Education and Child Health Insights from Linked Data (The ECHILD Research Database)

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

Datasets: + HES-ID to MPS-ID HES Admitted Patient Care

Objective for processing

BACKGROUND: Under the previous iteration of this agreement (v3), UCL have requested to convert and extend the existing Education and Child Health Insight Linked Data (ECHILD) Database used in the University College London (UCL) study 'Assessing the impact of the COVID-19 pandemic on vulnerable children’ to a Research Database for wider use, through a sub-licencing model. ECHILD includes linked data from health, education and children’s social care and this linkage is not currently supplied by NHS England directly as it combines data from different sectors. ECHILD is currently supported by funding from the National Institute of Health Research (NIHR) Policy Research Unit for Children, Young People and Families (CPRU), within University College London (UCL). CPRU works closely with the Department of Health and Social Care to determine priorities and provide evidence directly to the Secretary of State for Health, government departments and arms-length bodies, such as NHS England, the Office for Health Improvement and Disparities, and the UK Health Security Agency. Additional funding support is provided by Administrative Data Research UK, Health Data Research UK and NIHR. UCL have requested to sublicense the ECHILD data to accredited researchers. UCL remain the sole Data Controller who also process data. The London School of Hygiene and Tropical Medicine (LSHTM), the Office for National Statistics (ONS) and The Institute for Fiscal Studies (IFS) are also listed as data processors and will continue to process the data under this agreement. All access to ECHILD data is via the ONS Secure Research Service (SRS). UCL has been using ECHILD to understand the impact of COVID-19 and lockdown on children and individuals under the age of 35 years at the time of the COVID-19 pandemic. Children or individuals who are vulnerable due to social welfare or chronic health needs are expected to experience more adverse health and social effects of the COVID-19 lockdown than their peers. Key concerns for services are the effects of household confinement during the COVID-19 lockdown, combined with the limited access to support from health, social care and education services. The data processors have analysed the data to understand what impacts COVID-19 infection and related public health responses (such as lockdown) have had on children and individuals, including those that are deemed vulnerable, to inform strategies for the current wave of infection, and any future waves. The previous iteration of this agreement (v3) was built on UCL’s existing approval for COVID-19 work (DARS-NIC-381972-Q5F0V - approved), which includes data on approximately 14 million children and the previous Education and Child Health Insight Linked Data (ECHILD) project (DARS-NIC-27404-D5Z3F - approved), which used linked education and HES data for four one-year cohorts. The linkage under this application will be extended beyond the current specific purpose relating to COVID-19, to address a range of research questions aiming to generate improvements to the health and social care system, through a better understanding of the relationship between education and long term health outcomes. All children and individuals appearing in HES records from (the latest of) birth or April 1997 onwards and born since 01/09/1984 will be included, to allow researchers to evaluate how exposures in childhood and at school age are related to health outcomes in childhood and adulthood. The Research Database will hold information on approximately 20 million individuals, but researchers will only access minimised extracts required to answer specific research questions. PURPOSE [1 paragraph unchanged] The agreement is to also allow UCL to enhance information in ECHILD by including a pseudonymised mother-baby linkage flag, attached to the relevant pseudonymised HES record for mother or baby. To enable inclusion of maternal characteristics (such as age), and to follow up the health of young people to their 30s, UCL requested to extend the age range to all young people in England born on or after 01.09.1984. The strong interrelationship between health and education services in relation to the health and wellbeing of children is recognised by policy makers, but evidence is lacking on how services complement or compensate for each other and there have been calls for a stronger evidence base to be developed. It is imperative that government and researchers work together to fill this evidence gap – to improve the health, wellbeing, education, and safety of children, young people, and families, particularly the most vulnerable. The ECHILD Research Database will do this by using data to generate a comprehensive view of the journey through childhood to adulthood. This research database will be used to understand how trajectories of health, education and social care vary across children’s lives, and what works to improve the design and delivery of policies and systems which better meet the needs of children and young people. The ECHILD Research Database will fill this gap in evidence by facilitating research that will inform policy-makers and service commissioners about the associations between education risk factors and health outcomes. Extension of the age range and the mother-baby link allows evaluation of exposures and outcomes in early adulthood, including of children who themselves become parents. It enables linkage of adolescent education records to health records up to age 38 years in 2022. Researchers will also examine maternal characteristics (such as maternal age, chronic mental or physical conditions, previous birth), using the pseudonymised mother-child flag to assess the influence of risk factors (e.g., maternal history of social care, special educational needs and disabilities (SEND) support, or school exclusion) on health outcomes in her child. It is hoped findings will be relevant to policies receiving major investment such as ‘Start for Life’. UCL have enhanced information in ECHILD by including a pseudonymised mother-baby link between relevant pseudonymised HES records for mother and baby, to enable inclusion of maternal characteristics (such as age). The current ECHILD research database holds information on approximately 20 million individuals, born since 01 September 1984, and their mothers. Researchers only access minimised data extracts required to answer the specific research questions in their approved projects. [1 paragraph unchanged] UCL require administrative data for all children in England (England only, not including Wales) who appear in the specified national pupil database (NPD) (the NPD does not include all childrens data e.g. it doesnt hold date on children who are home schooled) and Hospital Episodes Statistics (HES) datasets to create longitudinal cohorts of children born on or after 1.9.1984. UCL requested a transfer of identifying variables from NPD datasets to NHS England to enable linkage to HES records. Linkage used names, date of birth and postcodes to link to NHS number. The output is pseudonymised linkage keys. UCL have previously demonstrated high quality linkage of 92% of NPD records to HES for individuals born in 1990/01 (NIC 27404). Researchers will evaluate linkage bias, but anticipate acceptable linkage rates from 1984/5. HES histories will be included for any mother linked through HES to a child with an anonymised Pupil Matching Reference, even if the mother is not included in NPD. ECHILD researchers at UCL have responsibility for the running of the ECHILD Research Database. The ECHILD Research Database will be configured to have two distinct classes of operational areas: AMENDMENT MADE UNDER PREVIOUS VERSION (V3) OF THE AGREEMENT: Under the previous iteration of this agreement (v3) it has been requested to update the ECHILD database and to turn it into a Research Database so that minimised extracts of the data can be sub-licensed to other accredited researchers for research aimed at generating improvements to health and social care services. The Department for Education (DfE) and Department of Health and Social Care (DHSC) also intend to access the data on the ONS SRS through separate data sharing agreements which will be subject to the NHS England Data Access Request Service (DARS) approvals process. DfE and DHSC are applying for access to the ECHILD data for their own specified purposes. The ECHILD Research Database will support a rich portfolio of research projects examining different aspects of child health, whilst ensuring data minimisation suitable for the requirements of specific research questions. In other words, researchers will be able to apply for an extract of ECHILD data and will need to justify the years of the data and the specific data modules that they require in order to answer their research question. For example, a researcher aiming to generate evidence about school attainment following liver transplantation may request HES APC (to identify the cohort of liver transplant patients and a comparator group), and the School Census, Key Stage 2 attainment, and Absences data from NPD (to evaluate outcomes), but would not necessarily require any information on Children’s Social care, Maternity Services data, Community Services data, etc. The study period would also need to be specified, for example HES APC data from 1997-2008 capturing children with liver transplantation would allow follow up at age 11 for children captured in NPD data from 2008-2019. The minimised extract would need to be justified for each project, based on the years of the data and the specific data modules, in a data access application to the ONS Research Accreditation Panel and approved by a UCL Data Access Committee (which will include lay members). All data will be stored and accessed securely on the ONS Secure Research Service. No re-identification of individuals will be allowed, and all outputs will be checked for statistical disclosure control before being released. Creating the Research Database to support a range of research projects is in the interests of data protection, as there will be less need for repeated transfers of personally identifiable information for multiple projects across different institutions. Recognising the value of the linked health and education data that have been linked as part of the existing study, there have been urgent calls to open up the ECHILD resource to more researchers, and for a wider range of research purposes. The strong interrelationship between health and education services in relation to the health and wellbeing of children is recognised by policy makers, but evidence is lacking on how services complement or compensate for each other and there have been calls for a stronger evidence base to be developed. It is imperative that government and researchers work together to fill this evidence gap – to improve the health, wellbeing, education, and safety of children, young people, and families, particularly the most vulnerable. The ECHILD Research Database will do this by using data to generate a comprehensive view of the journey through childhood to adulthood. This research database will be used to understand how trajectories of health, education and social care vary across children’s lives, and what works to improve the design and delivery of policies and systems which better meet the needs of children and young people. The ECHILD Research Database will fill this gap in evidence by facilitating research that will inform policy-makers and service commissioners about the associations between education risk factors and health outcomes. UCL are also requesting linkage to a broader range of health data because there is a wide range of potential research purposes. For example, linkage to mental health records will generate evidence on how schools can promote positive mental health in adolescence; linkage to health visiting activity in the Community Services Dataset will enable evaluation of how different levels of health visiting can improve health outcomes for families in contact with social care services. For the first time in England, the database will allow researchers to investigate long-term outcomes for a wide range of health conditions and treatments during childhood and early adulthood, alongside school attainment, absences, special needs support and exclusions, and social care support. ECHILD could also be used to investigate health benefits (or harms) of education practices. For example, to find out whether providing support for special educational needs for children with chronic health conditions improves health outcomes or reduces use of hospital services. Understanding the relationships between services provided within health and education will form the basis of evidence-based policy making at DHSC (and DfE) to ensure children in all settings, in all areas of the country, are healthy, safe and develop their potential. Linkage to the maternal HES record will allow inclusion of maternal chronic health conditions and demographic risk factors (maternal age, ethnicity, age at first birth) to improve understanding of health conditions within families, to guide healthcare support for families and better understanding of intergenerational adversity. In addition, for young mothers (<27y), it will be possible to assess risk factors recorded from school and social care, such as school attainment, exclusions, and previous care placement to inform early interventions before and during pregnancy. LEGAL BASIS, ETHICS AND RISK OF POTENTIAL HARM: UCL will be sole Data Controller under this agreement. The legal basis for processing personal data for this purpose at UCL falls under Article 6(1)(e) of the General Data Protection Regulations (GDPR), i.e. a task carried out in the public interest. It also falls under Article 9(2)(j), processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes. The data are required for research purposes in the public interest – meeting the conditions in the DPA 2018 Schedule 1 Part 1 (4) – which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data. DPA 2018 Schedule 1 Part 1 (4) - which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data. The Data Protection Act 2018 Schedule 1 Part 1 states that when processing special category data under the condition relating to research, the condition is met when: (a) 'necessary for archiving purposes, scientific or historical research purposes or statistical purposes'. As described above the data is considered necessary for the performance of the task (b) 'carried out in accordance with Article 89(1) of the GDPR'. In accordance with this article, processing is subject to appropriate safeguards. These include: i. The data will be pseudonymised prior to dissemination by NHS England to the data recipient; ii. The data recipient’s technical and organisational measures to safeguard the data have been assessed and meet NHS England’s acceptance criteria (see sections 2 and 5b of this application for further details); iii. The requested data has been assessed as proportionate to the aim pursued (see section 5a of this application for further details); iv. Controls, data retention and processing activities have been assessed to ensure respect to the essence of the right to data protection (see sections 5a, 5b and 8a of this application for further details); (c) 'is in the public interest'. NHS England is content that the information set out in the Benefits section of the application evidences that the data processing will be in the public interest. The processing of data for this study is a task of public interest as it will provide evidence on the relationship between health and education outcomes and use of healthcare services among children and individuals. This will benefit the provision of healthcare services by informing policy makers and service providers, and improving evidence-based information and interventions for children and their families. The case for this research being in the public interest is established through balancing the strong policy driver of improved understanding of the inter-relationships between education, health and social care, with consideration of potential harms to the patients/participants whose records are involved. Potential harms primarily relate to breach of confidentiality and the subsequent misuse of Personal Data or erosion of trust in longitudinal research/data science. The risk of this harm is minimised as this is: 1. an observational population-level research database which will not result in a direct intervention to any participant; 2. tried and tested IT infrastructure and governance frameworks specifically designed to minimise risks to privacy during health data science; 3. a fully de-identified research environment with sufficient controls that the risk of disclosure is not considered reasonably likely; 4. National Opt-Out of the use of health data for research will also be respected 5. all staff and users are vetted and approved professional researchers operating within controlled and auditable conditions. This research has strong scientific rationale, the process is transparent, best endeavours will be made to inform participants of the use of their data within ECHILD, with a right to object, that risks are mitigated and the participants (and wider public) are likely to directly benefit from the research through improved health care and government policy provision. Participants will be informed about how their data is used through ongoing fair processing communications including details of the process for opting-out through the National Data Opt-Out process. OPERATIONAL MANAGEMENT: UCL is the study sponsor and Data controller. ECHILD researchers at UCL have responsibility for the running of the ECHILD Research Database. The ECHILD Research Database will be configured to have two distinct classes of operational areas: [2 paragraphs unchanged] ONS SRS are data processors who provide data infrastructure and the Secure Data Environment (SDE). UCL, IFS and LSHTM all access the data for specific research purposes under the current agreement. ONS SRS will assist with the data management and processing (under UCL’s direction) and will conduct the output disclosure assessments. UCL, Institute of Fiscal Studies (IS) and London School of Hygiene and Tropical Medicine (LSHTM) all access the data for specific research purposes. ONS SRS are data processors who provide data infrastructure and the Secure Data Environment (SDE). ONS SRS will assist with the data management and processing (under UCL’s direction) and will conduct the output disclosure assessments. Amazon Web Services (AWS) is a processor acting under the instructions of UCL. AWS’ role is limited to secure back-up of data stored in UCL’s Data Safe Haven. UCL uses offsite data centre services provided by VIRTUS data centre. VIRTUS does not have access to the data. UCL uses offsite data centre services provided by VIRTUS data centre. VIRTUS does not have access to the data. SUB-LICENCING SUB-LICENCING: [2 paragraphs unchanged] UCL will be the data controller of the de-identified data collected within the ECHILD Research Database and stored on the ONS SRS, for the purpose of processing it, approving onward research use, and providing managed access to relevant sub-sets of data for purposes relating to generating benefits to the health and social care system (controlled through the Data Sharing Agreement between UCL and NHS England). In this sharing model of the linked data, ONS will be a data processor, as they will be processing the data on behalf of UCL. This includes holding the linked data in a secure environment, extraction of approved data and setting up access systems, and approving statistical outputs, following a statistical disclosure control procedure. The structure will enable UCL to determine the onward processing of the integrated data by reviewing applications from researchers for specific projects (with the ONS Research Accreditation Panel (RAP) and onwardly sharing sub-sets of relevant data within the ONS SRS to approved users (controlled through a Data Access Agreement between UCL and approved researchers’ institutions). NHS England requirement to audit data use is a requirement in the onward sharing contract and remains feasible in the ECHILD sub-license framework. The approved organisations and researchers who are granted an access to the linked data via the ONS SRS, agree to terms and conditions of use, their rights and responsibilities as users of the linked data, as defined by the ONS SRS. In addition to the agreements signed with the ONS SRS, the organisation of the researcher applying to use the linked data will enter into a Licence agreement with UCL. All data will be stored and accessed securely on the ONS Secure Research Service. No re-identification of individuals will be allowed, and all outputs will be checked for statistical disclosure control before being released. In line with this onward sharing model, the data sharing controls in place between NHS England and UCL are replicated between UCL and the other organisations. UCL is fully accountable for the actions of the parties involved in subsequent data share and use. The agreement mirrors the Data Sharing Framework Contract in place between NHS England and UCL. It also requests information about the research proposal, benefits to health and/or social care, organisational security assurance and terms and conditions regarding onward sharing of data, responsibilities and processing activities etc., including a GDPR legal basis. UCL will provide a sub-license to UK organisations undertaking research that will be of benefit to the public in England and Wales (this will be assessed in the project proposal form submitted to the ONS Project Accreditation Service for SRS [PASS] and to UCL). Applicants (potential licensees) will need to show that the provision of the sub licensing will be in the public interest and that the data will be used either (i) for the provision of health care or adult social care; or (ii) for the promotion of health. NHS England will retain the ability to directly audit UCL’s compliance with the outlined and agreed data access arrangements. The project proposal will be assessed to determine the details of the project, the people who will be accessing the data, and what data will be requested. Applicants will need to be accredited researchers or agree to undertake training and become accredited, prior to accessing the data. The anticipated volume / number of licences is 1-2 sub-licences per month, and the potential length of each sub-licence is 2-3 years in length. The territory of use in the sub-licence will be the same or narrower than the territory of use stated in this data sharing agreement, namely the United Kingdom. Additionally, an applicant's organisation will need to provide evidence that they have information governance and security assurances in place. Members of the ECHILD Data Access Committee will review and decide if the evidence provided satisfies the requirements of the ECHILD Data Access Agreement. The committee will comprise of the following members: The approved organisations and researchers who are granted an access to the linked data via the ONS SRS, agree to terms and conditions of use, their rights and responsibilities as users of the linked data, as defined by the ONS SRS. In addition to the agreements signed with the ONS SRS, the organisation of the researcher applying to use the linked data will enter into a Licence agreement with UCL. In the event of the termination or expiry of the Data Sharing Framework Contract between UCL and NHS England, all sub-licenses shall automatically terminate. ORGANISATIONAL AGREEMENTS: UCL will provide a sub-license to UK organisations undertaking research that will be of benefit to the public in England and Wales (this will be assessed in the project proposal form submitted to the ONS RAP and to UCL). Applicants (potential licensees) will need to show that the provision of the sub licensing will be in the public interest and that the data will be used either (i) for the provision of health care or adult social care; or (ii) for the promotion of health. The project proposal will be assessed to determine the details of the project, the people who will be accessing the data, and what data will be requested. Applicants will need to be accredited researchers or agree to undertake training and become accredited, prior to accessing the data. Additionally, an applicant's organisation will need to provide evidence that they have information governance and security assurances in place. Members of the ECHILD Data Access Committee will review and decide if the evidence provided satisfies the requirements of the ECHILD Data Access Agreement. The committee will comprise of the following members: [5 paragraphs unchanged] - Lay members Lay members will review requests to access data. Commercial purposes, e.g. where an applicant intends to allow use of data for purposes such as marketing, sales or insurance or where there may be international transfers of data (potentially including the EEA post Brexit), will not be considered unless there is a strong case to show the public interest. Use of the data for commercial purposes is not permitted. [2 paragraphs unchanged] -An agreement (this DSA) between NHS England and UCL to onwardly share linked ECHILD data under the sub-licensing sub- licensing model, which outlines the terms and conditions of use of the linked [13 words unchanged] actions of the parties involved in subsequent access to the linked data. [5 paragraphs unchanged] SUB LICENCE PURPOSES Analysis of ECHILD data will generate answers to questions about the interactions between education, children’s social care and health which have implications throughout the life course. ECHILD will be used for the following specific research themes. These themes, and the proposed research questions within each theme, have been collated from a number of potential research users, including researchers from: University of Bristol, Institute for Fiscal Studies, Applied Research Collaborations (ARCs) including major National Institute for Health and Care Research (NIHR) research groups working on applied health informatics (ARC West), and maternal and child health (PenARC, Exeter), Economists at the Centre for Health Economic, York University, and health researchers at Cardiff University, Swansea University, Imperial College London, and Kings College London. All data processed under the sub-licence will be completed using the same legal basis as mentioned above, namely GDPR (article 6(1)(e))and GDPR (article 9(2)(j)). The UCL Data Access Agreement will require licensees to provide the Legal Basis of their request to access ECHILD data and therefore ECHILD Data Access Committee will only grant approval to applications from researchers within public bodies who have a legal basis to process data under GDPR. ECHILD will only be used for research that has a clear public benefit in England and Wales relating to the provision of healthcare and to education. Analysis of ECHILD data will generate answers to questions about the interactions between education, children’s social care and health which have implications throughout the life course. ECHILD will be used for the following specific research themes. These themes, and the proposed research questions within each theme, have been collated from a number of potential research users, including researchers from: University of Bristol, Institute for Fiscal Studies, Applied Research Collobarations (ARCs) including major National Institute for Health and Care Research (NIHR) research groups working on applied health informatics (ARC West), and maternal and child health (PenARC, Exeter), Economists at the Centre for Health Economic, York University, and health researchers at Cardiff University, Swansea University, Imperial College London, and Kings College London. [14 paragraphs unchanged] • Does exposure to different aspects of children's social care modify the associations between child childhood adversity and later health and educational outcomes? • How does parental health and contact with social care services affect childhood health and educational outcomes? • What is the impact of the policy changes to Healthy Start vouchers on children's health and developmental outcomes? • What is the association between young people’s relative academic position within their school and the development of mental health problems, spanning from the end of primary school through to early adulthood? 2. INFORMING CHILDREN AND THEIR PARENTS The data will provide useful information for children and their parents, which is relevant to the promotion of their health and wellbeing and relevant to clinical practice. The data will show how school achievement and absence varies between children with and without particular health conditions, across the age range and between areas. Potential research questions: • What are the long-term educational outcomes for children who were born preterm (<37 weeks of gestation)? • What are the long-term health outcomes for care leavers? • What are the long-term health outcomes for children with special educational needs provision? • For young people living with diabetes (compared with peers without diabetes), what are the associations between child health outcomes (hospital admissions) and educational outcomes (absence and achievement) during school ages, and to what extent do these determine subsequent health outcomes after compulsory schooling is complete? 3. INFORMING CLINICAL PRACTICE Understanding variation across England in associations between health and school attainment can be used to inform clinical practice by identifying potentially better practices (e.g. to reduce school absence for children with chronic conditions) that could be adopted more widely. Potential research questions: • What is the best timing of liver transplantation, based on variation in educational outcomes for children with chronic liver disease? • Does treatment for childhood cancer affect children’s function in the longer term, for example, their cognitive ability, risk of mental health conditions, or their fertility? • Do children born to women who were treated for cancer in childhood have increased rates of a congenital malformation? 4. IDENTIFYING GROUPS WHO COULD BENEFIT FROM INTERVENTION The data will facilitate examination of whether indicators at school such as absenteeism and school failure can identify groups of vulnerable children and young people (e.g. those with particular health conditions), who could benefit from proactive or preventive healthcare input that might reduce emergency use of hospital services, and improve health and educational outcomes. Potential research questions: • How many children experiencing child abuse and neglect are known to healthcare services, based on comparing coding in health records with recorded contacts with children’s social services? • What are the long-term health outcomes for mothers who are care leavers (and their children)? • What are health and education outcomes for children born to care-experienced mothers? • What are the trajectories of unaccompanied asylum seekers through the social care and health systems? • Can schools be characterised as more or less stressful, measured through special educational needs support, rates of school absence and exclusion, and stress-related or mental health contacts with health care in adolescence? • Are markers of stressful versus less stressful schools associated with risky behaviour and mental health conditions in adolescence and adulthood? • What are the early health, education and social care patterns of children who develop mental disorders in adulthood? • What proportion of children experiencing school exclusion, repeated absence or off-rolling have underlying chronic mental or physical health conditions? • What are the health outcomes for mothers who are care leavers (and their children)? • What are the trajectories of unaccompanied asylum seekers through the care system? • Do children in contact with social care, or exposed to ACEs, have increased risks of admissions for serious mental disorder in adulthood? • What is the association between different dimensions of childhood adversity and later health and educational outcomes? • How does mental health need and service use amongst Looked After Children and Children In Need compare to that of other children? • Can we model differential trajectories of school performance that indicate high likelihood for adolescent / adult inpatient psychiatric admission? • What are the potential moderating and mediating effects of school performance outcomes for those with pre- existing vulnerabilities (e.g. the potential impact of school absence on recurrent self-harm, eating disorder readmission, young people admitted with psychosis)? • What is the influence of school-level factors on the patterns of health outcomes for young people with mental health conditions? • Can we identify and estimate peer contagion effects on health and educational outcomes? • What maternal health factors are associated with both health (diagnosis, service use patterns, mental health admissions, treatment and treatment response) and educational outcomes? • What proportion of the variation in health outcomes for children with diabetes can be accounted for by families and family structure? 5. UNDERSTANDING THE MOST EFFECTIVE METHODS FOR WORKING WITH LINKED HEALTH AND EDUCATION DATA Working with data on the large number of individuals included in ECHILD is challenging, and becomes more complex as the number of datasets increases. Methods to effectively work with these data need to be developed in order for the full potential of the linked data to be realised. UCL will use the linked data to inform descriptive and documentary analysis of the ECHILD dataset. This will describe the linked population and make clear the ECHILD denominator (the population of children included in ECHILD, whose data researchers are able to analyse, e.g. the representativeness of the population denominator of individuals included in ECHILD, in comparison to the general population in England. For example, ECHILD only includes children who have had a hospital contact within an NHS hospital in England, or who have attended a state school in England. Therefore the ECHILD denominator does not include children who have only ever attended a private school and who have only ever had private healthcare contacts), its characteristics and how this sub-set of the population relates to the wider population (through comparison with aggregated population statistics). This will inform research users and those (e.g. policy makers) seeking to draw inferences from findings. UCL envisage a number of methodological projects to be proposed for ECHILD. Potential research questions: • How can we use sibling control analyses to better understand the relationship between shared familial exposures such as maternal education and deprivation, and health outcomes? • What are the optimal ways of selecting control/comparison groups for population health research? • How can we account for clustering of health conditions within families when evaluating the association between individual-level exposures and health outcomes? • How can causal inference methods be best applied in linked administrative data? • How do we best evaluate and account for linkage quality to ensure that analyses of linked data are unbiased? To enable the analyses to address the research questions outlined above, and other research questions aimed at improving health outcomes, the researchers will link HES data (i.e. HES APC, outpatient, critical care, A&E and ECDS data, MSDS, CSDS, Mental Health, birth registrations/notifications, plus death registration data) to administrative data contained in the datasets collectively supplied within the National Pupil Dataset (NPD), provided by Department for Education (NPD data include information on education, child in need data, the Children Looked After return, and the Individual Learner Record; ILR). These datasets (HES-NPD) will be linked by NHS England for children and individuals in England who were born on or after 01/09/1984 using pseudonymised linkage keys

