementia, in particular investigating how infection and inflammation alter the course of disease • Multi-omic skin tissue biomarkers underlying longitudinal trajectory of melanoma risk progression • Identification of longitudinal molecular changes in blood, fat and skin tissue prior to development of cardiovascular disease in order to identify key molecular changes that precede clinical presentation. • Explore cancer phenotypes such as breast cancer screening and colonoscopy for frequency and results. • Investigate the granularity of imaging biomarkers, which relate to future cancer diagnoses • Explore patterns in the frequency and effectiveness of cancer treatments including chemotherapy & radiotherapy This data linkage programme will also establish a mechanism whereby future follow- up costs will be reduced. It is hoped to add general value to epidemiological research through creating methodological solutions that can be shared with other researchers and applied to other studies. Health records linkage will enhance case identification and track key traits such as body mass index, blood pressure, electronic frailty Index, healthcare and medication. All of this in turn will benefit the public by enabling better health research, such as those mentioned above, that is more representative of the public. A current limitation of UK-wide volunteer studies such as TwinsUK is that volunteer’s ability to attend our central clinic for physiological measures decreases as they become older and more frail. This leads to loss of data at critical points in an individual's healths pan. Linking electronic health records will allow the study to continue collecting critical endpoint data on individuals that are no longer physically or mentally willing to travel. It is hoped to also allow more inclusive retention of volunteers that find travel challenging, due to caring responsibilities, distance, disability or other reasons. TwinUK studies using linked data are hoped to benefits twins and their healthcare in the years to come. As aforementioned it is also beneficial to the twins as linkage is a non-intrusive way for twins to remain a part of the cohort. Linkage with health records data in the future will augment active COVID-19 research, and help patients and participants through enabling Twins UK to accurately identify biological markers which predict COVID-19 outcomes. These could be used for preventative strategies in the future. The drivers of most age-related diseases are a complex mix of genetic predisposition and environment. Comparing environmental exposure of identical twins, who share the same genome, but develop different diseases is a powerful strategy to identify environmental risk factors for disease development. In this twin cohort, health record data will be used to identify identical twin pairs who have been diagnosed with different diseases, thereby enabling research to identify whether different environmental exposures across the lifespan contributes to differential disease development independent of genetic factors. TwinsUK aims to maintain its position as a leader in the UK medical and health- science community, as well as within KCL. This will lead to further develop collaborations within KCL and other longitudinal population studies.

Benefits reported

There have been >1,000 articles published using TwinsUK data; details of these can be found on the TwinsUK study website (https://twinsuk.ac.uk/our-research/publications/). An example of past use of NHS benefits accrued from the existing linkage program has included identification of clinical and genomic risk factors for the development of cancer, including breast, prostate and skin cancer. The study based TwinsUK data has produced significant impact for the NHS through contribution to identification of genes, and genomic alterations in almost every health trait, as well as identifying proteins, metabolites and microbes influencing health. More recently TwinsUK have been able to identify personalised nutritional targets which are key in the development of metabolic syndrome. Cancer and mortality data provided under the Data Sharing Agreement in place for the past 10 years have been used in a wide range of studies and publications in line with the original purpose. Details on death certification and cancer registration have been utilised in 24 original scientific publications by members of the department. For example, these data were central to understanding the role of specific genes and genomic alterations (such as epigenetic changes and telomere length) in the development of many cancer types; identifying specific metabolite patterns associating with longevity and mortality (including biomarkers of ageing); identifying for the first time, changes in the gut microbiome associated with common cancers; and validating the prediction power of the study models of frailty, which have been used in many follow-on studies highly relevant to the ageing population. Much of this research is early in the translational pathway, contributing to understanding of mechanisms of disease which will ultimately influence the NHS by enabling earlier detection of biological ageing and disease, and intervention strategies.

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

Datasets: − MRIS - Personal Demographics Service; − MRIS - Scottish NHS / Registration

Objective for processing

Genetic Longitudinal Study of Ageing  1. To assess the extent to which age-related deterioration is correlated between different organ systems (cardiovascular, muscle, bone,  respiratory function, and vision).2. To assess how much of variation in longitudinal rates of physiological deterioration is due to genetic and environmental variation. 3. To investigate genetic associations with the five organ systems using candidate gene and genome wide association studies4. To determine the relative influences on biological ageing (as measured by loss of function/tissue) of environmental factors such as marital status, socio-economic status (income, education, occupation), levels of physical exercise, smoking and alcohol intake, number of children.5. To investigate the value of putative biomarkers of ageing: serum vitamin D, DHEAS, C-reactive protein, creatinine, retinal vascular calibre, and white cell telomere length Longitudinal cohorts like TwinsUK are an important source of information on life course on health and social development. TwinsUK is a very productive resource providing insight into many health questions, with over 1,000 research publications to date, producing insight particularly on the biology of ageing for public benefit. The purpose of the original study was to: • To assess the extent to which age-related deterioration is correlated between different organ systems (cardiovascular, muscle, bone, respiratory function, and vision). • To assess how much of variation in longitudinal rates of physiological deterioration is due to genetic and environmental variation. • To investigate genetic associations with the five organ systems using candidate gene and genome wide association studies • To determine the relative influences on biological ageing (as measured by loss of function/tissue) of environmental factors such as marital status, socio-economic status (income, education, occupation), levels of physical exercise, smoking and alcohol intake, number of children. • To investigate the value of putative biomarkers of ageing: serum vitamin D, DHEAS, C-reactive protein, creatinine, retinal vascular calibre, and white cell telomere length To achieve this, the following data was requested: Cancer registration, Mortality Data and Demographics data. Cancer registry data provided details on exact diagnosis and treatment given, which has been and will continue to be invaluable to understanding the biology of cancer, and long-term effects of treatment. The longitudinal nature of the study samples data means researchers can look for predictors and effects both before and after diagnosis. Mortality data provided hard endpoints, including lifespan and cause of death. Part of the Department of Twin Research, TwinsUK is the largest longitudinal cohort of community dwelling adult twins in the UK, comprising >14,000 volunteer twins now predominantly in midlife (mean age 59) (Verdi S, et al.). Participants have provided self-reported questionnaires, physical/cognitive measures and biological samples through clinical visits approximately every four years, for researchers to investigate how environmental factors and genetics interact to impact health over the life course. Importantly, longitudinal biobanking of samples over the last 28 years enables research on early detection of disease. As a longitudinal population study and an established research database, TwinsUK secured funding up to December 2022 by public funding from the Wellcome Trust (212904/Z/18/Z). The proposal funded by the Wellcome Trust has been approved following an open competitive funding process including scientific peer review. This project is part of the portfolio of research funded by the Wellcome Trust, Medical Research Council, European Union, the National Institute for Health Research (NIHR)-funded BioResource, Clinical Research Facility and Biomedical Research Centre based at Guy’s and St Thomas’ NHS Foundation Trust in partnership with King’s College London. TwinsUK has no envisaged end date and the maximum value of the resource will come from life-long follow-up of the cohort participants. The TwinsUK resource is sponsored by King’s College London who are the custodians of the resource, currently under the governance of the TwinsUK Resource Executive Committee (TREC), with the data linkage managed by Deputy Director. Data will solely be used to conduct statistical scientific research in line with the GDPR Articles 6 and 9 and will not be used for any other purpose. Aside from scientific and public good reasons, researchers consider this request reasonable as they will restrict the use of identifiable information to ensure accurate linkage and de-identify data prior to research use. Stringent data processing, analysis and security procedures alongside binding confidentiality agreements ensure patient confidentiality is protected. The following shows the datasets requested and the rational for requesting each of them. Cancer registration Help researchers establish what changes in the serum (metabolomics, proteomics) predicts cancer development 5 years prior to onset. This can be examined within twin pairs to identify factors which are independent of genetics. Mortality Data Aided the investigation of the biological and epidemiological predictors of death, and cause of death, and the predictors of death given the same genetic background. Demographic Data Data was used to maintain the TwinsUK study database in order to update contact details where necessary and to flag deaths to the study (in order to send a condolence letter and to block further insensitive contacts). TwinsUK has been set up for more than 25 years and previous linkage to health records datasets has been implemented 10 years ago. Researchers aim to strike a balance between maximising the opportunity for individuals to participate whilst minimising inappropriate intrusiveness. Following the approval of section 251 support, the fair processing campaign will offer an opportunity for all TwinsUK participants to object if they so wish. Organisations involved The sole Data Controller is King’s College London who also process the data.