Processing activities

UCL are requesting that the requested data are transferred to the ONS SRS in order to be merged with the education data from DfE. Merging will be on the basis of a link key that will be generated by NHS England, which will allow the DfE and health records of individuals to be brought together. UCL require administrative data for all children in England (England only, not including Wales) who appear in the national pupil database (NPD) (the NPD does not include all children’s data e.g. it doesn’t hold data on children who are home-schooled) and Hospital Episodes Statistics (HES) datasets to create longitudinal cohorts of children born since 1st September 1984. The researchers have requested the minimum data necessary for their research, which reflects the administrative history of the child/young adult for a subset of the available fields (e.g. the researchers have requested 60% of available inpatient fields, with no sensitive or identifiable fields). Data from the newly requested datasets (MSDS, CSDS, Mental health data) have not been minimised, due to the need to further explore these data in order to understand which variables are useful and sufficiently complete. Once this has been completed, the data will then be minimised, hopefully within 12 months. Data will be restricted to records relating to individuals in England born from 01/09/1984 onwards. Longitudinal data for all children and individuals in England born from 01/09/1984 is justified for the following reasons: The data will be limited to individuals in England born from 1 September 1984 onward and their mothers, as full longitudinal records are essential for identifying early-life factors that influence health outcomes into adulthood, particularly because key conditions may not be consistently recorded at every hospital admission. This cohort enables analyses of how childhood health, education and social care experiences affect later outcomes, including intergenerational impacts using the mother–baby link. National geographical coverage is required because health outcomes and service use vary substantially across regions and over time, especially during events like COVID 19, and finer grained data (at MSOA level) is needed to understand localised effects. Comprehensive coverage of all children is also necessary so researchers can compare vulnerable groups—such as care leavers or preterm children—with appropriate national and synthetic control groups; without full population data, comparisons risk bias and would not be reliable or generalisable. DfE will transfer data to NHS England. This Data is a list of NPD identifier variables. The data will consist of identifying details (specifically Name, Date of Birth, Postcode, Sex and a unique person ID) for the cohort to be linked with NHS England data. NHS England will match the identifiers from DfE to records held in the MPS/PDS using an algorithm that makes use of the chronology of postcodes in NPD and MPS/PDS. Matching to MPS/PDS data will be done internally within NHS England: no MPS/PDS data will be disseminated to ONS SRS or UCL Data Safe Haven. NHS England will link the NPD pseudonymised linkage key (i.e. anonymised PMR or young person ID) to MPS/PDS, and then to the ECDS data. For those children and young people whose NPD identifiers were matched to MPS/PDS, onward linkage to health data will occur within NHS England, linking aPMRs and Token Person IDs. NHS England will then transfer encrypted Token Person IDs, aPMRs, and indicators of match rank (denoting the step at which the match to HES and MPS/PDS was made) for these linked cases to the ONS SRS. NHS England will extract the health data for all children and young people born on or after 1.9.1984, plus health data for mothers of these children (Token Person IDs provided by UCL). including a pseudonymised mother-baby link and additional HES records of mothers, and link the aPMR and match rank statistics for those children and young people that were linked by NHS England from NPD. The deidentified health data will be transferred to the ONS SRS. Only month/year of birth and death will be transferred to the ONS SRS, in order to account for well-established effects of month of birth on school achievement (i.e. research consistently shows that children born in September do better than children born in July/August). DfE will supply ONS SRS with requested de-identified attribute data extracts, with the aPMR for all children and young people born on or after 1.9.1984. The deidentified attribute NPD and HES data will be linked within the ONS SRS by the research team, using the aPMR. Data will only be used by researchers authorised for the project or those who have been granted access to the data through a sub-license, with strict output controls applied by ONS SRS staff. The final data set that will be used for analyses will remain within the ONS SRS. The files will not contain any identifiable data. No additional record level data will be gathered or linked to the dataset. The aPMR is the only variable supplied from NPD data that is supplied by NHS England to UCL Data Safe Haven and then to ONS SRS. NHS England will retain the identifier file of all individuals linked in MPS/NPD-PDS and MPS/PDS-HES and all the postcodes used in linkage and postcode dates for 12 months after linkage, to address data queries or potential linkage errors. This data set will not contain any attribute data and will be accessible only to NHS England staff. At the end of the 12 months, NHS England will confirm deletion of the data to DfE. NHS England will not send confidential data to DfE or UCL DSH. UCL PhD students who are under the supervision of UCL substantive employees will also be able to access the data for the purposes covered in this DSA. No MSc or undergraduate students will access the data Non-UCL PhD students who are under the supervision of fully Accredited substantive employees may also access the data via sublicence arrangements. LSHTM and IFS are permitted to have named researchers who are undertaking analyses (on the ONS SRS). These named researchers will be substantive employees of the respective organisations. DfE and DHSC analysts will access the ECHILD data on the ONS SRS under separate agreements with NHS England. These applications will be considered separately to the UCL sublicensing application. The de-identified linked HES-NPD attribute data will be held on the ONS SRS and will only be accessible remotely. No record level data can be removed from the ONS SRS and statistical disclosure controls are applied by ONS staff. Access will be restricted to named users, with ONS accreditation. The researchers have requested the minimum data necessary for their research, which reflects the administrative history of the child/young adult for a subset of the available fields (e.g. the researchers have requested 60% of available inpatient fields, with no sensitive or identifiable fields). Data from the newly requested datasets (MSDS, CSDS, Mental health data) have not been minimised, due to the need to further explore these data in order to understand which variables are useful and sufficiently complete. Once this has been completed, the data will then be minimised, hopefully within 12 months. Data will be restricted to records relating to individuals in England born from 01/09/1984 onwards and their mothers. Longitudinal data for all children and individuals in England born from 01/09/1984 is justified for the following reasons: [1 paragraph unchanged] Examining health data from the time of birth to adulthood is critical [58 words unchanged] UCL have demonstrated the added value of using the whole longitudinal record. The study’s age range of all young people in England born on or after 01.09.1984 and use of the mother-baby flag The study’s age range of all young people in England born on or after 01.09.1984 and use of the mother-baby flag will allow evaluation of health outcomes in early adulthood that may be [22 words unchanged] factors among women who give birth on health outcomes in their child. [1 paragraph unchanged] The research aims to draw conclusions that are valid for all children and individuals in England. However, we know that health outcomes and service use vary across the country (e.g. both COVID infection rates and public health responses varied geographically) at different times. For example, surveys (e.g. The Royal College of Paediatrics and Child Health (RCPCH)) indicate geographical heterogeneity, including re-routing/re-deployment of healthcare staff and services, uptake of school access by eligible children, which are likely to disproportionately impact on areas with higher levels of overcrowding, less outside space, and greater deprivation. However, many surveys have incomplete coverage by geography or over time, making it difficult to accurately estimate the scale of the problem. Understanding time-varying patterns of change is increasingly important as public health responses shift towards localised management (e.g. local lockdowns) to control spread. The researchers therefore need data that makes it possible to understand local area impacts. The researchers have requested the minimum granularity possible, for example by requesting Middle layer Super Output Area rather than Lower Super Output Area. The research aims to draw conclusions that are valid for all children and individuals in England. However, we know that health outcomes and service use vary across the country (e.g. both COVID infection rates and public health responses varied geographically) at different times. For example, surveys (e.g. The Royal College of Paediatrics and Child Health (RCPCH)) indicate geographical heterogeneity, including re-routing/re-deployment of healthcare staff and services, uptake of school access by eligible children, which are likely to disproportionately impact on areas with higher levels of overcrowding, less outside space, and greater deprivation. However, many surveys have incomplete coverage by geography or over time, making it difficult to accurately estimate the scale of the problem. Understanding time-varying patterns of change is increasingly important as public health responses shift towards localised management (e.g. local lockdowns) to control spread. The researchers therefore need data that makes it possible to understand local area impacts. The researchers have requested the minimum granularity possible, for example by requesting Middle layer Super Output Area rather than Lower Super Output Area. [1 paragraph unchanged] In order for the research to draw meaningful conclusions, the researchers wish [9 words unchanged] care leavers, children born preterm, or other vulnerable groups) relative to a series of control children. The researchers will draw high level comparisons (e.g. to all other children) relevant to evaluating impacts at national level and for international comparisons, as well as detailed comparisons against synthetic control groups (e.g. through propensity score matching) to better understand the impacts of different groups of children in the context of related factors such as local environment, access to schools and healthcare needs. The researchers therefore require data for all children and individuals in England as without these data, comparisons would be incomplete, at greater risk of selection bias and not generalisable. DATA FLOWS series of control children. The researchers will draw high level comparisons (e.g. to all other children) relevant to evaluating impacts at national level and for international comparisons, as well as detailed comparisons against synthetic control groups (e.g. through propensity score matching) to better understand the impacts of different groups of children in the context of related factors such as local environment, access to schools and healthcare needs. The researchers therefore require data for all children and individuals in England as without these data, comparisons would be incomplete, at greater risk of selection bias and not generalisable. Linkage of identifiers from health and NPD has been conducted at NHS England. NHS England have transferred the pseudonymised linkage key to the UCL Data Safe Haven to flag linked records in the UCL-curated HES extract for transfer to the ONS Secure Research Statistics (SRS). NPD attribute data will only be available in the ONS SRS. Using the pseudonymised linkage key, merging of pseudonymised attribute data (clinical or education characteristics) will then occur separately, at the ONS Secure Research Service (SRS). The following outline describes the complete data flow for future linkages and details how identifiable and non-identifiable data extracts are handled. In the below, health data refers to the health datasets requested from NHS England (HES, mortality, ECDS, MSDS, CSDS, birth notifications/registrations, mental health data). 1) DfE supply the Trusted Third Party (NHS England) with a list of NPD identifier variables. These identifiers include name, date of birth, full postcode and sex, alongside a study specific pseudonymised linkage key known as the anonymized Pupil Matching Reference (aPMR). The identifying variables will be used for linkage to the Master Person Service (MPS)/Personal Demographic Service (PDS) (as previously done for NIC-27404 and NIC-381972). DfE will transfer the variables for any CYP born on or after cohort inception (1.9.84). 2) NHS England will match the identifiers from DfE to records held in the MPS/PDS using an algorithm that makes use of the chronology of postcodes in NPD and MPS/PDS. Matching to MPS/PDS data will be done internally within NHS England: no MPS/PDS data will be disseminated to ONS SRS or UCL Data Safe Haven. NHS England will link the NPD pseudonymised linkage key (i.e. anonymised PMR or young person ID) to MPS/PDS, and then to the ECDS data. 3) For those children and young people whose NPD identifiers were matched to MPS/PDS, onward linkage to health data will occur within NHS England, linking aPMRs and Token Person IDs. NHS England will then transfer encrypted Token Person IDs, aPMRs, and indicators of match rank (denoting the step at which the match to HES and MPS/PDS was made) for these linked cases to the ONS SRS. 4) NHS England will extract the health data for all children and young people born on or after 1.9.1984, including a pseudonymised mother-baby link and additional HES records of mothers, and link the aPMR and match rank statistics for those children and young people that were linked by NHS England in step (3) from NPD. The deidentified health data will be transferred to the ONS SRS. Only month/year of birth and death will be transferred to the ONS SRS, in order to account for well-established effects of month of birth on school achievement (i.e. research consistently shows that children born in September do better than children born in July/August). 5) DfE will supply ONS SRS with requested de-identified attribute data extracts, with the aPMR for all children and young people born on or after 1.9.1984. The deidentified attribute NPD and HES data will be linked within the ONS SRS by the research team, using the aPMR. Data will only be used by researchers authorised for the project or those who have been granted access to the data through a sub-license, with strict output controls applied by ONS SRS staff. 6) The final data set that will be used for analyses will remain within the ONS SRS. The files will not contain any identifiable data. No additional record level data will be gathered or linked to the dataset. The aPMR is the only variable supplied from NPD data that is supplied by NHS England to UCL Data Safe Haven and then to ONS SRS. 7) NHS England will retain the identifier file of all individuals linked in MPS/NPD-PDS and MPS/PDS-HES and all the postcodes used in linkage and postcode dates for 12 months after linkage, to address data queries or potential linkage errors. This data set will not contain any attribute data and will be accessible only to NHS England staff. At the end of the 12 months, NHS England will confirm deletion of the data to DfE. NHS England will not send confidential data to DfE or UCL DSH. ACCESS TO THE DATA BY UNIVERSITY COLLEGE LONDON, LONDON SCHOOL OF HYGIENCE AND TROPICAL MEDICINE (LSHTM) AND INSTITUTE FOR FISCAL STUDIES (IFS) (NAMED DATA PROCESSORS) UCL researchers who have a substantive contract with UCL will be authorised to access the data in the ONS SRS, for purposes covered in this DSA. UCL PhD students who are under the supervision of UCL substantive employees will also be able to access the data for the purposes covered in this DSA. No MSc or undergraduate students will access the data. All UCL PHD students are expected to undertake annual training on handling highly confidential information. All Trainees and students register for and complete NHS England’s Data Security Awareness (NHSD) course provided by e-Learning for Health. The course covers data security awareness, the law, threats to data security, breaches and incidents, and the General Data Protection Regulation. All UCL staff and students are also required to complete internal UCL GDPR training annually. UCL has a specific data protection and information security policy, which applies to all staff and students when processing personal data on behalf of UCL. All UCL students working on the study are bound by this policy, and that they will face potential sanctions in the event of a breach of the policy. All UCL students sign up to the UCL's Academic Manual. The Student Academic Misconduct section of the 2021-2022 manual Section 9.1, item 3 states "All instances of Research Misconduct whether by taught students, research students or members of staff will be investigated under UCL’s Procedure for Investigating and Resolving Allegations of Misconduct in Academic Research". LSHTM and IFS each have a named researcher and Principle Investigator (PI) who are undertaking analyses (on the ONS SRS) relating to a specific component of the original research question e.g. LSHTM will examine the impact (on health and education) of delays in time-sensitive procedures (e.g. surgical correction of cleft lip and palette) on children with underlying health conditions. These named researchers will be substantive employees of the respective organisations. DfE and DHSC analysts will access the ECHILD data on the ONS SRS under separate agreements with NHS England. These applications will be considered separately to the UCL sublicensing application. The de-identified linked HES-NPD attribute data will be held on the ONS SRS and will only be accessible remotely. No record level data can be removed from the ONS SRS and statistical disclosure controls are applied by ONS staff. Access will be restricted to named users, with ONS accreditation. Amazon Web Services provides cloud hosting services to UCL and will store the data as contracted by UCL. UCL uses offsite data centre services provided by VIRTUS data centre. DATA STORAGE The Office of National Statistics (ONS) and UCL have signed and maintain an organisational agreement to use the ONS Secure Research Statistics (SRS) service for the purposes of secure statistical research, signed on 21/03/2019 with an indefinite expiry date. HES data will be stored in the ONS SRS, plus the anonymised PMRs for those records that link to NPD. The HES data will not include any identifiable data. For security and resource reasons the SRS is a Managed Service. Equiniti Ltd (based in Belfast) maintains the system, on behalf of the ONS SRS. They do so through encrypted (TLS1.2) VPN tunnel and Remotely Access (RA) the SRS. All Equiniti Ltd administrators are SC cleared and have no access to any data. ONS SRS Research Support Administrative staff only have permissions to carry out such tasks as creating users, updating patches, testing and installing software applications, arranging DR, ITHC for the SRS environment, closing SRS sessions down, i.e. all the SRS environment Admin maintenance - essentially they are power users. There have been no data infractions by Equiniti Ltd staff in the last 5 years of them maintaining the environment. The high level security document that Equiniti Ltd provided states: 6.3. Service Management support for the SRS Service is provided from Equiniti offices in Belfast, all staff are SC cleared. The office hosting the SRS Desk is IS0/IEC27001 2018 certified. Equiniti Ltd nor any their staff process the data. Therefore Equinity Ltd is not considered to be a Data Processor. The ONS SRS environment is an isolated system. It has no connectivity to the internet other than using it as a bearer to pass TLS1.2 encrypted image packages for a virtual desktop infrastructure (VDI), hosted on an accredited cloud server hosted by UKCloud Ltd on the mainland UK. UKCloud Ltd merely host the environment, they have no access to data. Therefore CloudUK Ltd is not considered to be a Data Processor. CLOUD SECURITY NHS England security has provided assurance regarding the use of the Office of National Statistics' Secure Research Statistics service (ONS SRS), hosted by CloudUK Ltd in this application. The Office of National Statistics has submitted a selection of security documentation to support the use of cloud storage. NHS England Security have reviewed the documentation and provided relevant feedback, where necessary. NHS England are satisfied that the documentation demonstrates the level of security and governance in place. The Office of National Statistics have supplied evidence to support: * The use of the Data Risk Model to assess the Risk Profile Class. * Risk Management of the use of the Cloud for this data, taking into consideration Confidentiality, Integrity and Availability. * The use of Pseudonymisation. * Board level involvement in the Risk Management Process evidenced through Minutes of these meetings. ͻUnderstanding of the Shared Responsibility Model The Office of National Statistics have a very good understanding of the security controls available to them to provide the appropriate controls to secure data in the Cloud. Using the Cloud, benefits from the inherited controls that cannot practically be replicated locally such as Physical Controls, Resilience of Systems, Power Supplies, Communications and Geographically dispersed Data Centres within a region. Elasticity in provisioning is also a consideration that benefits organisations in managing workloads. The Cloud provider, CloudUK, will use UK Data Centres only. DISCLOSURE CONTROL RULES Community Services Dataset The suppression rules are as follows (applicable if one of something is necessarily one person e.g. length of stay - one day as an inpatient can only relate to one person in many data sets): · Zeros should be shown. · 1-7 to be rounded to 5. · Any other numbers rounded to nearest 5. · Rounding unnecessary for averages etc. · Percentages calculated from rounded values Maternity Services Dataset (MSDS) In order to prevent disclosure of identities or information about service users, · all figures (except national) for all organisations which submitted, are rounded to the nearest five; · all figures between zero and four are suppressed (*); · averages with a denominator of less than five have been replaced by '*' and other values have been rounded to one decimal place. Mental Health Services Dataset (MSDS) and Mental Health Services Dataset (MHSDS) v5.0 The publication covers sensitive topics. There would be a risk of identification/self-identification from small numbers. In order to minimise the disclosure risk associated with small numbers, all figures presented within the report and within the reference data tables have had the following measures applied: · Present zeros; · 1-7 rounded to 5 · Other values have been rounded to the nearest 5; · Percentages calculated from rounded values.

Expected output

All outputs will contain aggregate level data only and all small numbers [104 words unchanged] for export by an ONS data scientist not involved in the project. All ECHILD Publications can be found on the ECHILD website: https://www.echild.ac.uk/publications. The researchers will submit full reports for publication in peer reviewed journals and produce briefing reports for policy stakeholders and lay summaries. For example, for research conducted for the NIHR Children and Families Policy Research Unit (CPRU), reports will be produced for DHSC. Findings will also be used in public involvement and engagement events. Study [12 words unchanged] Journal (BMJ), Lancet Public Health), and social media including lay summaries. UCL would expect that findings from the research will be presented at conferences such [7 words unchanged] International Population Data Linkage Conference within two years of obtaining the data. Sublicense agreements will encourage relevant findings to be shared with policy makers, [34 words unchanged] Street Hospital (GOSH) Patient Engagement group. Lay summaries of the study findings can will be published on the ECHILD website, and linked through websites for sub-licensee organisations. Sublicensees will be required to report yielded benefits to UCL and NHS England at the time of each annual review of the sub-license agreement. Sublicensees will also be required to notify UCL of all publications in advance of publication. Details of the publications will then be recorded on the ECHILD Release Register (https://www.ucl.ac.uk/child-health/research/population-policy-and-practice-research-and-teaching-department/cenb-clinical-22 and also see ECHILD Data Access Decision Making Process Standard Operating Procedure). The release register will also include details of the research project (including a summary of the purpose), the licensee’s organisation and Chief Investigator, and the licence end date. Sublicensees will also be required to notify UCL of all publications in advance of publication. Details of any sublicensed project is recorded on the ECHILD Release Register (https://www.echild.ac.uk/data-release-register). The release register will also include details of the research project (including a summary of the purpose), the licensee’s organisation and Chief Investigator, and the licence end date. [1 paragraph unchanged]

Expected measurable benefits

[12 paragraphs unchanged] Analyses will hopefully explore outcomes for the whole population, to possibly inform [21 words unchanged] poor outcomes. Findings may be reported directly to DfE and NHS policy makers. makers

Benefits reported

The expected outputs and benefits outlined in our original and prior application have not yet been fully achieved, primarily due to delays in data availability. Although the agreement was signed in May 2023, data was received late in the year (December 2023), and throughout 2024, the ONS uploaded data at a slow pace. Full access for the ECHILD team was only granted in August 2024, at which point applications were opened to external researchers for the first time. However, ongoing data issues, some related to the ONS and the DEA, others involving NHSE, are currently being addressed in collaboration with the relevant teams. Data sharing metrics: ECHILD currently has 118 users: 53 internal, and 65 from 23 external institutions. Since August 2024, UCL have approved 42 projects through the sublicensing model. Alongside these, there are 38 internal UCL projects currently ongoing, and 23 more external projects in preparation under consultation with the ECHILD team. For a typical application, feasibility review and guidance for refinement reflects reviewing at least two iterations of their protocol and often requires a meeting with the applicant to ensure that the protocol is ready to submit to ONS. This close working with researchers has meant that no eligible data access requests have been denied; researchers have been able to present feasible projects and/or to ensure that significant exploratory components are included for projects where data quality is uncertain. UCL have worked extensively with the ONS and applicants to ensure that data access processes are as efficient as possible (see Data Management section): the UCL Data Access Committee (DAC) aim to respond with feedback to completed applications within 2 weeks; ONS currently take an average 10 weeks to approve applications. Sublicensing Progress A collaborative approach: The UCL ECHILD team is embedded within the research community and is closely engaged with government. This collaborative approach allows us to attract high-quality research projects and support policy and practice-relevant, impactful research through data curation and understanding. Researchers using ECHILD generate information on methods, data quality, analysis tools, and findings from Patient and Public Involvement and Engagement (PPIE), which are fed-back to help the ECHILD team improve the quality of the data resource, justify further enhancements and advise potential users on questions that could be feasibly answered. Despite these challenges, we successfully met our key objective of enabling external accredited researchers to apply for access to ECHILD data via the ONS Secure Research Service (SRS) by August 2024. To date, the ECHILD team has approved 26 external research projects, including seven funded through the ADR UK Fellowship programme (as ECHILD is one of ADR UK’s flagship datasets). Of these, 20 projects have received approval from ONS RAS, with the remaining applications still under review. Policy-related impact and relevance to the public: Since September 2020, six ECHILD studies have been commissioned by the Department of Health and Social Care (DHSC) through the Children and Families Policy Research Unit (CPRU), including research on prioritising adolescents with additional needs for catch-up healthcare after COVID lockdowns, the relationship between chronic health conditions and school absence, mortality rates of adolescents receiving Special Educational Needs and Disability (SEND) provision or social care services, and the characteristics of children with high service utilisation across SEND, social and hospital care.(1)(2)(3)(4) ECHILD has also generated important information on health and education outcomes for different population groups (e.g. children with neurodisablity, congenital malformations, and Hirschsprung’s disease,) relevant to parents and children, clinicians and services.(5)(6)(7)A recent NESTA-funded study on the effect of the two-child limit on children’s school readiness by the Institute for Fiscal Studies (IFS) received widespread media and policy attention, concluding that although scrapping the two-child benefit limit would be one of the most effective ways to reduce child poverty, more still needs to be done to improve child development and school readiness at age five.(8) The NIHR-funded HOPE study (£1.46m) assessed variation and inequalities in access to SEND provision across England and used causal methods to assess the impact of SEND provision on health and education outcomes in primary school.(9) A £2m NIHR programme evaluating the implementation of the Transforming Children and Young People's Mental Health Provision Green Paper plans to use ECHILD alongside mixed methods approaches.(10) UCL share preliminary research findings with government (DfE and DHSC) through seminars, pre-published material and meetings with analysts and policy makers. For example, UCL have recently presented to DfE on mental health presentations (n=126 attendees) and community services (n=77). Therefore, UCL are requesting a 12-month extension to our current DSA to allow sufficient time for processing the requested data and effectively demonstrating its benefits. Capacity-building impact: ECHILD has contributed extensively to academic training and capacity building. UCL currently support 15 UCL PhD students and 7 ADR UK research fellows; UCL expect to recruit more ADR fellows in the current round, and a number of other fellowship applications using ECHILD are currently in progress or under review. UCLs two-day in person ECHILD training course was delivered to 30 participants in March 2024 and 42 participants online in November 2024. UCL currently offer on-demand training material via Instats, which has had 14 additional participants, and are planning a further in-person training course in September 2026, to align with the new ADR UK fellowship start date. UCLs online Seminar series, running 2-3 times per year, has been extremely popular, attracting ~100 academic/government participants each time. During 2025, the HOPE and wider ECHILD teams delivered targeted seminars to DfE and DHSC analysts on critical research areas, and presented to the Children’s Commissioner Office. UCL have also contributed to ADR UK Data Insights. In addition to the yielded benefits outlined in the previous iteration of this agreement (signed in March 2024), we have achieved the following outputs/benefits: Publications: ECHILD has generated 44 journal publications, including 17 peer-reviewed journal articles (with ~10 others currently under review), 10 published research protocols, and 17 pre-prints. UCL have given 33 conference presentations, 6 published reports, 17 blogs, and 23 webinars • Awards: Awards and recognition: The ECHILD team were awarded the ONS Research Excellence Award in 2024 for Secure Data Creation, an ADR UK Partnership Award (Raising to a Challenge) in 2024, and the ONS Research Excellence Award for Early Career Researchers in 2022. ECHILD team members have also been awarded two conference presentation prizes. UCL were nominated and shortlisted for a HDRUK Impact Award, featuring in the associated report. ECHILD has been highlighted as a key initiative in UK data sharing across various settings: at the launch of the Cathie Sudlow Review (2024); by Dr Emma Gordon in her OECD Symposium speech (2024); at the UK Parliament House of Lords Preterm Birth Committee (2024), by the Academy of Medical Sciences in their 2024 report (11) on child health; and in an NIHR Policy Research Programme Research Specification launch (2024). ECHILD was also recognised as an example of good practice by the Office for Statistics Regulation.(12) ECHILD has been highlighted within ADR UK Impact Case Studies for UCLs work on code sharing and PPIE. - ECHILD awarded the ONS Research Excellence Award 2024 in Secure Data Creation Funding: Leveraging the £1.3M ADR funding since 2020, the UCL ECHILD team have contributed to >£31 million in grant and fellowship income through ~30 grant applications to various funders (including NIHR, MRC, NESTA, ESRC, HDRUK). - ECHILD Shortlisted for the HDR UK Susannah Boddie Award for Impact of the Year (results in March 2025) Community Building: An important component of the current grant has been to foster collaboration, exchange knowledge and share learning. A key initiative is the annual User Day which brings together users from across the country to share their experiences and future research plans, discuss emerging challenges, and develop solutions that enhance the impact and accessibility of ECHILD. UCLs 2023 and 2024 events attracted ~70 attendees each: UCLs next event is planned for Spring 2026. UCL maintain a large mailing list (>500 people) with whom UCL communicate important updates, upcoming events and news. UCLs Github Discussion Forum is an online space empowering users to ask data questions, raise issues, and start discussions, helping to foster collaboration and knowledge sharing. - Poster Award: Children with high needs for healthcare and special educational support in England, UK: a national cohort using novel linkage of health and education records from the ECHILD database (Ania Zylbersztejn) for the Highest Scoring Poster Abstract at the Paediatric Academic Societies meeting. May 2024, Toronto, Canada Academic Collaborations and Partnerships: UCL collaborate widely and partner with a large number of research programmes. For example, RG co-directs the NIHR CPRU and is deputy director of the HDRUK Social and Environmental Determinants of Health Research Driver Programme; KH is co-I for the ESRC Centre for Lifecourse Health Equity (Equalise), the ESRC Centre for Longitudinal Cohort Studies, and the NIHR Global Health Unit for Social and Environmental Determinants of Health Inequalities; RB is lead for Children and Young People’s steam of the UKRI Population Mental Health Consortium. UCL also work closely with the Kid’s Environment and Health Cohort team (PI: Pia Hardelid) and the UCL Great Ormond Street Biomedical Research Centre. These collaborations ensure that ECHILD is recognised as a core resource within the UK data landscape. - ECHILD has been showcased as an Impactful resource in the latest HDRUK Impact Report “Making Data Count” Technical enhancements: UCLs data enhancements have been popular with users: 32 of the 42 approved projects since August 2024 will use the Mental Health data, and 20 will use the mother-baby link. Use of the Maternity Services Dataset (18 applications) has been more limited due to low data quality, but UCLs curation has highlighted where variables can be used (Data Profile accepted for publication). UCLs extension of the cohort to include individuals born since 1984, alongside the mother-baby link, has allowed follow up of mothers through children’s social care data and their child outcomes.(13) - ECHILD is featured in the Cathie Sudlow Review launch and report, Uniting the UK’s Health Data: A Huge Opportunity for Society, as a key example of the benefits of linking data from multiple sources. Patient and Public Involvement and Engagement: Since 2022, UCL have organised >25 PPIE activities with at least 350 participants, encompassing events and initiatives with parents, carers, and wider community groups to ensure lived experiences inform UCLs research. These engagements reflect research projects across the lifecycle: PPIE has been used to steer the development of ECHILD linkages at a very early stage right through to late-stage dissemination. UCL have also established UCLs Lay Advisory Group (22 contributors), which has played a key role in the development of the website • Peer Reviewed Publications: An exemplar for early-stage work informing future linkages includes UCLs work with an expert public panel (HDR UK’s Public Advisory Panel) for data driven research to co-produce guidance on linkage of place-based data for research. This guidance includes signposting to PEDRI’s Good Practice Standards for PPIE in data research and to existing high quality public-facing resources explaining place-based linkages. The learning resources are disseminated to researchers through development of a learning module hosted on the HDR UK Futures platform and promoted at conferences and seminars - School-recorded special educational needs provision in children with major congenital anomalies: a linked administrative records study of births in England, 2003-2013 (Maria Peppa et al, 2025) - Hospital-recorded chronic health conditions in children with and without Down syndrome in England: a national cohort of births from 2003 to 2019 (Julia Shumway et al, 2025) - Cumulative incidence of chronic health conditions recorded in hospital inpatient admissions from birth to age 16 in England (Matthew Jay et al, 2024) - Data Resource Profile: ECHILD only-children and siblings (ECHILD-oCSib): a national cohort of linked health, education and social care data on mothers and children in England (Qi Feng et al 2024) - Data Resource Profile: A national linked mother-baby cohort of health, education and social care data in England (ECHILD-MB) (Qi Feng et al 2024) • Working Papers: - Hospital contacts amongst high achieving adolescents from disadvantaged socio-economic backgrounds (John Jerrim) - Academic selection, relative achievement and hospital contacts due to mental health and adjustment issues in England (John Jerrim) - Educational outcomes of children with congenital anomalies (trajectories in attainment) (Joachim Tan) - Primary school attainment outcomes in children with neurodisability: A population-based cohort study using linked education and hospital data from England (Ayana Cant) - Planned and unplanned hospital admissions and health-related school absence rates in children with neurodisability: A population-based study using linked education and hospital data from England (Laura Gimeno) - Health and education outcomes of primary school children with Down Syndrome in England (Julia Shumway) - Phenotyping neurodisability in hospital admissions records in England: a national birth cohort from linked administrative data (Ania Zylbersztejn) - Sociodemographic variation in the timing of recorded special educational needs provision in primary school in England amongst children with cerebral palsy (Kate Lewis) - SEN provision during primary education for children born in 2007-08 in England: demographic and social characteristics (prelim title) (Vincent Nguyen) - Local authority variation in primary school-recorded special educational needs provision among children with major congenital anomalies (Kate Lewis) - Impact of special educational needs provision on hospital utilisation, school attainment and absences for children in English primary schools for children with cerebral palsy: A target trial emulation (Vincent Nguyen) - Short- and long-term effects of Special Educational Needs (SEN) provision on school attendance: Insights from G-Computation and Dynamic Panel Models (Lorraine Dearden) - Causal analysis of educational policies: a review of econometrics and biostatistics methods (Bianca De Stavola) - Early special educational needs provision and its impact on unplanned hospital utilisation and school absences in children with isolated cleft lip and/or palate: a demonstration target trial emulation study using ECHILD (Vincent Nguyen) • Protocols: - Planned and unplanned hospital admissions and health-related school absence rates in children with neurodisability: Protocol for a population-based study using linked education and hospital data from England (Laura Gimeno et al, 2024) - Primary school attainment outcomes in children with neurodisability: Protocol for a population-based cohort study using linked education and hospital data from England (Ayana Cant et al, 2024) - Educational outcomes of children with major congenital anomalies: Study protocol for a population-based cohort study using linked hospital and education data from England (Joachim Tan et al, 2025) • Conference Abstracts: 1- Cumulative incidence of chronic health conditions recorded in hospital admissions from birth to age 16 in England: a cohort study using linked hospital and education administrative data (IPDLN 2024) 2- Screening for psychological distress in childhood and mental health-related hospital attendance among young adults (IPDLN 2024) 3- Health and education needs of primary school children with Down Syndrome in England: What can we learn from linked administrative data? (IPDLN 2024) 4- 10 year risk of death in adolescents with learning disability or autism in England: a national cohort study using linked health and education data from ECHILD (IPDLN 2024) 5- Hospital admissions among adolescents in out-of-home care or special educational needs in England: insights from the ECHILD database of linked health, education and social care records (IPDLN 2024) 6- Creating an intergenerational cohort of mothers and their children using administrative data from health, education and social care in England (ECHILD) (IPDLN 2024) 7- Mortality in adolescents receiving special education and social care provision: A population-level cohort study in England (IPDLN 2024) 8- Implementing Transparency Standards: The ECHILD Project in Action (May 2024) • Knowledge sharing outputs/events: - ECHILD Seminar Series, online - ECHILD User Days, in-person - Houses of Parliament call for evidence on the SEND crisis, - Talks at the Children’s Commissioner, Department for Health, NIHR Applied Research Collaborations and Biomedical Research Centres (public and practitioners), UKRI UK Regenerative Medicine Platform, Medical Research Council (MRC), Institute for Clinical Evaluation Services, UKRI SPF Programmes, National Foundation For Educational Research, BHF Data Science Centre, Children’s Rights Conference, ADRUK and HDRUK Conferences and Symposiums. • Resources and tools: - ECHILD How-To Guides - INSTATS ECHILD Course - on Demand - ECHILD Phenotype Code List Repository - ECHILD Video and Illustrations - Maintained ECHILD Data Catalogue v2 - Maintained ECHILD User Guide • Funded Projects: - 7 ADRUK Funded Fellowships for 18 months each. - 33 Funding sources supporting UCL research projects - At least 6 External funding sources supporting sublicensees • Users: Currently, 45 internal users (UCL and collaborators) have access to ECHILD. The ONS SRS has only recently begun setting up access for external researchers, with 6 external users currently accessing the data via sublicensing. This number is expected to grow, as at least 14 more projects have already been approved by ONS RAS.