Processing activities

Not stated in the previous version; added here.

TwinsUK systems and policies are bound by the study's commitment to comply with information governance standards included in the NHS Digital DSP Toolkit 2019. To comply with these, TwinsUK are required to put in place measures to ensure the appropriate receipt and processing of confidential data which contains personal identifiers. The term “data safe haven” defines an organisational structure and methodologies that provide a means to meet these requirements.

TwinsUK Administrative Database

The Administrative database contains identifiable information about the participants and is kept segregated from any other database within the KCL firewall. Logged and role-based access is allowed for authorised personnel only. In order to gain access to health-related data, a very small but identifiable set of data has to be divulged to NHS Digital. Participants’ Name, Date of Birth (DOB), Gender and Address are provided to the NHS Digital Personal Demographics Service (PDS) and the NHS ID, DOB and General Practices registration details are received to upscale the data held within the TwinsUK Administrative Database. NHS ID and DOB of participants are provided to NHS Digital and a number of health-related datasets are received into the TwinsUK Data Safe Haven. Typically, these are yearly tasks and the encrypted transfer of these personal identifiers uses high standard security protocols secure sockets layer (SSL)/Transport layer security (TLS) and AES 256 encryption. Using these protocols ensures the security of the end-to-end internet connection safeguarding any sensitive data that is passed between systems.

Linkage Data Provider (NHS Digital)

Personal identifiers received by NHS Digital and the PDS will be limited to the minimum requirements to fulfil the particular request. Whilst the NHS Digital PDS will be used to confirm or upscale the TwinsUK Administrative Data, NHS Digital will feed Cancer data and mortality data into the TwinsUK Data Safe Haven. These identifiable data are transferred in a secure data tunnel as above.

TwinsUK Data Safe Haven

The linkage data, when procured from NHS Digital, arrive at the TwinsUK Data Safe Haven which provides a secure and segregated area for the safe storage and processing of these sensitive data by authorised personnel only.

Data within the Safe Haven are held on encrypted servers, in password protected databases and data files accessible by authorised TwinsUK data personnel only. The function of the Data Safe Haven is described in detail below. All identifiable information is stripped during this process and pseudonymised linkage data files are created.

The use of identifiable NHS Digital data in TwinsUK research requires either consent from the cohort participant or approval under legislation including Section 251 of the NHS 2006 Act. The consent sought by TwinsUK under which researchers collect administrative and research data are based on the assurance that data are de-identified as soon as possible and that access to identifiable data is strictly controlled and not permissible for research purposes. The TwinsUK Data Safe Haven provides the business processes required to meet these commitments.

Once the data is received in the Safe Haven, it will go through the processes of quality control, pseudonymisation and harmonisation with the rest of the longitudinal data. In order to keep the data clean, accurate and relevant, there is an element of data feedback in the Safe Haven process.

Data Storage

The data is housed in various secure databases on servers that sit within the King’s College London (KCL) network, protected from the outside world by a firewall. TwinsUK never stores NHS (or any other) identifiable data on laptops or desktop computers. The proprietary database called “Phenobase” is a SQL Server database application holding both the data and associated metadata. Data is extracted by the team using this application. The health record linkage data will sit on a database that will not be physically linked with Phenobase, increasing the former’s security. TwinsUK minimises the number of people who have access to identifiable health records to a small team responsible for record linkage and maintaining the TwinsUK Data. All desktops, laptops and servers used in the KCL network are encrypted.