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

Benefits reported

The expected outputs and benefits outlined in our original application have not been fully achieved as UCL have not yet received the data that were requested under the DSA (the data have been released but have not yet been ingested to the SRS). UCL have not yet released any data through our sublicense. UCL are therefore requesting a 12 month extension to our DSA in order to have enough time to receive and process the requested data, and to demonstrate benefits. The expected outputs and benefits outlined in our original and prior application have not yet been fully achieved, primarily due to delays in data availability. Although the agreement was signed in May 2023, data was received late in the year (December 2023), and throughout 2024, the ONS uploaded data at a slow pace. Full access for the ECHILD team was only granted in August 2024, at which point applications were opened to external researchers for the first time. However, ongoing data issues, some related to the ONS and the DEA, others involving NHSE, are currently being addressed in collaboration with the relevant teams. In addition to the yielded benefits outlined in the previous iteration of this agreement (signed in May 2023), UCL have continued to work on the data previously released under our DSA, and have achieved the following outputs/benefits: Sublicensing Progress 1. ECHILD has been highlighted as a key initiative to improve access to and use of data to improve health, including in the early years in the recent Academy of Medical Sciences report on “Prioritising early childhood to promote the nations’ health, wellbeing and prosperity”: https://acmedsci.ac.uk/file-download/16927511 Despite these challenges, we successfully met our key objective of enabling external accredited researchers to apply for access to ECHILD data via the ONS Secure Research Service (SRS) by August 2024. To date, the ECHILD team has approved 26 external research projects, including seven funded through the ADR UK Fellowship programme (as ECHILD is one of ADR UK’s flagship datasets). Of these, 20 projects have received approval from ONS RAS, with the remaining applications still under review. 2. ECHILD was mentioned in the parliament at the Preterm Birth Committee 2024 (see 15:20 https://www.parliamentlive.tv/Event/Index/bc47da21-c39c-44d8-9dfe-f9cb61e1a6d8?_gl=1*1wnshlm*_ga*MTUwNzQwMTk1My4xNzA3NzU4MTA1*_ga_L0NJWDWMGN*MTcwNzc1ODEwNC4xLjEuMTcwNzc1ODEwOC41Ni4wLjA) Therefore, UCL are requesting a 12-month extension to our current DSA to allow sufficient time for processing the requested data and effectively demonstrating its benefits. 3. ADRUK have funded three fellowships to use ECHILD. In addition to the yielded benefits outlined in the previous iteration of this agreement (signed in March 2024), we have achieved the following outputs/benefits: 4. ADRUK currently have a call out advertising for new fellowships using ECHILD: https://www.ukri.org/opportunity/adr-uk-research-fellowships-2024/#:~:text=Administrative%20Data%20Research%20UK%20(ADR,organisation%20eligible%20for%20ESRC%20funding. • Awards: 5. ADRUK have classified ECHILD as one of their flagship datasets (https://www.adruk.org/fileadmin/uploads/adruk/Documents/ADR-England-flagship-dataset-brochure.pdf) and have funded three fellowships to work on ECHILD through this funding call: https://www.ukri.org/opportunity/adr-uk-research-fellowships-2023/ - ECHILD awarded the ONS Research Excellence Award 2024 in Secure Data Creation 6. ECHILD has been highlighted as a key data resource in a 2024 NIHR Programme Development Grant call (see 56:30 https://youtu.be/NO-7AGgxTtg) - ECHILD Shortlisted for the HDR UK Susannah Boddie Award for Impact of the Year (results in March 2025) 7. As part of a Research Community Catalyst for Children at Risk of Poor Outcomes, ADRUK have funded an ECHILD researcher to analyse children’s social care data (https://www.adruk.org/news-publications/news-blogs/two-new-projects-will-develop-research-communities-and-drive-transformative-insights-about-children-and-young-people/) - Poster Award: Children with high needs for healthcare and special educational support in England, UK: a national cohort using novel linkage of health and education records from the ECHILD database (Ania Zylbersztejn) for the Highest Scoring Poster Abstract at the Paediatric Academic Societies meeting. May 2024, Toronto, Canada 8. UCL have published a protocol in BMJ Open for the HOPE study, which aims to evaluate special educational needs provision and its impact on health and education outcomes: Evaluation of variation in special educational needs provision and its impact on health and education using administrative records for England: umbrella protocol for a mixed-methods research programme. The HOPE study aims to build the evidence base for fairer and more effective SEN provision and, by informing national and local policy and the public and changing practice, to improve health and education outcomes of children with SEN. - ECHILD has been showcased as an Impactful resource in the latest HDRUK Impact Report “Making Data Count” 9. UCL published a paper on cleft lip and palate repair surgeries before and during COVID: Number and timing of primary cleft lip palate repair surgeries in England: whole nation study of electronic health records before and during the COVID-19 pandemic. This paper showed significant reductions in the number and delays in timing of first primary CLP repair procedures in England during the first year of the pandemic, which may affect long-term outcomes. - ECHILD is featured in the Cathie Sudlow Review launch and report, Uniting the UK’s Health Data: A Huge Opportunity for Society, as a key example of the benefits of linking data from multiple sources. 10. UCL published a paper on hospital admissions for stress-related presentations among school-aged adolescents: Hospital admissions for stress-related presentations among school-aged adolescents during term time versus holidays in England: weekly time series and retrospective cross-sectional analysis. This paper showed that hospital admissions for stress-related presentations are common, affecting around two girls and one boy in every classroom in England. • Peer Reviewed Publications: 11. UCL published an abstract on the educational outcomes of children with chronic liver disease and presented this as the RCPCH conference: Educational outcomes in children with chronic liver disease in England are inferior to peers: evidence from 5 million children. This analysis can be used to inform educational and health services policies of the urgent need for neuro-developmental assessment to be included in the routine care for children with chronic liver disease, to ensure early detection and referral to specialist services. - School-recorded special educational needs provision in children with major congenital anomalies: a linked administrative records study of births in England, 2003-2013 (Maria Peppa et al, 2025) 12. UCL published an abstract on special educational needs of primary school aged children with neurodevelopmental conditions and presented this work at the Society for Social Medicine conference: Special educational needs of primary school aged children with neurodevelopmental conditions: a population cohort study using linked health and education records. This analysis showed that children with neurodevelopmental conditions have varying levels of SEN provision, with more intensive provision increasing and less intensive provision decreasing during primary school. - Hospital-recorded chronic health conditions in children with and without Down syndrome in England: a national cohort of births from 2003 to 2019 (Julia Shumway et al, 2025) 13. UCL published an abstract on self-harm hospitalisations in secondary school pupils in England and presented this at the RCPCH and YPHSIG Adolescent Health conference: Quantifying emergency hospital admissions with self-harm in secondary school pupils in England: whole nation study of linked data from health and education. This analysis showed that Individually targeted interventions (e.g. to girls with a history of self-harm) may be more effective than universal strategies for reducing self-harm admissions. - Cumulative incidence of chronic health conditions recorded in hospital inpatient admissions from birth to age 16 in England (Matthew Jay et al, 2024) 14. UCL published a study protocol for a target trial emulation study of the impact of special educational needs provision on hospital utilisation for children according to gestational age at birth: Impact of special educational needs provision on hospital utilisation, school attainment and absences for children in English primary schools stratified by gestational age at birth: A target trial emulation study protocol. This work will inform whether SEN provision can improve educational and health outcomes. - Data Resource Profile: ECHILD only-children and siblings (ECHILD-oCSib): a national cohort of linked health, education and social care data on mothers and children in England (Qi Feng et al 2024) 15. UCL published a study protocol for a study of the impact of special educational needs on children with cleft lip and/or palate: Early special educational needs provision and its impact on unplanned hospital utilisation and school absences in children with isolated cleft lip and/or palate: a demonstration target trial emulation study protocol using ECHILD. This analysis will inform whether reasonable adjustments at the start of compulsory education can improve health and educational outcomes in the cleft lip and palate population. - Data Resource Profile: A national linked mother-baby cohort of health, education and social care data in England (ECHILD-MB) (Qi Feng et al 2024) 16. UCL published an abstract on special educational needs and school readiness in children with neurodevelopmental conditions and presented this at the British Academy of Childhood Disability conference: Special educational needs and school readiness of children with neurodevelopmental conditions: a national cohort using linked health and education records. This analysis showed that the small proportion of children with neurodevelopmental conditions account for 22% of Education and Health Care Plans. • Working Papers: 17. UCL published a study protocol for a study on local authority variation in primary school-recorded special educational need provision in children with major congenital anomalies: Local authority variation in primary school-recorded special educational needs provision among children with major congenital anomalies: A research protocol. This analysis will help UCL to understand how equal special educational needs provision is across England and inform policies to best support children. - Hospital contacts amongst high achieving adolescents from disadvantaged socio-economic backgrounds (John Jerrim) 18. UCL have been maintaining an ECHILD user guide, data catalogue and analysis code, and stakeholder work to support wider use: - Academic selection, relative achievement and hospital contacts due to mental health and adjustment issues in England (John Jerrim) o ECHILD User Guide V2: https://www.ucl.ac.uk/child-health/sites/child_health/files/echild_user_guide_v2.pdf (updated June 2023) - Educational outcomes of children with congenital anomalies (trajectories in attainment) (Joachim Tan) o ECHILD Data Catalogue: https://www.ucl.ac.uk/child-health/sites/child_health/files/echild_datacatalogue_v2.1_202310.xlsx (updated October 2023) - Primary school attainment outcomes in children with neurodisability: A population-based cohort study using linked education and hospital data from England (Ayana Cant) o Github for reusable code: https://github.com/UCL-CHIG - Planned and unplanned hospital admissions and health-related school absence rates in children with neurodisability: A population-based study using linked education and hospital data from England (Laura Gimeno) - Health and education outcomes of primary school children with Down Syndrome in England (Julia Shumway) - Phenotyping neurodisability in hospital admissions records in England: a national birth cohort from linked administrative data (Ania Zylbersztejn) - Sociodemographic variation in the timing of recorded special educational needs provision in primary school in England amongst children with cerebral palsy (Kate Lewis) - SEN provision during primary education for children born in 2007-08 in England: demographic and social characteristics (prelim title) (Vincent Nguyen) - Local authority variation in primary school-recorded special educational needs provision among children with major congenital anomalies (Kate Lewis) - Impact of special educational needs provision on hospital utilisation, school attainment and absences for children in English primary schools for children with cerebral palsy: A target trial emulation (Vincent Nguyen) - Short- and long-term effects of Special Educational Needs (SEN) provision on school attendance: Insights from G-Computation and Dynamic Panel Models (Lorraine Dearden) - Causal analysis of educational policies: a review of econometrics and biostatistics methods (Bianca De Stavola) - Early special educational needs provision and its impact on unplanned hospital utilisation and school absences in children with isolated cleft lip and/or palate: a demonstration target trial emulation study using ECHILD (Vincent Nguyen) • Protocols: - Planned and unplanned hospital admissions and health-related school absence rates in children with neurodisability: Protocol for a population-based study using linked education and hospital data from England (Laura Gimeno et al, 2024) - Primary school attainment outcomes in children with neurodisability: Protocol for a population-based cohort study using linked education and hospital data from England (Ayana Cant et al, 2024) - Educational outcomes of children with major congenital anomalies: Study protocol for a population-based cohort study using linked hospital and education data from England (Joachim Tan et al, 2025) • Conference Abstracts: 1- Cumulative incidence of chronic health conditions recorded in hospital admissions from birth to age 16 in England: a cohort study using linked hospital and education administrative data (IPDLN 2024) 2- Screening for psychological distress in childhood and mental health-related hospital attendance among young adults (IPDLN 2024) 3- Health and education needs of primary school children with Down Syndrome in England: What can we learn from linked administrative data? (IPDLN 2024) 4- 10 year risk of death in adolescents with learning disability or autism in England: a national cohort study using linked health and education data from ECHILD (IPDLN 2024) 5- Hospital admissions among adolescents in out-of-home care or special educational needs in England: insights from the ECHILD database of linked health, education and social care records (IPDLN 2024) 6- Creating an intergenerational cohort of mothers and their children using administrative data from health, education and social care in England (ECHILD) (IPDLN 2024) 7- Mortality in adolescents receiving special education and social care provision: A population-level cohort study in England (IPDLN 2024) 8- Implementing Transparency Standards: The ECHILD Project in Action (May 2024) • Knowledge sharing outputs/events: - ECHILD Seminar Series, online - ECHILD User Days, in-person - Houses of Parliament call for evidence on the SEND crisis, - Talks at the Children’s Commissioner, Department for Health, NIHR Applied Research Collaborations and Biomedical Research Centres (public and practitioners), UKRI UK Regenerative Medicine Platform, Medical Research Council (MRC), Institute for Clinical Evaluation Services, UKRI SPF Programmes, National Foundation For Educational Research, BHF Data Science Centre, Children’s Rights Conference, ADRUK and HDRUK Conferences and Symposiums. • Resources and tools: - ECHILD How-To Guides - INSTATS ECHILD Course - on Demand - ECHILD Phenotype Code List Repository - ECHILD Video and Illustrations - Maintained ECHILD Data Catalogue v2 - Maintained ECHILD User Guide • Funded Projects: - 7 ADRUK Funded Fellowships for 18 months each. - 33 Funding sources supporting UCL research projects - At least 6 External funding sources supporting sublicensees • Users: Currently, 45 internal users (UCL and collaborators) have access to ECHILD. The ONS SRS has only recently begun setting up access for external researchers, with 6 external users currently accessing the data via sublicensing. This number is expected to grow, as at least 14 more projects have already been approved by ONS RAS.

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

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

Objective for processing

BACKGROUND BACKGROUND: This application requests approval Under the previous iteration of this agreement (v3), UCL have requested to convert and extend the existing Education and Child Health Insight Linked Data (ECHILD) Database used in the University Collele College London (UCL) study 'Assessing the impact of the COVID-19 pandemic on vulnerable [20 words unchanged] children’s social care and this linkage is not currently supplied by NHS Digital England directly as it combines data from different sectors. [1 paragraph unchanged] UCL are requesting have requested to sublicense the ECHILD data to accredited researchers. UCL remain the sole [45 words unchanged] access to ECHILD data is via the ONS Secure Research Service (SRS). [1 paragraph unchanged] This amendment builds The previous iteration of this agreement (v3) was built on UCL’s existing approval for COVID-19 work (DARS-NIC-381972-Q5F0V - approved), which includes [138 words unchanged] researchers will only access minimised extracts required to answer specific research questions. [5 paragraphs unchanged] UCL require administrative data for all children in England (England only, not [47 words unchanged] UCL requested a transfer of identifying variables from NPD datasets to NHS Digital England to enable linkage to HES records. Linkage used names, date of birth [63 words unchanged] Pupil Matching Reference, even if the mother is not included in NPD. AMENDMENT AMENDMENT MADE UNDER PREVIOUS VERSION (V3) OF THE AGREEMENT: The current amendment requests approval Under the previous iteration of this agreement (v3) it has been requested to update the ECHILD database and to turn it into a Research [50 words unchanged] through separate data sharing agreements which will be subject to the NHS Digital England Data Access Request Service (DARS) approvals process. DfE and DHSC are applying for access to the ECHILD data for their own specified purposes. [7 paragraphs unchanged] LEGAL BASIS, ETHICS AND RISK OF POTENTIAL HARM HARM: UCL will be sole Data Controller under this agreement. UCL will be sole Data Controller under this agreement. The legal basis for processing personal data for this purpose at UCL falls under Article 6(1)(e) of the General Data Protection Regulations (GDPR), i.e. a task carried out in the public interest. It also falls under Article 9(2)(j), processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes. The legal basis for processing personal data for this purpose at UCL falls under Article 6(1)(e) of the General Data Protection Regulations (GDPR), i.e. a task carried out in the public interest. It also falls under Article 9(2)(j), processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes. [4 paragraphs unchanged] i. The data will be pseudonymised prior to dissemination by NHS Digital England to the data recipient; ii. The data recipient’s technical and organisational measures to safeguard the data have been assessed and meet NHS Digital’s England’s acceptance criteria (see sections 2 and 5b of this application for further details); [2 paragraphs unchanged] (c) 'is in the public interest'. NHS Digital England is content that the information set out in the Benefits section of the application evidences that the data processing will be in the public interest. [9 paragraphs unchanged] OPERATIONAL MANAGEMENT MANAGEMENT: [6 paragraphs unchanged] SUB-LICENCING SUB-LICENCING: ECHILD includes linked data from health, education and children’s social care and this linkage is not currently supplied by NHS Digital England directly as it combines data from different sectors. Significant value has therefore been added to the data prior to sub-licencing. NHS Digital’s England's Controller requirements are mirrored in the sub-licencing agreement that researchers will sign with UCL. [1 paragraph unchanged] UCL will be the data controller of the de-identified data collected within [40 words unchanged] care system (controlled through the Data Sharing Agreement between UCL and NHS Digital). England). In this sharing model of the linked data, ONS will be a [86 words unchanged] through a Data Access Agreement between UCL and approved researchers’ institutions). NHS Digital England requirement to audit data use is a requirement in the onward sharing contract and remains feasible in the ECHILD sub-license framework. In line with this onward sharing model, the data sharing controls in place between NHS Digital England and UCL are replicated between UCL and the other organisations. UCL is [15 words unchanged] The agreement mirrors the Data Sharing Framework Contract in place between NHS Digital England and UCL. It also requests information about the research proposal, benefits to [14 words unchanged] of data, responsibilities and processing activities etc., including a GDPR legal basis. NHS Digital England will retain the ability to directly audit UCL’s compliance with the outlined and agreed data access arrangements. [2 paragraphs unchanged] In the event of the termination or expiry of the Data Sharing Framework Contract between UCL and NHS Digital, England, all sub-licenses shall automatically terminate. ORGANISATIONAL AGREEMENTS AGREEMENTS: [10 paragraphs unchanged] -An agreement (this DSA) between NHS Digital England and UCL to onwardly share linked ECHILD data under the sub-licensing model, [24 words unchanged] actions of the parties involved in subsequent access to the linked data. [3 paragraphs unchanged] If the data sharing agreement between NHS Digital England and UCL were to cease, the license agreement between UCL and the licensee organisation will be terminated. [18 paragraphs unchanged] • Does exposure to different aspects of children's social care modify the associations between childhood adversity and later health and educational outcomes? child • How does parental health and contact with social care services affect childhood health and educa

Benefits reported

Analyses resulting from the original research question on the impact of COVID-19 on vulnerable children and individuals have been shared with and used to inform strategies by DHSC, PHE and NHS England, and with DfE through the ECHILD project and a project advisory group. The expected outputs and benefits outlined in our original application have not been fully achieved as UCL have not yet received the data that were requested under the DSA (the data have been released but have not yet been ingested to the SRS). UCL have not yet released any data through our sublicense. UCL are therefore requesting a 12 month extension to our DSA in order to have enough time to receive and process the requested data, and to demonstrate benefits. Since June 2021, when linked ECHILD data could first be accessed, the researchers have published the following research findings: In addition to the yielded benefits outlined in the previous iteration of this agreement (signed in May 2023), UCL have continued to work on the data previously released under our DSA, and have achieved the following outputs/benefits: i) UCL showed that the quality of data linkage between schools and hospital data was good and improved over time. However, those not linked were disproportionately poor or from certain ethnic groups. This information can be used to reduce biases in linkage and in analyses. (please see the "Linking education and hospital data in England: linkage process and quality" paper published in IJPDS Vol. 6 No. 1 (2021), available: https://doi.org/10.23889/ijpds.v6i1.1671 and "Ethnic bias in data linkage" correspondence in The Lancet Digital Health, Vol 3, Issue 6, E339, (2021), available: https://doi.org/10.1016/S2589-7500(21)00081-9). This evaluation has informed other linkages between health and education data in order to improve the quality of linkage of cross-sectoral data and improve the quality of data being used to generate benefits for health. 1. ECHILD has been highlighted as a key initiative to improve access to and use of data to improve health, including in the early years in the recent Academy of Medical Sciences report on “Prioritising early childhood to promote the nations’ health, wellbeing and prosperity”: https://acmedsci.ac.uk/file-download/16927511 ii) Analyses showed that children who were vulnerable, due to contact with social services or because they received special educational needs support for additional learning needs, had a much greater deficit in hospital care during the COVID pandemic than their peers. Findings have generated evidence on the need for targeted ‘catchup’ funding and resources for child health, particularly for vulnerable children who were affected disproportionately. For example, the ring-fenced resource for ‘catch-up’ of NHS care (Health and Social Care Levy) might be further targeted for the vulnerable groups that have disproportionally missed out on hospital contacts. Secondary school pupils receiving special educational needs support or social care services may need to be prioritised for face-to-face outpatient care as it is unclear how effective remote care is for these children (Please see report: “Changes in hospital contacts during the COVID-19 pandemic among vulnerable children and young people”, 4th November 2021. available: https://www.ucl.ac.uk/children-policy-research/projects/assessing-impact-covid-19-pandemic-vulnerable-children). 2. ECHILD was mentioned in the parliament at the Preterm Birth Committee 2024 (see 15:20 https://www.parliamentlive.tv/Event/Index/bc47da21-c39c-44d8-9dfe-f9cb61e1a6d8?_gl=1*1wnshlm*_ga*MTUwNzQwMTk1My4xNzA3NzU4MTA1*_ga_L0NJWDWMGN*MTcwNzc1ODEwNC4xLjEuMTcwNzc1ODEwOC41Ni4wLjA) iii) Analysis showed that for children aged between 0 to 4 years, deficits in hospital care during the pandemic were much higher for clinically vulnerable children than peers. 1 in 6 clinically vulnerable accounted for one-third to one half of the deficit in hospital care during the pandemic. During the pandemic, weekly rates of planned care returned to pre-pandemic levels for infants with chronic conditions but not older children. Deficits in care differed by ethnic group and level of deprivation. These findings may be used to inform health services on who should be targeted for additional care as the pandemic slows. “Deficits in hospital care among clinically vulnerable children aged 0 to 4 years during the COVID-19 pandemic” accepted for publication, 17th December 2021 pre-print available https://www.medrxiv.org/content/10.1101/2021.12.16.21267904v1. December 2021. 3. ADRUK have funded three fellowships to use ECHILD. iv) Methodology paper showed how the ECHILD administrative data may be used in lieu of randomised controlled trials, through target trial emulation methods, to answer causal questions whilst reducing confounding and other biases likely to arise with such data. Methods focus on trial emulation to understand the impact of SEN support from the start of school on unplanned hospital utilisation in children with cleft lip and palate. This work may inform future studies using linked administrative data to generate benefits for health. “SEN support from the start of school and its impact on unplanned hospital utilisation in children with cleft lip and palate: a demonstration target trial emulation protocol using ECHILD” 5th April 2022 preprint available https://www.medrxiv.org/content/10.1101/2022.04.01.22273280v1 (DOI: https://doi.org/10.1101/2022.04.01.22273280). 4. ADRUK currently have a call out advertising for new fellowships using ECHILD: https://www.ukri.org/opportunity/adr-uk-research-fellowships-2024/#:~:text=Administrative%20Data%20Research%20UK%20(ADR,organisation%20eligible%20for%20ESRC%20funding. v) Systematic review and thematic analysis of administrative data research showed the characteristics of research-ready administrative data to define a common understanding of what constitutes research-ready administrative data. In turn, the analysis helps data owners and researchers develop common principles and standards to establish clear principles and frameworks for data’s development and the realisation of admin data’s full research potential. This work may inform future studies using linked administrative data to generate benefits for health. “What makes administrative data research-ready? A systematic review and thematic analysis of published literature” 27th April 2022 available https://ijpds.org/article/view/1718 (DOI:https://doi.org/10.23889/ijpds.v7i1.1718). 5. ADRUK have classified ECHILD as one of their flagship datasets (https://www.adruk.org/fileadmin/uploads/adruk/Documents/ADR-England-flagship-dataset-brochure.pdf) and have funded three fellowships to work on ECHILD through this funding call: https://www.ukri.org/opportunity/adr-uk-research-fellowships-2023/ vi) Analysis showed that those who were receiving special educational needs (SEN) support or children’s social care (CSC) services had greater decreases in planned hospital care than their peers during the COVID-19 pandemic. Those receiving SEN support or CSC services bore a proportionally larger decrease in outpatient attendances, planned hospital admissions and were also less likely than their peers to have face-to-face outpatient care during the pandemic. The large decreases young people experienced during the pandemic could mean that some health needs have gone unmet which may have long-lasting impacts on their health and well-being. These findings could be used to inform health services on who should be targeted for additional care as the pandemic slows. (Please see publication: “Changes in adolescents’ planned hospital care during the COVID-19 pandemic: analysis of linked administrative data” 16th May 2022. available https://adc.bmj.com/content/early/2022/05/15/archdischild-2021-323616 (DOI: http://dx.doi.org/10.1136/archdischild-2021-323616). 6. ECHILD has been highlighted as a key data resource in a 2024 NIHR Programme Development Grant call (see 56:30 https://youtu.be/NO-7AGgxTtg) vii) Children born even a few weeks too early are less likely to achieve expected levels of attainment at age 7 and 11 and are more likely to have Special Educational Needs provision than those born at 40 weeks of gestation. This association is not fully explained by maternal risk factors including deprivation, age and parity, or by size-for-gestation at birth. Chronic conditions in school-aged children contribute more to the burden of Special Educational Needs and low academic attainment than preterm birth. Additional support prior to school entry to improve school readiness could be targeted at high-risk groups based on early health indicators shown to influence later outcomes. This evidence could be used to inform strategies for allowing for the disadvantage that children born preterm experience as they start school, e.g. by allowing parents to make decisions about whether to delay school entry. “Gestational age at birth, chronic conditions and school outcomes: a population-based data linkage study of children born in England”, 19 May 2022. Available https://academic.oup.com/ije/advance-article/doi/10.1093/ije/dyac105/6589377?login=false (DOI: https://doi.org/10.1093/ije/dyac105) 7. As part of a Research Community Catalyst for Children at Risk of Poor Outcomes, ADRUK have funded an ECHILD researcher to analyse children’s social care data (https://www.adruk.org/news-publications/news-blogs/two-new-projects-will-develop-research-communities-and-drive-transformative-insights-about-children-and-young-people/) viii) UCL have authored the Data Resource Profile: The Education and Child Health Insights from Linked Data (ECHILD) Database. This resource helps researchers interested in the ECHILD Database, explaining what information the database contains and discusses ECHILD’s key strengths and limitations for research. Data Resource Profile available: https://academic.oup.com/ije/advance-article/doi/10.1093/ije/dyab149/6425590 (DOI: https://doi.org/10.1093/ije/dyab149). November 2021. 8. UCL have published a protocol in BMJ Open for the HOPE study, which aims to evaluate special educational needs provision and its impact on health and education outcomes: Evaluation of variation in special educational needs provision and its impact on health and education using administrative records for England: umbrella protocol for a mixed-methods research programme. The HOPE study aims to build the evidence base for fairer and more effective SEN provision and, by informing national and local policy and the public and changing practice, to improve health and education outcomes of children with SEN. ix) UCL have additionally been maintaining an ECHILD user guide, data catalogue, analysis code and PPIE findings to support for wider use. The ECHILD user guide may be downloaded here: The Education and Child Health Insights from Linked Data (ECHILD) Database - An Introductory Guide for Researchers - v1.1.1 Updated February 2022, with PPIE findings published here: Echild website / Engaging the public, this additionally includes ECHILD Stakeholder Event Report. The event acted as an opportunity for stakeholders to ask questions about and share their views on the ECHILD Database and its use in future research for the public benefit. Report from the event highlights the key messages from government and stakeholders regarding the ECHILD Database, available: https://www.adruk.org/news-publications/news-blogs/the-potential-of-linked-administrative-data-for-understanding-the-relationships-between-childrens-health-and-education-434/ . Lastly, UCL’s ECHILD and reusable HES coding (such as R and Stata code, as well as code lists on Github: here https://github.com/UCL-CHIG. 9. UCL published a paper on cleft lip and palate repair surgeries before and during COVID: Number and timing of primary cleft lip palate repair surgeries in England: whole nation study of electronic health records before and during the COVID-19 pandemic. This paper showed significant reductions in the number and delays in timing of first primary CLP repair procedures in England during the first year of the pandemic, which may affect long-term outcomes. Other benefits from using the ECHILD data can be found on the website - https://www.ucl.ac.uk/child-health/research/population-policy-and-practice-research-and-teaching-department/cenb-clinical-22 10. UCL published a paper on hospital admissions for stress-related presentations among school-aged adolescents: Hospital admissions for stress-related presentations among school-aged adolescents during term time versus holidays in England: weekly time series and retrospective cross-sectional analysis. This paper showed that hospital admissions for stress-related presentations are common, affecting around two girls and one boy in every classroom in England. 11. UCL published an abstract on the educational outcomes of children with chronic liver disease and presented this as the RCPCH conference: Educational outcomes in children with chronic liver disease in England are inferior to peers: evidence from 5 million children. This analysis can be used to inform educational and health services policies of the urgent need for neuro-developmental assessment to be included in the routine care for children with chronic liver disease, to ensure early detection and referral to specialist services. 12. UCL published an abstract on special educational needs of primary school aged children with neurodevelopmental conditions and presented this work at the Society for Social Medicine conference: Special educational needs of primary school aged children with neurodevelopmental conditions: a population cohort study using linked health and education records. This analysis showed that children with neurodevelopmental conditions have varying levels of SEN provision, with more intensive provision increasing and less intensive provision decreasing during primary school. 13. UCL published an abstract on self-harm hospitalisations in secondary school pupils in England and presented this at the RCPCH and YPHSIG Adolescent Health conference: Quantifying emergency hospital admissions with self-harm in secondary school pupils in England: whole nation study of linked data from health and education. This analysis showed that Individually targeted interventions (e.g. to girls with a history of self-harm) may be more effective than universal strategies for reducing self-harm admissions. 14. UCL published a study protocol for a target trial emulation study of the impact of special educational needs provision on hospital utilisation for children according to gestational age at birth: Impact of special educational needs provision on hospital utilisation, school attainment and absences for children in English primary schools stratified by gestational age at birth: A target trial emulation study protocol. This work will inform whether SEN provision can improve educational and health outcomes. 15. UCL published a study protocol for a study of the impact of special educational needs on children with cleft lip and/or palate: Early special educational needs provision and its impact on unplanned hospital utilisation and school absences in children with isolated cleft lip and/or palate: a demonstration target trial emulation study protocol using ECHILD. This analysis will inform whether reasonable adjustments at the start of compulsory education can improve health and educational outcomes in the cleft lip and palate population. 16. UCL published an abstract on special educational needs and school readiness in children with neurodevelopmental conditions and presented this at the British Academy of Childhood Disability conference: Special educational needs and school readiness of children with neurodevelopmental conditions: a national cohort using linked health and education records. This analysis showed that the small proportion of children with neurodevelopmental conditions account for 22% of Education and Health Care Plans. 17. UCL published a study protocol for a study on local authority variation in primary school-recorded special educational need provision in children with major congenital anomalies: Local authority variation in primary school-recorded special educational needs provision among children with major congenital anomalies: A research protocol. This analysis will help UCL to understand how equal special educational needs provision is across England and inform policies to best support children. 18. UCL have been maintaining an ECHILD user guide, data catalogue and analysis code, and stakeholder work to support wider use: o ECHILD User Guide V2: https://www.ucl.ac.uk/child-health/sites/child_health/files/echild_user_guide_v2.pdf (updated June 2023) o ECHILD Data Catalogue: https://www.ucl.ac.uk/child-health/sites/child_health/files/echild_datacatalogue_v2.1_202310.xlsx (updated October 2023) o Github for reusable code: https://github.com/UCL-CHIG

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

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

Datasets: + Civil Registration - Births

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

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

Datasets: + Birth Notification Data; + Community Services Data Set (CSDS); + MSDS (Maternity Services Data Set) v1.5; + MSDS (Maternity Services Data Set) v2.0; + Mental Health Minimum Data Set (MHMDS); + Mental Health Services Data Set (MHSDS); + Mental Health Services Data Set (MHSDS) v5.0; + Mental Health and Learning Disabilities Data Set (MHLDDS) · − HES-ID to MPS-ID HES Admitted Patient Care; − HES:Civil Registration (Deaths) bridge