Expected output

Not stated in the previous version; added here.

TwinsUK will continue to engage with the scientific community, the community of twins who participate in the study, policymakers and the wider public. TwinsUK has a Communications and Engagement Manager who is dedicated to these responsibilities.

Scientific community: Researchers will continue to engage with the scientific community through participation in international conferences and collaborative workshops, in addition to the usual channels of peer-reviewed publications and presentations. Many of the researchers have established presences on social media, which further aids dissemination and communication of novel research and findings.

TwinsUK participants: TwinsUK has maintained and encouraged engagement of cohort participants in various ways. Researchers have a Communications and Engagement Manager who provides the main link between the TwinsUK research team and cohort participants. TwinsUK have a regular newsletter which share research updates, and actively disseminate study findings through local and national media. TwinsUK are active on social media and engage the twins particularly through the Facebook page, and also Instagram. TwinsUK have a Volunteer Advisory Panel (VAP) made up of 12 TwinsUK participants and an eVAP made up of 17 participants. TwinsUK meet at least twice a year in person with the VAP and consult through email both the VAP and eVAP throughout the year. The VAP and eVAP advise on study direction, priorities and assessment methods. The VAP have been heavily involved in the development of the data linkage research programme, and the eVAP too to a lesser extent.

Wider public: Staff members regularly take part in public engagement activities, including interactive stands at events, collaborations with artists, and talks e.g. in schools. These opportunities allow us to communicate the research and findings to date with a wider audience. These activities are coordinated and organised by the Communications and Engagement Manager. TwinsUK also seek national media interest for research papers and findings, and the study is regularly featured in the news and on TV.

Policymakers: TwinsUK already work with policymakers and influence policy through participation in All-Party Parliamentary Groups (APPGs) relevant to the research. For example, TwinsUK are members of the Human Microbiome APPG and linked to the APPG on Longevity.

The outputs will be an ongoing process, where each year different papers, presentations will be made.

All outputs will be restricted to aggregate data with small numbers suppressed in line with the HES analysis guide.

No target dates or deliverables were stated in the original Data Sharing Agreement valid up to the 20th July 2020.

Expected measurable benefits

Not stated in the previous version; added here.

An example of benefits accrued from the existing linkage program has included identification of clinical and genomic risk factors for the development of cancer, including breast, prostate and skin cancer. The study-based TwinsUK data has produced significant impact for the NHS through contribution to identification of genes, and genomic alterations in almost every health trait, as well as identifying proteins, metabolites and microbes influencing health. More recently the team have been able to identify personalised nutritional targets which are key in the development of metabolic syndrome.

In the future the NHS will benefit from this research through a wide range of routes, including early identification of potential drug targets, nutritional interventions, as well as critical data to identify environmental structures and pollutants influencing disease while controlling for the influence of genetic and early life factors. All these research streams will be significantly enhanced by better ascertainment of outcomes through triangulation of Health Records data with direct clinical measures and self-reported data.

Benefits reported

Yielded Benefits is not a requirement for new applications. There have been >1,000 articles published using TwinsUK data; details of these can be found on the TwinsUK study website (https://twinsuk.ac.uk/our-research/publications/). Cancer and mortality data provided under the Data Sharing Agreement in place for the past 10 years have been used in a wide range of studies and publications in line with the original purpose. Details on death certification and cancer registration have been utilised in 24 original scientific publications by members of the department. For example, these data were central to understanding the role of specific genes and genomic alterations (such as epigenetic changes and telomere length) in the development of many cancer types; identifying specific metabolite patterns associating with longevity and mortality (including biomarkers of ageing); identifying for the first time, changes in the gut microbiome associated with common cancers; and validating the prediction power of the study models of frailty, which have been used in many follow-on studies highly relevant to the ageing population. Much of this research is early in the translational pathway, contributing to understanding of mechanisms of disease which will ultimately influence the NHS by enabling earlier detection of biological ageing and disease, and intervention strategies.