Objective for processing

Version 1 of this agreement is for the same purpose as version 0 but will be for all young people born in England on or after 01.09.1984 (rather than 01.09.1995). The following detail outlines further detail for this change: BACKGROUND To enhance information on vulnerable children, UCL request an amendment to transfer to ONS SRS a pseudonymised mother and baby/babies linkage flag, attached to the relevant pseudonymised HES record for mother and baby/babies (this is pre-linked by NHS Digital). To enable inclusion of maternal characteristics (such as age), and to follow up the health of young people born on or after 01.09.1984; UCL request to extend the age range to all individuals in England born on or after 01.09.1984, from the current age limit of 01.09.1995. This application requests approval to convert and extend the existing Education and Child Health Insight Linked Data (ECHILD) Database used in the University Collele London (UCL) study 'Assessing the impact of the COVID-19 pandemic on vulnerable children’ to a Research Database for wider use, through a sub-licencing model. ECHILD includes linked data from health, education and children’s social care and this linkage is not currently supplied by NHS Digital directly as it combines data from different sectors. This project aims to improve understanding of the effects of the COVID-19 lockdown and restricted access to schools and health care on outcomes for children and young people. UCL require data over the child and adult life course because they will be taking a longitudinal perspective across life. UCL will compare how exposure to vulnerability during childhood and youth, in relation to health conditions, school factors (e.g. special needs, exclusion) and social care influence health outcomes during childhood and adulthood. These relationships need to be assessed in cohorts followed through the child to adult life course before the pandemic, and compare with similar cohorts who experienced deficits in health and social care and education, during the pandemic. In this way, UCL can estimate potential impacts of these deficits in care that might be attributable to the pandemic. ECHILD is currently supported by funding from the National Institute of Health Research (NIHR) Policy Research Unit for Children, Young People and Families (CPRU), within University College London (UCL). CPRU works closely with the Department of Health and Social Care to determine priorities and provide evidence directly to the Secretary of State for Health, government departments and arms-length bodies, such as NHS England, the Office for Health Improvement and Disparities, and the UK Health Security Agency. Additional funding support is provided by Administrative Data Research UK, Health Data Research UK and NIHR. Extension of the age range and the mum-baby link will allow evaluation of exposures and outcomes in adulthood, including of children who themselves become parents. Extension of the age range will enable linkage of adolescents with education records to their health records as adults up to the oldest age. Researchers will examine vulnerability indicators in the mother (e.g., maternal history of social care, SEND support, or school exclusion in her childhood, and characteristics of a child’s mother before or after delivery, such as maternal age at delivery, chronic mental or physical conditions, previous teenage motherhood) on child health outcomes. These linkages will be re-run for all children and extended to incorporate linkage to social care data held by DfE, and to the Emergency Care Data Set (ECDS) held by NHS Digital. The mum-baby link is already approved for use in NIC-393510-D6H1D. and is pre-linked by NHS Digital. UCL are requesting to sublicense the ECHILD data to accredited researchers. UCL remain the sole Data Controller who also process data. The London School of Hygiene and Tropical Medicine (LSHTM), the Office for National Statistics (ONS) and The Institute for Fiscal Studies (IFS) are also listed as data processors and will continue to process the data under this agreement. All access to ECHILD data is via the ONS Secure Research Service (SRS). UCL require administrative data for all children in England who appear in the specified NPD and HES datasets to create longitudinal cohorts of children born on or after 1.9.1984 - 95. UCL request a transfer of identifying variables from NPD datasets to NHS Digital to enable linkage to HES records. Initially, linkage between names, date of birth and postcodes will be via the Personal Demographic Service (PDS) and then, using NHS number from PDS to hospital episode statistics (HES). The output will be pseudonymised linkage keys (following the process used in the ECHILD project). UCL have previously demonstrated high quality linkage of 92% of NPD records to HES for individuals born in 1990/01 (NIC-27404-D5Z3F). Researchers will evaluate linkage bias, but anticipate acceptable linkage rates from 1984/5. HES histories will be included for any mother linked through HES to a child with an anonymised PMR, even if the mother is not included in NPD. UCL has been using ECHILD to understand the impact of COVID-19 and lockdown on children and individuals under the age of 35 years at the time of the COVID-19 pandemic. Children or individuals who are vulnerable due to social welfare or chronic health needs are expected to experience more adverse health and social effects of the COVID-19 lockdown than their peers. Key concerns for services are the effects of household confinement during the COVID-19 lockdown, combined with the limited access to support from health, social care and education services. The data processors have analysed the data to understand what impacts COVID-19 infection and related public health responses (such as lockdown) have had on children and individuals, including those that are deemed vulnerable, to inform strategies for the current wave of infection, and any future waves. Clarifications of funding sources and outputs to better describe the development of vulnerability cohorts before and after COVID are also included in this amendment. This amendment builds on UCL’s existing approval for COVID-19 work (DARS-NIC-381972-Q5F0V - approved), which includes data on approximately 14 million children and the previous Education and Child Health Insight Linked Data (ECHILD) project (DARS-NIC-27404-D5Z3F - approved), which used linked education and HES data for four one-year cohorts. The linkage under this application will be extended beyond the current specific purpose relating to COVID-19, to address a range of research questions aiming to generate improvements to the health and social care system, through a better understanding of the relationship between education and long term health outcomes. All children and individuals appearing in HES records from (the latest of) birth or April 1997 onwards and born since 01/09/1984 will be included, to allow researchers to evaluate how exposures in childhood and at school age are related to health outcomes in childhood and adulthood. The Research Database will hold information on approximately 20 million individuals, but researchers will only access minimised extracts required to answer specific research questions. The following detail from version 0 has been updated to reflect the above change: PURPOSE The data is requested for a programme of research relevant to the aims of the of the National Institute of Health Research Policy Research Unit for Children, Young People and Families (CPRU), within University College London (UCL). ECHILD aims to improve understanding of the relationship between child health, child development, and contact with social care services. UCL require data over the child and adult life course because they are taking a longitudinal perspective across life. Long-term follow-up is required as it is known that exposures in early life (such as entry into care, or early disability such as extreme prematurity at birth) can have lifelong consequences for health. In addition, parental exposures, related to child maltreatment, poverty, or poor mental or physical health, and school factors such as special needs, influence the outcomes of children into adulthood. CPRU is one of 15 NIHR Policy Research Units formed to undertake research to inform decision-making by government and arms-length bodies. CPRU works closely with the Department of Health and Social Care to determine priorities and provide evidence directly to the Secretary of State for Health, government departments and arms-length bodies, such as NHS England and Public Health England. Additional funding support is provided by Administrative Data Research UK, Health Data Research UK, and NIHR. The agreement is to also allow UCL to enhance information in ECHILD by including a pseudonymised mother-baby linkage flag, attached to the relevant pseudonymised HES record for mother or baby. To enable inclusion of maternal characteristics (such as age), and to follow up the health of young people to their 30s, UCL requested to extend the age range to all young people in England born on or after 01.09.1984. For this programme of research, UCL are the sole Data Controller who also process data. The London School of Hygiene and Tropical Medicine (LSHTM), the Office for National Statistics (ONS) and The Institute for Fiscal Studies (IFS) are also data processors. Extension of the age range and the mother-baby link allows evaluation of exposures and outcomes in early adulthood, including of children who themselves become parents. It enables linkage of adolescent education records to health records up to age 38 years in 2022. Researchers will also examine maternal characteristics (such as maternal age, chronic mental or physical conditions, previous birth), using the pseudonymised mother-child flag to assess the influence of risk factors (e.g., maternal history of social care, special educational needs and disabilities (SEND) support, or school exclusion) on health outcomes in her child. It is hoped findings will be relevant to policies receiving major investment such as ‘Start for Life’. The study is looking at the impact of COVID-19 and lockdown on Children and individuals under the age of 35 years at the start time of the COVID-19 pandemic. Children or individuals who are vulnerable due to social welfare or chronic health needs are expected to experience more adverse health and social effects of the COVID-19 lockdown than their peers. The Department for Education (DfE) and the Department of Health and Social Care (DHSC) will also access the ECHILD asset under their own Data Sharing Agreements (DSAs) for their own purposes, which are beyond the remit of the UCL agreement. The DfE and DHSC DSAs will have DfE/DHSC as the data controller who also process the data along with ONS as the data processor. Key concerns for services are the effects of household confinement during the COVID-19 lockdown, combined with the limited access to support from health, social care and education services. The researchers urgently need to understand what impacts COVID-19 infection and related public health responses (such as lockdown) have had on children and individuals, including those that are deemed vulnerable, to inform strategies for the current wave of infection, and any future waves. UCL require administrative data for all children in England (England only, not including Wales) who appear in the specified national pupil database (NPD) (the NPD does not include all childrens data e.g. it doesnt hold date on children who are home schooled) and Hospital Episodes Statistics (HES) datasets to create longitudinal cohorts of children born on or after 1.9.1984. UCL requested a transfer of identifying variables from NPD datasets to NHS Digital to enable linkage to HES records. Linkage used names, date of birth and postcodes to link to NHS number. The output is pseudonymised linkage keys. UCL have previously demonstrated high quality linkage of 92% of NPD records to HES for individuals born in 1990/01 (NIC 27404). Researchers will evaluate linkage bias, but anticipate acceptable linkage rates from 1984/5. HES histories will be included for any mother linked through HES to a child with an anonymised Pupil Matching Reference, even if the mother is not included in NPD. This study; Education and Child Health Insights from Linked Data - COVID (ECHILD-COVID) builds on the Education and Child Health Insight Linked Data (ECHILD) project (DARS-NIC-27404-D5Z3F) which uses linked education and HES data for four one-year cohorts amounting to two million children and young people in England. The linkage under this agreement will be extended urgently to address the impact of COVID-19 on all children and individuals who are aged between 0 and 34 years in the COVID-pandemic year (linkage of 15 million individuals). The subject of the research will include all children and individuals appearing in HES records from (the latest of) birth or April 1997 onwards, who are aged between 0 and 34 years in the COVID-pandemic year (hence start date for birth is the start of school year 1.9.1984). AMENDMENT ECHILD-COVID addresses four priority areas raised by the Department of Health and Social Care (DHSC) with the Children’s Policy Research Unit (CPRU) team relating to the secondary impacts of infection and lockdown on: The current amendment requests approval to update the ECHILD database and to turn it into a Research Database so that minimised extracts of the data can be sub-licensed to other accredited researchers for research aimed at generating improvements to health and social care services. The Department for Education (DfE) and Department of Health and Social Care (DHSC) also intend to access the data on the ONS SRS through separate data sharing agreements which will be subject to the NHS Digital Data Access Request Service (DARS) approvals process. DfE and DHSC are applying for access to the ECHILD data for their own specified purposes. ~ Children who need safeguarding The ECHILD Research Database will support a rich portfolio of research projects examining different aspects of child health, whilst ensuring data minimisation suitable for the requirements of specific research questions. In other words, researchers will be able to apply for an extract of ECHILD data and will need to justify the years of the data and the specific data modules that they require in order to answer their research question. ~ low income families For example, a researcher aiming to generate evidence about school attainment following liver transplantation may request HES APC (to identify the cohort of liver transplant patients and a comparator group), and the School Census, Key Stage 2 attainment, and Absences data from NPD (to evaluate outcomes), but would not necessarily require any information on Children’s Social care, Maternity Services data, Community Services data, etc. The study period would also need to be specified, for example HES APC data from 1997-2008 capturing children with liver transplantation would allow follow up at age 11 for children captured in NPD data from 2008-2019. ~ Children with special educational needs The minimised extract would need to be justified for each project, based on the years of the data and the specific data modules, in a data access application to the ONS Research Accreditation Panel and approved by a UCL Data Access Committee (which will include lay members). All data will be stored and accessed securely on the ONS Secure Research Service. No re-identification of individuals will be allowed, and all outputs will be checked for statistical disclosure control before being released. ~ health inequalities Creating the Research Database to support a range of research projects is in the interests of data protection, as there will be less need for repeated transfers of personally identifiable information for multiple projects across different institutions. The researchers will draw on the published DfE definition for vulnerable individuals. This relates to children assessed as being in need under section 17 of the Children Act 1989 (i.e. have records indicating contact with social care services, or are being looked after) special educational needs or additional needs (such as prolonged absences or school exclusion). Children or individuals assessed as high risk by educational providers or local authorities will also be considered vulnerable (e.g. children on the edge of receiving support or those at risk of becoming not in employment, education or training). Recognising the value of the linked health and education data that have been linked as part of the existing study, there have been urgent calls to open up the ECHILD resource to more researchers, and for a wider range of research purposes. The strong interrelationship between health and education services in relation to the health and wellbeing of children is recognised by policy makers, but evidence is lacking on how services complement or compensate for each other and there have been calls for a stronger evidence base to be developed. It is imperative that government and researchers work together to fill this evidence gap – to improve the health, wellbeing, education, and safety of children, young people, and families, particularly the most vulnerable. The ECHILD Research Database will do this by using data to generate a comprehensive view of the journey through childhood to adulthood. This research database will be used to understand how trajectories of health, education and social care vary across children’s lives, and what works to improve the design and delivery of policies and systems which better meet the needs of children and young people. The ECHILD Research Database will fill this gap in evidence by facilitating research that will inform policy-makers and service commissioners about the associations between education risk factors and health outcomes. The vulnerable cohort is defined by their exposure during childhood and early adulthood (up to the age 24). ECHILD is focused on the exposure to vulnerability for children and youths and the outcomes related to that exposure. The age groups of those aged 25-37 will allow follow up on vulnerable and non-vulnerable young people into adulthood to assess differences in health outcomes across these groups. Outcomes include health, education and social care outcomes in children and young people and health outcomes measured in adulthood. The period from age 25 onwards will examine outcomes for these groups in adulthood. UCL are also requesting linkage to a broader range of health data because there is a wide range of potential research purposes. For example, linkage to mental health records will generate evidence on how schools can promote positive mental health in adolescence; linkage to health visiting activity in the Community Services Dataset will enable evaluation of how different levels of health visiting can improve health outcomes for families in contact with social care services. For the first time in England, the database will allow researchers to investigate long-term outcomes for a wide range of health conditions and treatments during childhood and early adulthood, alongside school attainment, absences, special needs support and exclusions, and social care support. UCL draw on work on definitions of vulnerability by DfE (as above) and Public Health England, and consider three broad groups: ECHILD could also be used to investigate health benefits (or harms) of education practices. For example, to find out whether providing support for special educational needs for children with chronic health conditions improves health outcomes or reduces use of hospital services. Understanding the relationships between services provided within health and education will form the basis of evidence-based policy making at DHSC (and DfE) to ensure children in all settings, in all areas of the country, are healthy, safe and develop their potential. Linkage to the maternal HES record will allow inclusion of maternal chronic health conditions and demographic risk factors (maternal age, ethnicity, age at first birth) to improve understanding of health conditions within families, to guide healthcare support for families and better understanding of intergenerational adversity. In addition, for young mothers (<27y), it will be possible to assess risk factors recorded from school and social care, such as school attainment, exclusions, and previous care placement to inform early interventions before and during pregnancy. - clinically vulnerable children and young adults (those with chronic mental or physical health conditions) LEGAL BASIS, ETHICS AND RISK OF POTENTIAL HARM - socially vulnerable (those receiving statutory support from social care as a child in need, or from education services, for example as special educational needs support or pupil referral unit) UCL will be sole Data Controller under this agreement. - those at high risk of being vulnerable due to social circumstances but not known to be receiving state support (eg: referred to social care but not receiving services (i.e. not a child in need); receiving free school meals, living in a deprived neighbourhood, high risk of becoming not employed, in education or training (NEET)). The legal basis for processing personal data for this purpose at UCL falls under Article 6(1)(e) of the General Data Protection Regulations (GDPR), i.e. a task carried out in the public interest. It also falls under Article 9(2)(j), processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes. The researchers also explore whether children with long-term health conditions such as asthma or poor mental health, and those allocated any special educational needs (as indicators of underlying health or behavioural problems), are at greater risk of adverse impacts of infection or lockdown. The data are required for research purposes in the public interest – meeting the conditions in the DPA 2018 Schedule 1 Part 1 (4) – which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data. The researcher will focus on two specific research questions: DPA 2018 Schedule 1 Part 1 (4) - which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data. The Data Protection Act 2018 Schedule 1 Part 1 states that when processing special category data under the condition relating to research, the condition is met when: RQ1: What are the differences in emergency hospital contacts during the COVID-19 pandemic for vulnerable children and individuals compared with other children and individuals? Is there any evidence that differences are related to COVID-19 infection or the secondary effects of lockdown? (a) 'necessary for archiving purposes, scientific or historical research purposes or statistical purposes'. As described above the data is considered necessary for the performance of the task RQ2: What is the predicted deferred health care use and what are the long-term health, education and social care outcomes due to restrictions during the COVID-19 pandemic? (b) 'carried out in accordance with Article 89(1) of the GDPR'. In accordance with this article, processing is subject to appropriate safeguards. These include: Addressing these questions requires understanding of the causal effects of vulnerability status on outcomes pre- and post- COVID-19. i. The data will be pseudonymised prior to dissemination by NHS Digital to the data recipient; The researcher will therefore: develop phenotypes and coding clusters to define vulnerable groups; evaluate relationships between vulnerability groups across the life course; and evaluate health, education and social care outcomes in comparator cohorts before and after the onset of COVID-19. Vulnerable cohorts are defined by exposure during childhood and youth. Health, education and social care outcomes can be measured before age 25. The period from age 25 onwards will examine health outcomes in adulthood for cohorts with and without exposure to vulnerability during childhood and youth. The mid- to long-term impact of delays, or withdrawal of healthcare, education and social care during the COVID pandemic (the deficit in care) require information on expected outcomes without such delays from similar cohorts followed up pre-COVID (i.e. what would have happened had COVID not occurred?). ii. The data recipient’s technical and organisational measures to safeguard the data have been assessed and meet NHS Digital’s acceptance criteria (see sections 2 and 5b of this application for further details); As vulnerable children or individuals are hard to identify in healthcare records, the researcher will use administrative data histories of ever being a Child in Need (CiN), having special educational needs (SEN), frequent absences, a chronic health condition requiring hospitalisation, or combinations of these exposures. The researcher will derive these vulnerability indicators from the linked longitudinal ECHILD dataset. Examining health data across the life course is critical for identifying markers of vulnerability and health outcomes in childhood and adulthood. For example, previous work completed by UCL has shown that chronic underlying conditions, or congenital disorders associated with special education needs may not be recorded at every admission (e.g., asthma may not be recorded when a child is admitted for an operation) and UCL have demonstrated the added value of using the whole longitudinal record. iii. The requested data has been assessed as proportionate to the aim pursued (see section 5a of this application for further details); Previous birth contributes to vulnerability status because teenage motherhood at first live-birth is recognised as a social risk factor, which influences a child’s health outcomes, even for subsequent children born to older mothers. iv. Controls, data retention and processing activities have been assessed to ensure respect to the essence of the right to data protection (see sections 5a, 5b and 8a of this application for further details); Ethnic background will not be categorised as a component of any vulnerability indicator. UCL consider ethnic background as potentially influencing associations between vulnerability indicators and outcomes (i.e. ethnic background will be analysed as a confounder). For example, UCL will explore whether the proportion of children with vulnerability indicators or with adverse health outcomes, vary according to ethnic background, and if so, include ethnic background in statistical models that seek to determine the influence of vulnerability on child outcomes, after adjusting for ethnic background. (c) 'is in the public interest'. NHS Digital is content that the information set out in the Benefits section of the application evidences that the data processing will be in the public interest. To enable the analyses to address these research questions, the researcher will link HES data (i.e. HES APC, outpatient, critical care, A&E and ECDS data, plus death registration data) to administrative data contained in the datasets collectively supplied within the National Pupil Dataset (NPD), provided by DfE (the researcher refers to NPD data as education, CiN, and children looked after (CLA)). These datasets (HES-NPD) will be linked by NHS Digital for children and individuals in England who are aged between 0 and 34 years at the time of the COVID-19 pandemic using pseudonymised linkage keys. The processing of data for this study is a task of public interest as it will provide evidence on the relationship between health and education outcomes and use of healthcare services among children and individuals. This will benefit the provision of healthcare services by informing policy makers and service providers, and improving evidence-based information and interventions for children and their families. The legal basis for processing personal data for this purpose data at UCL falls under Article 6(1)(e) of the General Data Protection Regulations (GDPR), i.e. “a task carried out in the public interest”. It also falls under Article 9(2)(j), “processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes”. The processing of data for this study is a task of public interest as it will provide evidence on the effect of the COVID-19 pandemic on health outcomes and use of healthcare services among vulnerable children and individuals. This will benefit and inform policy makers, service providers, vulnerable children and their families. The case for this research being in the public interest is established through balancing the strong policy driver of improved understanding of the inter-relationships between education, health and social care, with consideration of potential harms to the patients/participants whose records are involved. Potential harms primarily relate to breach of confidentiality and the subsequent misuse of Personal Data or erosion of trust in longitudinal research/data science. All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract ie: employees, agents, PhD students and contractors of the Data Recipient who may have access to that data). The risk of this harm is minimised as this is: Patient and public information groups have been engaged with. 1. an observational population-level research database which will not result in a direct intervention to any participant; o The project has seen 8 engagements so far, with 2 more planned during 2022. 2. tried and tested IT infrastructure and governance frameworks specifically designed to minimise risks to privacy during health data science; o This includes advocacy and representative groups that were brought together for the ECHILD public stakeholder event on 29 April 2021. Amongst others, this included the Children’s Commissioner for England, NSPCC, SCOPE, NASEN, Contact, Council for Disabled Children, Down's Syndrome Association, MENCAP, GenerationR Alliance, Parentkind (PTA UK). 3. a fully de-identified research environment with sufficient controls that the risk of disclosure is not considered reasonably likely; o A nationwide survey of parents and carers of disabled children and young people is being finalised with SCOPE. Fieldwork is expected to begin in September. 4. National Opt-Out of the use of health data for research will also be respected 5. all staff and users are vetted and approved professional researchers operating within controlled and auditable conditions. This research has strong scientific rationale, the process is transparent, best endeavours will be made to inform participants of the use of their data within ECHILD, with a right to object, that risks are mitigated and the participants (and wider public) are likely to directly benefit from the research through improved health care and government policy provision. Participants will be informed about how their data is used through ongoing fair processing communications including details of the process for opting-out through the National Data Opt-Out process. OPERATIONAL MANAGEMENT UCL is the study sponsor and Data controller. ECHILD researchers at UCL have responsibility for the running of the ECHILD Research Database. The ECHILD Research Database will be configured to have two distinct classes of operational areas: 1. Data management and processing: All operational control and access for data processing and management of de-identified data, including but not limited to data processing and storage, and provision of data to approved users is restricted to UCL and ONS SRS staff. UCL and ONS staff are the only individuals who can access all data within ECHILD and sources in their raw and processed forms. 2. Onward sharing to research analysts: Secure operating partitions will be created for each approved ECHILD project. ECHILD approved users for the projects will have access to their folder, which will contain a sub-set of data relevant to that study where onward sharing conditions of data providers are met. ONS SRS are data processors who provide data infrastructure and the Secure Data Environment (SDE). UCL, IFS and LSHTM all access the data for specific research purposes under the current agreement. ONS SRS will assist with the data management and processing (under UCL’s direction) and will conduct the output disclosure assessments. Amazon Web Services (AWS) is a processor acting under the instructions of UCL. AWS’ role is limited to secure back-up of data stored in UCL’s Data Safe Haven. UCL uses offsite data centre services provided by VIRTUS data centre. VIRTUS does not have access to the data. SUB-LICENCING ECHILD includes linked data from health, education and children’s social care and this linkage is not currently supplied by NHS Digital directly as it combines data from different sectors. Significant value has therefore been added to the data prior to sub-licencing. NHS Digital’s Controller requirements are mirrored in the sub-licencing agreement that researchers will sign with UCL. For ECHILD to be a useful and sustainable resource for researchers to interrogate, and obtain meaningful data in a timely manner, contractual arrangements will need to facilitate high-volume and rapid turn-around of data requests. UCL expect 1-2 applications per month and the potential length of each sub-licence is 2-3 years in length. Sub-licencing will remove the burden of a high number of additional data sharing agreements across multiple institutions, which would impede the speed of research. Therefore, a contract structure is needed to control the data flows, to control the purposes and way these data are processed, and to enforce the governance requirements of the individual studies and data owners and the legal basis under which they are permitted to operate. This will allow efficient re-use of existing data through dissemination of minimised data extracts for research where the purpose is consistent with this DSA. UCL will be the data controller of the de-identified data collected within the ECHILD Research Database and stored on the ONS SRS, for the purpose of processing it, approving onward research use, and providing managed access to relevant sub-sets of data for purposes relating to generating benefits to the health and social care system (controlled through the Data Sharing Agreement between UCL and NHS Digital). In this sharing model of the linked data, ONS will be a data processor, as they will be processing the data on behalf of UCL. This includes holding the linked data in a secure environment, extraction of approved data and setting up access systems, and approving statistical outputs, following a statistical disclosure control procedure. The structure will enable UCL to determine the onward processing of the integrated data by reviewing applications from researchers for specific projects (with the ONS Research Accreditation Panel (RAP) and onwardly sharing sub-sets of relevant data within the ONS SRS to approved users (controlled through a Data Access Agreement between UCL and approved researchers’ institutions). NHS Digital requirement to audit data use is a requirement in the onward sharing contract and remains feasible in the ECHILD sub-license framework. In line with this onward sharing model, the data sharing controls in place between NHS Digital and UCL are replicated between UCL and the other organisations. UCL is fully accountable for the actions of the parties involved in subsequent data share and use. The agreement mirrors the Data Sharing Framework Contract in place between NHS Digital and UCL. It also requests information about the research proposal, benefits to health and/or social care, organisational security assurance and terms and conditions regarding onward sharing of data, responsibilities and processing activities etc., including a GDPR legal basis. NHS Digital will retain the ability to directly audit UCL’s compliance with the outlined and agreed data access arrangements. The anticipated volume / number of licences is 1-2 sub-licences per month, and the potential length of each sub-licence is 2-3 years in length. The territory of use in the sub-licence will be the same or narrower than the territory of use stated in this data sharing agreement, namely the United Kingdom. The approved organisations and researchers who are granted an access to the linked data via the ONS SRS, agree to terms and conditions of use, their rights and responsibilities as users of the linked data, as defined by the ONS SRS. In addition to the agreements signed with the ONS SRS, the organisation of the researcher applying to use the linked data will enter into a Licence agreement with UCL. In the event of the termination or expiry of the Data Sharing Framework Contract between UCL and NHS Digital, all sub-licenses shall automatically terminate. ORGANISATIONAL AGREEMENTS UCL will provide a sub-license to UK organisations undertaking research that will be of benefit to the public in England and Wales (this will be assessed in the project proposal form submitted to the ONS RAP and to UCL). Applicants (potential licensees) will need to show that the provision of the sub licensing will be in the public interest and that the data will be used either (i) for the provision of health care or adult social care; or (ii) for the promotion of health. The project proposal will be assessed to determine the details of the project, the people who will be accessing the data, and what data will be requested. Applicants will need to be accredited researchers or agree to undertake training and become accredited, prior to accessing the data. Additionally, an applicant's organisation will need to provide evidence that they have information governance and security assurances in place. Members of the ECHILD Data Access Committee will review and decide if the evidence provided satisfies the requirements of the ECHILD Data Access Agreement. The committee will comprise of the following members: - Co-Chairs: ECHILD leads - ECHLD Project Manager - ECHILD researcher - ECHILD Senior Data Scientist - ECHILD Senior Data Resource Manager - Lay members Commercial purposes, e.g. where an applicant intends to allow use of data for purposes such as marketing, sales or insurance or where there may be international transfers of data (potentially including the EEA post Brexit), will not be considered unless there is a strong case to show the public interest. Applicants (licensees) will have to sign two agreements to obtain a sub-license, one with the ONS SRS and another with UCL. In both cases the licensee will agree with the terms stated in the Confidentiality Section of the ECHILD Data Access Agreement and with the Confidentiality Terms stated in the Access Agreement which will be signed with the ONS SRS. By signing these agreements, the licensee agrees to adhere to these terms, including respecting the privacy of health services user data they will receive. Licensees are also reminded of the penalties they are likely to incur if they do not comply with the terms they have agreed. In addition to the above, the ONS SRS agreement stipulates that data users must complete mandatory training before they are allowed to access the data. To ensure the security of the linked information, UCL will put in place the following controls employed at the different steps of the process of depositing, approving and sharing of the linked information: -An agreement (this DSA) between NHS Digital and UCL to onwardly share linked ECHILD data under the sub-licensing model, which outlines the terms and conditions of use of the linked data via the ONS SRS, and the full accountability of UCL to the actions of the parties involved in subsequent access to the linked data. -An agreement between the ONS SRS and the approved researcher and organisation, which outlines the terms and conditions of use of the linked data in the ONS SRS (Accredited Research Assurance Registration Form). -An Agreement between UCL and the organisation requesting to use the linked data via the ONS SRS (the ECHILD Data Access Agreement), which outlines the terms and conditions of use of the linked data. This agreement specifies how any data breaches will be dealt with. The researcher accessing the data via the ONS SRS will not be able to download any record level pseudonymised data. Once the researcher has finished their research, ONS SRS will destroy the data folder with the tailored dataset for the specific project. Any outputs produced under sub-license will be subject to strict disclosure control methods with small numbers suppressed in line with the HES analysis guide. If the data sharing agreement between NHS Digital and UCL were to cease, the license agreement between UCL and the licensee organisation will be terminated. Sub-licensing activity is not currently addressed in the UCL DPIA, since the UCL DPIA only covers processing of data within UCL. The ONS are currently working on their own DPIA, which should cover the processing of data for sub-licensees. SUB LICENCE PURPOSES All data processed under the sub-licence will be completed using the same legal basis as mentioned above, namely GDPR (article 6(1)(e))and GDPR (article 9(2)(j)). The UCL Data Access Agreement will require licensees to provide the Legal Basis of their request to access ECHILD data and therefore ECHILD Data Access Committee will only grant approval to applications from researchers within public bodies who have a legal basis to process data under GDPR. ECHILD will only be used for research that has a clear public benefit in England and Wales relating to the provision of healthcare and to education. Analysis of ECHILD data will generate answers to questions about the interactions between education, children’s social care and health which have implications throughout the life course. ECHILD will be used for the following specific research themes. These themes, and the proposed research questions within each theme, have been collated from a number of potential research users, including researchers from: University of Bristol, Institute for Fiscal Studies, Applied Research Collobarations (ARCs) including major National Institute for Health and Care Research (NIHR) research groups working on applied health informatics (ARC West), and maternal and child health (PenARC, Exeter), Economists at the Centre for Health Economic, York University, and health researchers at Cardiff University, Swansea University, Imperial College London, and Kings College London. 1. INFORMING PREVENTATIVE STRATEGIES BY HEALTH CARE AND EDUCATION SERVICES The ECHILD dataset will facilitate research that will be used to inform health care and education services about whether certain types of schools or local authorities are associated with increased or decreased rates of hospital contacts for children with particular health conditions. The findings from such research will inform preventive strategies by local authorities, schools and healthcare that might reduce adverse outcomes for children and adolescents. For example, studies could explore whether children with serious learning impairing conditions may have lower rates of emergency hospital admission if they attend a special school than if they attend a mainstream school; school type or area may affect rates of emergency admissions and A&E attendance for adversity-related conditions (e.g. self-harm, violence or mental health), after adjusting for underlying chronic conditions, previous admissions, age and socioeconomic factors. The data will generate hypotheses about how interventions in schools, or improved feedback from hospitals, could improve the health and well-being of children and adolescents and thereby impact healthcare utilisation. Wider social determinants of health, such as education and adverse childhood experiences, are a key focus of ongoing work within DHSC and across government on a new strategy for health promotion, focused on the prevention of poor health. Understanding drivers of poor health at a young age is critical to achieving government ambitions to improve the health of the nation. Analysis of ECHILD data will generate answers to questions about the inter-relationships between education, children’s social care and health, which have implications throughout the life course. For example, many NIHR Applied Research Centres (ARCs), including ARC West, have research themes around improving health and addressing health inequality in vulnerable and disadvantaged children. A national data resource linking administrative records across secondary care, education and social care data will be invaluable in helping them do this. Longitudinal population-based research cohorts are limited in their ability to do this as they tend to be small and typically don't include the most vulnerable children. Potential research questions: • What are the characteristics and health outcomes for children placed in social care out of local authority compared with those placed closer to home? • Do children with serious learning impairing conditions have lower rates of emergency hospital admission if they attend a special school than if they attend a mainstream school? • Does school type or area affect rates of emergency admissions and A&E attendance for adversity-related conditions (e.g. self-harm, violence or mental health), after adjusting for underlying chronic conditions, previous admissions, age and socioeconomic factors such as maternal country of birth? • Which maternal factors (including education history and contact with social care services) mediate the effect of health visiting on child health outcomes? • Does early provision of special educational needs support for children with chronic conditions improve their participation in school, and reduce behaviour or mental health problems in adolescence and adulthood? • Can we identify child abuse and neglect through medical coding (comparing how many are known to services with how many we can identify in health data)? • How do characteristics and health outcomes for children placed out of county compare with those placed closer to home? • Does poor school attainment or frequent absences during adolescence predict risk-taking behaviour, or early pregnancy? • How does hospital contacts for health problems during pregnancy affect child health and education outcomes? • How does special educational needs provision influence health outcomes for children with different health conditions? • Does exposure to different aspects of children's social care modify the associations between childhood adversity and later health and educational outcomes? • How does parental health and contact with social care services affect childhood health and educa

Processing activities

The activities outlined below happened under v0 of this agreement but only for those born on or after 1.9.1995. Version 1 of the agreement will follow the same activities but for those born on or after 1.9.1984. UCL are requesting that the requested data are transferred to the ONS SRS in order to be merged with the education data from DfE. Merging will be on the basis of a link key that will be generated by NHS England, which will allow the DfE and health records of individuals to be brought together. The application shows data held and requested which is the data that is held and requested under NIC-393510-D6H1D. This agreement (NIC-381972-Q5F0V) will allow the data to be linked to that which is already held at UCL. The only data requested under NIC-381972-Q5F0V is a record identifier which allow the customer to link those matched with the ECDS data held under NIC-393510-D6H1D, this is because there are no HES IDs in ECDS. The researchers have requested the minimum data necessary for their research, which reflects the administrative history of the child/young adult for a subset of the available fields (e.g. the researchers have requested 60% of available inpatient fields, with no sensitive or identifiable fields). Data from the newly requested datasets (MSDS, CSDS, Mental health data) have not been minimised, due to the need to further explore these data in order to understand which variables are useful and sufficiently complete. Once this has been completed, the data will then be minimised, hopefully within 12 months. Data will be restricted to records relating to individuals in England born from 01/09/1984 onwards. Longitudinal data for all children and individuals in England born from 01/09/1984 is justified for the following reasons: Analyses for RQ1: The researcher will analyse outcomes for all children and individuals and compare differences in health outcomes (e.g. emergency hospital contacts, deaths) between groups classified as vulnerable or not (stratified by age group), in the years before, compared with after, COVID-19 onset. Descriptive analyses will explore whether changes post-COVID-19 reflect increased risks of contacts related to infection, mental health, adversity, or acute complications of chronic or complex health conditions for all children and individuals, and according to whether they had indices for vulnerability or not. The researcher will use diagnostic and procedure codes, and types of hospital contact (eg emergency/elective admission), recorded in HES, to infer whether hospital contacts are related to underlying chronic conditions. Preliminary results October 2021. Analyses for RQ2: The researcher will model expected healthcare contacts for all children and individuals after the onset of the COVID-19 pandemic, based on observed trajectories of healthcare contact for children and individuals for periods prior to the start of COVID-19. The researcher will assess potential unmet healthcare need, according to age and vulnerability status, by comparing expected vs observed healthcare contacts after COVID-19 onset. Differences between expected and observed new diagnoses and interventions, will be used to infer potential unmet need. The researcher will vary their assumptions about whether and when these healthcare needs may manifest, and whether deferred presentation is likely to be more severe. The researcher will estimate how much deferred healthcare use could add to predicted rates of healthcare contacts in the post-COVID-19 period, and potential long-term outcomes of COVID-19 restrictions. Preliminary results in January 2022. The researcher will also conduct analyses of all children and individuals to explore associations between inequalities (using index of multiple deprivation), ethnic group, and vulnerable vs other children and individuals, and outcomes measured in health care, NPD data (i.e. education and CiN/CLA) in periods before, during and after the COVID-19 pandemic. Funding for these analyses is provided through the NIHR Policy Research Programme, NIHR Programme for Applied Research, Health Data Research UK and Administrative Data Research UK (see section 8). The researcher will refresh the linked NPD datasets annually in 2021 and 2022 to evaluate the longer-term impacts of COVID-19 infection and response on health, education and social care outcomes for vulnerable, compared with other children or individuals. To address these research questions, HES (APC, A&E, OPD, ECDS, critical care) and death registration data (HES-mortality data) provided by NHS Digital will be linked to administrative data from the National Pupil Dataset (NPD) provided by the Department for Education (DfE) to the ONS SRS. Using a pseudonymised linkage key, these datasets will be linked for all children and individuals born in England on or after 1.9.1984. Further information with regards to how data minimisation has been applied including why the full cohort is required (longitudinal data for all children and individuals born in England on or after 1.9.1984) is justified for the following reasons: [1 paragraph unchanged] Examining health data from the time of birth to current age (up to, but not including, age 35 years) adulthood is critical for identifying markers of vulnerability or other predictors of adverse health conditions in administrative data. For example, previous work completed by UCL has shown [34 words unchanged] UCL have demonstrated the added value of using the whole longitudinal record. The researchers have requested the minimum data necessary for their research, which reflects the administrative history of the child for a subset of the available fields (e.g. the researchers have requested 60% of available inpatient fields, with no sensitive or identifiable fields). The study’s age range of all young people in England born on or after 01.09.1984 and use of the mother-baby flag will allow evaluation of health outcomes in early adulthood that may be influenced by health, education and social care in childhood. UCL will also assess the influence of health, education and social care risk factors among women who give birth on health outcomes in their child. [1 paragraph unchanged] The research aims to draw conclusions that are valid for all children and individuals in England. Yet the pandemic has had differential impacts However, we know that health outcomes and service use vary across the country (reflecting (e.g. both COVID infection rates and public health responses) responses varied geographically) at different times. For example example, surveys (e.g. RCPCH) The Royal College of Paediatrics and Child Health (RCPCH)) indicate geographical heterogeneity, including re-routing/re-deployment of healthcare staff and services, uptake of [93 words unchanged] requesting Middle layer Super Output Area rather than Lower Super Output Area. [1 paragraph unchanged] The research focuses on the impact of the pandemic and lockdown on vulnerable children and individuals (further defined below). Reliable identification of children and individuals meeting this definition is not trivial, requiring longitudinal data from birth across health, education and social care. As a result there are relatively few robust estimates of the size of this population. In order for the research to draw meaningful conclusions, the researchers wish to draw comparisons between different groups of children (e.g. care leavers, children born preterm, or other vulnerable groups) relative to a series of control children. The researchers will draw high level comparisons (e.g. to all other children) relevant to evaluating impacts at national level and for international comparisons, as well as detailed comparisons against synthetic control groups (e.g. through propensity score matching) to better understand the impacts of different groups of children in the context of related factors such as local environment, access to schools and healthcare needs. The researchers therefore require data for all children and individuals in England as without these data, comparisons would be incomplete, at greater risk of selection bias and not generalisable. However, there is good evidence that these indicators of vulnerability are common. For example, new research estimates that 25% of all children are ever designated a child in need and that 44% (of the 25%) are ever referred to children’s social care before the age of 16 years. A further subset of children will have other indicators of vulnerability reflecting health or educational needs. In order for the research to draw meaningful conclusions the researchers wish to draw comparisons between the impact of the pandemic and lockdown on different groups of vulnerable children relative to a series of control children. The researchers will draw high level comparisons (e.g. to all other children) relevant to evaluating impacts at national level and for international comparison, as well as detailed comparisons against synthetic control groups (e.g. through propensity score matching) to better understand the impacts of vulnerability in the context of related factors such as local environment, access to schools and healthcare needs. DATA FLOWS The researchers therefore require data for all children and individuals born in England on or after 1.9.1984 as without these data our comparisons would be incomplete, at greater risk of selection bias and not generalisable. Linkage of identifiers from health and NPD has been conducted at NHS England. NHS England have transferred the pseudonymised linkage key to the UCL Data Safe Haven to flag linked records in the UCL-curated HES extract for transfer to the ONS Secure Research Statistics (SRS). NPD attribute data will only be available in the ONS SRS. Using the pseudonymised linkage key, merging of pseudonymised attribute data (clinical or education characteristics) will then occur separately, at the ONS Secure Research Service (SRS). The following outline describes the complete data flow for future linkages and details how identifiable and non-identifiable data extracts are handled. In the below, health data refers to the health datasets requested from NHS England (HES, mortality, ECDS, MSDS, CSDS, birth notifications/registrations, mental health data). Linkage of identifiers from HES-mortality data and NPD will be conducted at NHS Digital which will then only transfer the pseudonymised linkage key to the UCL Data Safe Haven to flag linked records in the UCL-curated HES extract for transfer to the ONS Secure Research Statistics (SRS). NPD attributable data will only be available in the ONS SRS. Using the pseudonymised linkage key, linkage of pseudonymised attribute data (clinical or education characteristics) will then occur separately, at the ONS Secure Research Service (SRS). The following outline describes the complete data flow and details how identifiable and non-identifiable data extracts will be handled: 1) DfE supply the Trusted Third Party (NHS England) with a list of NPD identifier variables. These identifiers include name, date of birth, full postcode and sex, alongside a study specific pseudonymised linkage key known as the anonymized Pupil Matching Reference (aPMR). The identifying variables will be used for linkage to the Master Person Service (MPS)/Personal Demographic Service (PDS) (as previously done for NIC-27404 and NIC-381972). DfE will transfer the variables for any CYP born on or after cohort inception (1.9.84). 1) DfE will supply the Trusted Third Party (NHS Digital) with a list of NPD identifier variables, these identifiers include name, date of birth, full postcode and sex, alongside a study specific pseudonymised linkage key known as the anonymised Pupil Matching Reference (aPMR). The identifying variables will be used for linkage to the Master Person Service (MPS)/Personal Demographic Service (PDS). DfE will transfer the variables for any CYP born on or after cohort inception (1.9.84). 2) NHS England will match the identifiers from DfE to records held in the MPS/PDS using an algorithm that makes use of the chronology of postcodes in NPD and MPS/PDS. Matching to MPS/PDS data will be done internally within NHS England: no MPS/PDS data will be disseminated to ONS SRS or UCL Data Safe Haven. NHS England will link the NPD pseudonymised linkage key (i.e. anonymised PMR or young person ID) to MPS/PDS, and then to the ECDS data. 2) NHS Digital will match the identifiers from DfE to records held in the MPS/PDS using an algorithm that makes use of the chronology of postcodes in NPD and MPS/PDS. Matching to MPS/PDS data will be done internally within NHS Digital, no MPS/PDS data will be disseminated to ONS SRS or UCL Data Safe Haven. NHS Digital will link the NPD pseudonymised linkage key (i.e. anonymised PMR or young person ID) to MPS/PDS, and then to the ECDS data. 3) For those children and young people whose NPD identifiers were matched to MPS/PDS, onward linkage to health data will occur within NHS England, linking aPMRs and Token Person IDs. NHS England will then transfer encrypted Token Person IDs, aPMRs, and indicators of match rank (denoting the step at which the match to HES and MPS/PDS was made) for these linked cases to the ONS SRS. 3) For those CYP whose NPD identifiers were matched to MPS/PDS, onward linkage to HES-mortality data will occur within NHS Digital to link aPMRs and HES-IDs. NHS Digital will then transfer encrypted HES-IDs, aPMRs, and indicators of match rank (denoting the step at which the match to HES and MPS/PDS was made) for these linked cases to the UCL Data Safe Haven for linkage to the existing HES-mortality extract held by UCL (NIC-393510-D6H1D). 4) NHS England will extract the health data for all children and young people born on or after 1.9.1984, including a pseudonymised mother-baby link and additional HES records of mothers, and link the aPMR and match rank statistics for those children and young people that were linked by NHS England in step (3) from NPD. The deidentified health data will be transferred to the ONS SRS. Only month/year of birth and death will be transferred to the ONS SRS, in order to account for well-established effects of month of birth on school achievement (i.e. research consistently shows that children born in September do better than children born in July/August). 4) UCL will extract the HES-mortality data for all CYP born on or after 1.9.1984, using the existing HES extract, including a pseudonymised mother-baby link and additional HES records of mothers, and link the aPMR and match rank statistics for those CYP that were linked by NHS Digital in step (3) from NPD. The de-identified HES-mortality extract will be transferred to the ONS SRS. Only month/year of birth and death will be transferred to the ONS SRS, in order to account for well-established effects of month of birth on school achievement (i.e. research consistently shows that children born in September do better than children born in July/August). The attribute data will also include high-level categorical indicators relevant to birth (e.g. parity – 0, 1, 2+ prior births), which have an important bearing on child health. These derived health indicators are created by analysing variables in our analysis files, that are covered by this DSA (e.g. diagnosis codes and mother or baby tail) and for which we have the appropriate permissions. 5) DfE will supply ONS SRS with requested de-identified attribute data extracts, with the aPMR for all children and young people born on or after 1.9.1984. The deidentified attribute NPD and HES data will be linked within the ONS SRS by the research team, using the aPMR. Data will only be used by researchers authorised for the project or those who have been granted access to the data through a sub-license, with strict output controls applied by ONS SRS staff. 5) DfE will supply ONS SRS with requested de-identified attribute data extracts, with the aPMR for all CYP born on or after 1.9.1984. The deidentified attribute NPD and HES data will be linked within the ONS SRS by the research team, using the aPMR. Data will only be used by researchers authorised for the project, with strict output controls applied by ONS SRS staff. 6) The final data set that will be used for analyses will remain within the ONS SRS. The files will not contain any identifiable data. No additional record level data will be gathered or linked to the dataset. The aPMR is the only variable supplied from NPD data that is supplied by NHS England to UCL Data Safe Haven and then to ONS SRS. 6) The final data set that will be used for analyses will remain within the ONS SRS. The files will not contain any identifiable data. No additional record level data will be gathered or linked to the dataset. The aPMR is the only variable supplied from NPD data that is supplied by NHS Digital to UCL Data Safe Haven and then to ONS SRS. 7) NHS England will retain the identifier file of all individuals linked in MPS/NPD-PDS and MPS/PDS-HES and all the postcodes used in linkage and postcode dates for 12 months after linkage, to address data queries or potential linkage errors. This data set will not contain any attribute data and will be accessible only to NHS England staff. At the end of the 12 months, NHS England will confirm deletion of the data to DfE. NHS England will not send confidential data to DfE or UCL DSH. 7) NHS Digital will retain the identifier file of all individuals linked in NPD-MPS/PDS and MPS/PDS-HES and all the postcodes used in linkage and postcode dates for 24 months to address data queries or potential linkage errors (v1 note: still being retained as 24 months has not yet passed since the linkage has been undertaken). This data set will not contain any attribute data and will be accessible only to NHS Digital staff. At the end of the 24 months, NHS Digital will confirm deletion of the data to DfE. NHS Digital will not send confidential data to DfE or UCL DSH. ACCESS TO THE DATA BY UNIVERSITY COLLEGE LONDON, LONDON SCHOOL OF HYGIENCE AND TROPICAL MEDICINE (LSHTM) AND INSTITUTE FOR FISCAL STUDIES (IFS) (NAMED DATA PROCESSORS) LSHTM and IFS will each have a named researcher and investigator who will undertake analyses (on the ONS SRS) relating to a specific component of the wider research question on the impact of the pandemic on vulnerable children and young people. e.g. LSHTM will examine the impact (on health and education) of delays in time-sensitive procedures (e.g. surgical correction of cleft lip and palette) on children with underlying health conditions. These named researchers will be substantive employees of the respective organisations. These researchers are undertaking this analyses under instruction of UCL. UCL researchers who have a substantive contract with UCL will be authorised to access the data in the ONS SRS, for purposes covered in this DSA. UCL PhD students who are under the supervision of UCL substantive employees will also be able to access the data for the purposes covered in this DSA. No MSc or undergraduate students will access the data. The de-identified linked HES-NPD attribute data will be held on the ONS SRS and will only be accessible remotely from the UCL Data Safe Room which has restricted and monitored access. No record level data can be removed from the ONS SRS and statistical disclosure controls are applied by ONS staff. Access will be restricted to named users, who are part of the study team and are accessing the data for the purposes outlined in this DSA. Access to the data is via the ONS SRS environment. All UCL PHD students are expected to undertake annual training on handling highly confidential information. All Trainees and students register for and complete NHS England’s Data Security Awareness (NHSD) course provided by e-Learning for Health. The course covers data security awareness, the law, threats to data security, breaches and incidents, and the General Data Protection Regulation. All UCL staff and students are also required to complete internal UCL GDPR training annually. UCL has a specific data protection and information security policy, which applies to all staff and students when processing personal data on behalf of UCL. All UCL students working on the study are bound by this policy, and that they will face potential sanctions in the event of a breach of the policy. Office of National Statistics (ONS) and UCL have signed and maintain an organisational agreement to use the ONS Secure Research Statistics (SRS) service for the purposes of secure statistical research, signed on 21/03/2019 with an indefinite expiry date. The only HES data stored in the ONS SRS are the 4 one-year cohorts of HES, plus the anonymised PMRs for those records that link to NPD. The HES data will include month of death and month of birth so no identifying data. All UCL students sign up to the UCL's Academic Manual. The Student Academic Misconduct section of the 2021-2022 manual Section 9.1, item 3 states "All instances of Research Misconduct whether by taught students, research students or members of staff will be investigated under UCL’s Procedure for Investigating and Resolving Allegations of Misconduct in Academic Research". For security and resource reasons the SRS is a Managed Service. Equiniti Ltd (based in Belfast) maintains the system, on behalf of the ONS SRS. They do so through encrypted (TLS1.2) VPN tunnel and Remotely Access (RA) the SRS. All Equiniti Ltd administrators are SC cleared and have no access to any data. ONS SRS Research Support “Admin” staff only have permissions to carry out such tasks as creating users, updating patches, testing and installing software applications, arranging DR, ITHC for the SRS environment, closing SRS sessions down, i.e. all the SRS environment Admin maintenance - essentially they are “power users”. There have been no data infractions by Equiniti Ltd staff in the last 5 years of them maintaining the environment, they have been very professional. LSHTM and IFS each have a named researcher and Principle Investigator (PI) who are undertaking analyses (on the ONS SRS) relating to a specific component of the original research question e.g. LSHTM will examine the impact (on health and education) of delays in time-sensitive procedures (e.g. surgical correction of cleft lip and palette) on children with underlying health conditions. These named researchers will be substantive employees of the respective organisations. DfE and DHSC analysts will access the ECHILD data on the ONS SRS under separate agreements with NHS England. These applications will be considered separately to the UCL sublicensing application. The de-identified linked HES-NPD attribute data will be held on the ONS SRS and will only be accessible remotely. No record level data can be removed from the ONS SRS and statistical disclosure controls are applied by ONS staff. Access will be restricted to named users, with ONS accreditation. Amazon Web Services provides cloud hosting services to UCL and will store the data as contracted by UCL. UCL uses offsite data centre services provided by VIRTUS data centre. DATA STORAGE The Office of National Statistics (ONS) and UCL have signed and maintain an organisational agreement to use the ONS Secure Research Statistics (SRS) service for the purposes of secure statistical research, signed on 21/03/2019 with an indefinite expiry date. HES data will be stored in the ONS SRS, plus the anonymised PMRs for those records that link to NPD. The HES data will not include any identifiable data. For security and resource reasons the SRS is a Managed Service. Equiniti Ltd (based in Belfast) maintains the system, on behalf of the ONS SRS. They do so through encrypted (TLS1.2) VPN tunnel and Remotely Access (RA) the SRS. All Equiniti Ltd administrators are SC cleared and have no access to any data. ONS SRS Research Support Administrative staff only have permissions to carry out such tasks as creating users, updating patches, testing and installing software applications, arranging DR, ITHC for the SRS environment, closing SRS sessions down, i.e. all the SRS environment Admin maintenance - essentially they are power users. There have been no data infractions by Equiniti Ltd staff in the last 5 years of them maintaining the environment. [1 paragraph unchanged] The ONS SRS environment is an isolated system. It has no connectivity [36 words unchanged] UKCloud Ltd merely host the environment, they have no access to data. Therefore CloudUK Ltd is not considered to be a Data Processor. AMENDMENT Therefore CloudUK Ltd is not considered to be a Data Processor. This study’s extension of age range to all individuals in England born on or after 01.09.1984 and use of the mother and baby/babies flag will allow evaluation of health outcomes in adulthood that may be influenced by health, education and social care in childhood. UCL will also assess the influence of health, education and social care risk factors among women who give birth on health outcomes in their child. CLOUD SECURITY NHS England security has provided assurance regarding the use of the Office of National Statistics' Secure Research Statistics service (ONS SRS), hosted by CloudUK Ltd in this application. The Office of National Statistics has submitted a selection of security documentation to support the use of cloud storage. NHS England Security have reviewed the documentation and provided relevant feedback, where necessary. NHS England are satisfied that the documentation demonstrates the level of security and governance in place. The Office of National Statistics have supplied evidence to support: * The use of the Data Risk Model to assess the Risk Profile Class. * Risk Management of the use of the Cloud for this data, taking into consideration Confidentiality, Integrity and Availability. * The use of Pseudonymisation. * Board level involvement in the Risk Management Process evidenced through Minutes of these meetings. ͻUnderstanding of the Shared Responsibility Model The Office of National Statistics have a very good understanding of the security controls available to them to provide the appropriate controls to secure data in the Cloud. Using the Cloud, benefits from the inherited controls that cannot practically be replicated locally such as Physical Controls, Resilience of Systems, Power Supplies, Communications and Geographically dispersed Data Centres within a region. Elasticity in provisioning is also a consideration that benefits organisations in managing workloads. The Cloud provider, CloudUK, will use UK Data Centres only. DISCLOSURE CONTROL RULES Community Services Dataset The suppression rules are as follows (applicable if one of something is necessarily one person e.g. length of stay - one day as an inpatient can only relate to one person in many data sets): · Zeros should be shown. · 1-7 to be rounded to 5. · Any other numbers rounded to nearest 5. · Rounding unnecessary for averages etc. · Percentages calculated from rounded values Maternity Services Dataset (MSDS) In order to prevent disclosure of identities or information about service users, · all figures (except national) for all organisations which submitted, are rounded to the nearest five; · all figures between zero and four are suppressed (*); · averages with a denominator of less than five have been replaced by '*' and other values have been rounded to one decimal place. Mental Health Services Dataset (MSDS) and Mental Health Services Dataset (MHSDS) v5.0 The publication covers sensitive topics. There would be a risk of identification/self-identification from small numbers. In order to minimise the disclosure risk associated with small numbers, all figures presented within the report and within the reference data tables have had the following measures applied: · Present zeros; · 1-7 rounded to 5 · Other values have been rounded to the nearest 5; · Percentages calculated from rounded values.

Expected output

All outputs will contain aggregate level data only and all small numbers will be suppressed in line with the HES and Department for Education analysis guide. Outputs will be monitored for compliance with ADRN statistical output controls and the HES Analysis Guides. guidance (small numbers suppressed). No potentially disclosive outputs will be shared or published. The data analyses are conducted on the ONS Secure Research Service. Detailed individual level data cannot leave the ONS Secure Research Service. Results of analyses can be exported by a secure encrypted transfer system on the ONS SRS, which is audited. Any outputs from analyses that are published have to meet statistical disclosure controls that prevent small sizes in accordance with NHS England and DfE requirements. Tabulations of aggregate data are assessed for statistical disclosure control and authorized for export by an ONS data scientist not involved in the project. Preliminary reports will be shared with DHSC, PHE and NHS England, and with DfE through the ECHILD project and a project advisory group. Preliminary results will be produced for October 2021 (RQ1) and January 2022 (RQ2). Reports on linkage evaluation will be published in 2021. An introductory guide to ECHILD has been published on the ECHILD website www.ucl.ac.uk/child-health/echild. Reports describing patterns of vulnerability status, and the impact of vulnerability status on outcomes before and after the pandemic, will be published in 2022. The researchers will submit full reports for publication in peer reviewed journals and produce briefing reports for policy stakeholders and lay summaries. For example, for research conducted for the NIHR Children and Families Policy Research Unit (CPRU), reports will be produced for DHSC. Findings will also be used in public involvement and engagement events. Study findings will be also disseminated through peer-reviewed academic journals (e.g. British Medical Journal (BMJ), Lancet Public Health), and social media including lay summaries. UCL would expect that findings from the research will be presented at conferences such as the Lancet Public Health conference, and International Population Data Linkage Conference within two years of obtaining the data. Reports on linkage evaluation will be published in 2021. An introductory guide to ECHILD has been will be published on the ECHILD website www.ucl.ac.uk/child-health/echild. Reports describing patterns of vulnerability status, and the impact of vulnerability status on outcomes before and after the pandemic, will be published in 2022-2024. Sublicense agreements will encourage relevant findings to be shared with policy makers, clinicians/health professionals, educators and parent/family groups in accessible formats (e.g. lay summaries, videos or animations). This could include forums such as the National Children's Bureau (NCB) Young Person and Parent group, the Great Ormond Street Hospital (GOSH) Patient Engagement group. Lay summaries of the study findings can be published on the ECHILD website, and linked through websites for sub-licensee organisations. The researcher will submit full reports for publication in peer reviewed journals and produce briefing reports for DHSC, DfE and other public bodies through the Children and Families Policy Research Unit (CPRU). Findings will also be used in public involvement and engagement events. Study findings will be also disseminated through peer-reviewed academic journals (e.g. BMJ, Lancet Public Health), and social media including lay summaries. Sublicensees will be required to report yielded benefits to UCL and NHS England at the time of each annual review of the sub-license agreement. Sublicensees will also be required to notify UCL of all publications in advance of publication. Details of the publications will then be recorded on the ECHILD Release Register (https://www.ucl.ac.uk/child-health/research/population-policy-and-practice-research-and-teaching-department/cenb-clinical-22 and also see ECHILD Data Access Decision Making Process Standard Operating Procedure). The release register will also include details of the research project (including a summary of the purpose), the licensee’s organisation and Chief Investigator, and the licence end date. UCL would expect to present findings at conferences such as the Lancet Public Health conference, and International Population Data Linkage Conference within two years of obtaining the data. All outputs will be required to meet strict disclosure control rules and small numbers will required to be supressed in line with the HES analysis guide. Relevant findings will be shared with policy makers, clinicians/health professionals, educators and parent groups particularly in accessible formats (e.g. lay summaries, videos or animations). This could include forums such as the National Children's Bureau (NCB) Young Person and Parent group, the Great Ormond Street Hospital (GOSH) Patient Engagement group. These groups can be accessed through the joint institute of UCL Great Ormond Street Hospital Institute of Child Health, through the North Thames ARC (led by UCL) and through the CPRU. Lay summaries of the study findings can be published on the CPRU website, and linked through websites for these organisations. The data analyses are conducted on the ONS Secure Research Service. Detailed individual level child data cannot leave the ONS Secure Research Service. Results of analyses can be exported by a secure encrypted transfer system, which is audited. Any outputs from analyses that are published have to meet statistical disclosure controls that prevent small sizes in accordance with NHS Digital and DfE requirements. Tabulations of aggregate data are assessed for statistical disclosure control and authorized for export by an ONS data scientist not involved in the project.

Expected measurable benefits

This project aims to produce urgent results on the impact of COVID-19 infection and lockdown on the health of children and individuals and in particular vulnerable children, characterised by education and social care indices from the NPD linked datasets. It is hoped the study will provide vital understanding of the repercussions of the current response strategies on the health and well-being of key population groups, and provide insight into how infection control and lockdown strategies should be developed to better meet the needs of children and individuals in England who are aged between 0 and 34 years at the time of the COVID-19 pandemic. These results (preliminary results in October 2021 and January 2022) are critical for addressing current health needs arising from COVID-19 infection and responses, and also for informing strategies for future waves of infection. Better understanding of how and when to intervene with families early on is a cross-government policy priority. For example, Public Health England’s 2020 report on addressing vulnerability in childhood (No child left behind) emphasises that intervening early can mitigate negative impacts of early adversity, and that family settings that provide a safe and secure environment are an essential protective factor. The Children’s Commissioner ‘Best Beginnings’ report in 2020 described a system that fails to target the most vulnerable and disadvantaged children. It specifically recommended better sharing of data between different services, including more effective use of NHS number and Unique Pupil Number, in order to ensure that all families are given the support they need to help their children to thrive, and to prevent early challenges turning into serious problems. The Early Intervention Foundation 2020 report on Adverse Childhood Experiences (ACEs) emphasised that good data on the prevalence of childhood adversity and wider risk factors is lacking, and that more accurate estimates are essential in order to plan services and to ensure that effective interventions are available for the children and families who most need them. However, population-level data linking parental and family characteristics with children’s health and education outcomes across the life course is lacking. It is hoped the study will compare different groups of vulnerable and non-vulnerable groups of children and individuals, using indicators of vulnerability drawn from health, social care and education histories in administrative data. It is hoped the study will estimate impacts of service support (from schools, social care or hospitals) pre-COVID in order to predict detrimental effects of reduced support during COVID. It is hoped the analyses will address a priority for policy makers, that COVID-19 and lockdown have resulted in disproportionate impacts on health for vulnerable groups. The analyses aim to explore this question for the whole population, to inform policy to better support children and individuals, and to better understand which types of vulnerability are most affected. The ECHILD database may help fill this gap in evidence. For example, the Director of the NIHR PenARC has stated that “Children’s ability to participate in education and their health are closely linked. Better responses by schools to children’s physical and mental health difficulties could have important impacts on their health and use of services. The ECHILD database may offer the possibility of a step change in research that may inform decision makers how schools can improve health outcomes, and work more effectively with health care services. I anticipate this data resource being hugely valuable for research to improve child health” Concerns have been raised about the impact of the resulting treatment delays, yet much more evidence is required in to quantify the scale of the problem in different population groups and to predict future/ongoing needs. ECHILD has also been recognised as an example of good practice by the UK Statistics Authority. The research seeks to help to fill this gap, firstly by evaluating high level differences in impacts for groups of vulnerable (in terms of clinical, socio-demographic, social care and educational needs) versus other children and individuals, which will guide the development of more detailed, in depth research within groups for which there is evidence of the most significant adverse impacts. Specific examples include examining the impact of delays in time-sensitive procedures, where delays are expected to have significant and prolonged negative impacts on health and education outcomes (e.g. surgical correction of cleft lip and palate). It is hoped the results will establish the scale and urgency (e.g. how many children, how extensive were the delays, what are the expected unmet healthcare and education needs) of these impacts and guide the development of reactive policies and changes in service provision to mitigate long-term impacts for specific population groups. It is hoped the research will also add evidence on the impact of COVID-19 on health inequalities - including for black and minority ethnic groups – and the mechanisms that drive these. The comprehensive geographical coverage and population base of the research is a real strength of this research and it is hoped it will allow the researchers to draw conclusions for all vulnerable children and individuals in England who are aged between 0 and 34 years at the time of the COVID-19 pandemic, and to identify groups who might be disadvantaged by current policy or need more support from services. The ECHILD Research Database may generate important evidence on the inter-relationships between health and education, especially for groups exposed to different education and social care characteristics as determined from the NPD linked datasets. A list of potential research questions is provided in the Objective for Processing section: being able to answer these questions would hopefully provide huge impact in terms of informing preventative strategies, informing children and their parents, informing clinical practice, and identifying groups in need of targeted interventions. Findings will be reported directly to DfE and NHS policy makers. The project also addresses two priorities on the impact of COVID19 on vulnerable patients set out by Health Data Research UK. The study is reporting preliminary results to DHSC policy makers through our regular 2-monthly meetings, through seminars with wider NHS staff (DHSC, PHE and NHS England), and through briefing reports and papers published in peer reviewed journals. Widening data access through sublicensing of ECHILD will hopefully bring a step change in evidence for informing policy interventions to improve health outcomes as well as education and social care. Many more of the potential research questions listed in the Objectives for Processing will hopefully be answered than would be the case without a sublicensing model to enable data access. The sub-license model will also hopefully accelerate and scale up analytic capacity among a wide range of analysts in government, academia and the third sector. Widened access through the sublicence could improve rigour and interpretability of research using complex data as other researchers may be able to contest and replicate findings reported by others. Access to ECHILD data could have the following potential impact and benefits: Evaluation of policy interventions. For example, to evidence impacts of changes in Special Educational Needs (SEN) for children exposed to ACEs. Child exposure to ACEs is strongly related to parental mental health conditions. Linked records on children’s school and health longitudinal trajectories may help improve targeting of provision (eg early day care) for children exposed to ACEs, by using ECHILD to predict additional social, education or health needs. The data could be used to help monitor improved provision. Evidence from ECHILD will hopefully be used to support policies and services. For example, ECHILD could be used to improve surveillance of childhood conditions related to maternal health during pregnancy, which impact on SEND provision, and cognitive ability measured through school attainment; Local Authorities can hopefully improve commissioning and targeting of the Healthy Child Programme to improve school readiness and attainment, and target school nurse support, disability services, and day care. The DHSC-commissioned, NIHR Children and Families Policy Research Unit will hopefully be able to use the data to evaluate policies for vulnerable and disadvantaged children, and to hopefully reduce health inequities. Health services may benefit from the evidence generated from the data, by improving access to early preventive services for the most vulnerable groups, for example vulnerable youth before and after becoming parents, thereby hopefully reducing later interventions (e.g. out of home care for their child, repeat hospitalisation). The data could be used to inform existing and future cohort investments by mapping onto groups missed from traditional research studies and providing insights into the representativeness of cohorts based on longitudinal trajectories. Children, young people, families and the wider public may benefit from the data if it leads to more effective policies and services (and thereby more efficient use of public funds), and better health and educational outcomes for children and young people. ECHILD could provide a means for understanding the overall population of children and young people in England, with information about their household and their needs. Use of the linked data will hopefully inform population surveys and censuses, informing better targeting of expensive surveys and research studies. Analyses will hopefully explore outcomes for the whole population, to possibly inform policy to better support children and individuals, and to hopefully better understand which types of vulnerability are most at risk of poor outcomes. Findings may be reported directly to DfE and NHS policy makers.

Benefits reported

Since June 2021, when linked ECHILD data could first be accessed, UCL have published the following research findings. Analyses resulting from the original research question on the impact of COVID-19 on vulnerable children and individuals have been shared with and used to inform strategies by DHSC, PHE and NHS England, and with DfE through the ECHILD project and a project advisory group. i) UCL showed that the quality of data linkage between schools and hospital data was good and improved over time. However, those not linked were disproportionately poor or from certain ethnic groups. This information can be used to reduce biases in linkage and in analyses. (please see the "Linking education and hospital data in England: linkage process and quality" paper published in IJPDS Vol. 6 No. 1 (2021), available: https://doi.org/10.23889/ijpds.v6i1.1671 and "Ethnic bias in data linkage" correspondence in The Lancet Digital Health, Vol 3, Issue 6, E339, (2021), available: https://doi.org/10.1016/S2589-7500(21)00081-9) Since June 2021, when linked ECHILD data could first be accessed, the researchers have published the following research findings: ii) Recent analyses showed that children who were vulnerable, due to contact with social services or because they received special educational needs support for additional learning needs, had a much greater deficit in hospital care during the COVID pandemic than their peers. Findings indicate a need for targeted ‘catchup’ funding and resources for child health, particularly for vulnerable children who were affected disproportionately. i) UCL showed that the quality of data linkage between schools and hospital data was good and improved over time. However, those not linked were disproportionately poor or from certain ethnic groups. This information can be used to reduce biases in linkage and in analyses. (please see the "Linking education and hospital data in England: linkage process and quality" paper published in IJPDS Vol. 6 No. 1 (2021), available: https://doi.org/10.23889/ijpds.v6i1.1671 and "Ethnic bias in data linkage" correspondence in The Lancet Digital Health, Vol 3, Issue 6, E339, (2021), available: https://doi.org/10.1016/S2589-7500(21)00081-9). This evaluation has informed other linkages between health and education data in order to improve the quality of linkage of cross-sectoral data and improve the quality of data being used to generate benefits for health. ii) Analyses showed that children who were vulnerable, due to contact with social services or because they received special educational needs support for additional learning needs, had a much greater deficit in hospital care during the COVID pandemic than their peers. Findings have generated evidence on the need for targeted ‘catchup’ funding and resources for child health, particularly for vulnerable children who were affected disproportionately. For example, the ring-fenced resource for ‘catch-up’ of NHS care (Health and [62 words unchanged] pandemic among vulnerable children and young people”, 4th November 2021. available: https://www.ucl.ac.uk/children-policy-research/projects/assessing-impact-covid-19-pandemic-vulnerable-children). iii) Analysis showed that for children aged between 0 to 4 years, deficits in hospital care during the pandemic were much higher for clinically vulnerable children than peers. 1 in 6 clinically vulnerable accounted for one-third to one half of the deficit in hospital care during the pandemic. During the pandemic, weekly rates of planned care returned to pre-pandemic levels for infants with chronic conditions but not older children. Deficits in care differed by ethnic group and level of deprivation. These findings may be used to inform health services on who should be targeted for additional care as the pandemic slows. “Deficits in hospital care among clinically vulnerable children aged 0 to 4 years during the COVID-19 pandemic” accepted for publication, 17th December 2021 pre-print available https://www.medrxiv.org/content/10.1101/2021.12.16.21267904v1. December 2021. iv) Methodology paper showed how the ECHILD administrative data may be used in lieu of randomised controlled trials, through target trial emulation methods, to answer causal questions whilst reducing confounding and other biases likely to arise with such data. Methods focus on trial emulation to understand the impact of SEN support from the start of school on unplanned hospital utilisation in children with cleft lip and palate. This work may inform future studies using linked administrative data to generate benefits for health. “SEN support from the start of school and its impact on unplanned hospital utilisation in children with cleft lip and palate: a demonstration target trial emulation protocol using ECHILD” 5th April 2022 preprint available https://www.medrxiv.org/content/10.1101/2022.04.01.22273280v1 (DOI: https://doi.org/10.1101/2022.04.01.22273280). v) Systematic review and thematic analysis of administrative data research showed the characteristics of research-ready administrative data to define a common understanding of what constitutes research-ready administrative data. In turn, the analysis helps data owners and researchers develop common principles and standards to establish clear principles and frameworks for data’s development and the realisation of admin data’s full research potential. This work may inform future studies using linked administrative data to generate benefits for health. “What makes administrative data research-ready? A systematic review and thematic analysis of published literature” 27th April 2022 available https://ijpds.org/article/view/1718 (DOI:https://doi.org/10.23889/ijpds.v7i1.1718). vi) Analysis showed that those who were receiving special educational needs (SEN) support or children’s social care (CSC) services had greater decreases in planned hospital care than their peers during the COVID-19 pandemic. Those receiving SEN support or CSC services bore a proportionally larger decrease in outpatient attendances, planned hospital admissions and were also less likely than their peers to have face-to-face outpatient care during the pandemic. The large decreases young people experienced during the pandemic could mean that some health needs have gone unmet which may have long-lasting impacts on their health and well-being. These findings could be used to inform health services on who should be targeted for additional care as the pandemic slows. (Please see publication: “Changes in adolescents’ planned hospital care during the COVID-19 pandemic: analysis of linked administrative data” 16th May 2022. available https://adc.bmj.com/content/early/2022/05/15/archdischild-2021-323616 (DOI: http://dx.doi.org/10.1136/archdischild-2021-323616). vii) Children born even a few weeks too early are less likely to achieve expected levels of attainment at age 7 and 11 and are more likely to have Special Educational Needs provision than those born at 40 weeks of gestation. This association is not fully explained by maternal risk factors including deprivation, age and parity, or by size-for-gestation at birth. Chronic conditions in school-aged children contribute more to the burden of Special Educational Needs and low academic attainment than preterm birth. Additional support prior to school entry to improve school readiness could be targeted at high-risk groups based on early health indicators shown to influence later outcomes. This evidence could be used to inform strategies for allowing for the disadvantage that children born preterm experience as they start school, e.g. by allowing parents to make decisions about whether to delay school entry. “Gestational age at birth, chronic conditions and school outcomes: a population-based data linkage study of children born in England”, 19 May 2022. Available https://academic.oup.com/ije/advance-article/doi/10.1093/ije/dyac105/6589377?login=false (DOI: https://doi.org/10.1093/ije/dyac105) viii) UCL have authored the Data Resource Profile: The Education and Child Health Insights from Linked Data (ECHILD) Database. This resource helps researchers interested in the ECHILD Database, explaining what information the database contains and discusses ECHILD’s key strengths and limitations for research. Data Resource Profile available: https://academic.oup.com/ije/advance-article/doi/10.1093/ije/dyab149/6425590 (DOI: https://doi.org/10.1093/ije/dyab149). November 2021. ix) UCL have additionally been maintaining an ECHILD user guide, data catalogue, analysis code and PPIE findings to support for wider use. The ECHILD user guide may be downloaded here: The Education and Child Health Insights from Linked Data (ECHILD) Database - An Introductory Guide for Researchers - v1.1.1 Updated February 2022, with PPIE findings published here: Echild website / Engaging the public, this additionally includes ECHILD Stakeholder Event Report. The event acted as an opportunity for stakeholders to ask questions about and share their views on the ECHILD Database and its use in future research for the public benefit. Report from the event highlights the key messages from government and stakeholders regarding the ECHILD Database, available: https://www.adruk.org/news-publications/news-blogs/the-potential-of-linked-administrative-data-for-understanding-the-relationships-between-childrens-health-and-education-434/ . Lastly, UCL’s ECHILD and reusable HES coding (such as R and Stata code, as well as code lists on Github: here https://github.com/UCL-CHIG. Other benefits from using the ECHILD data can be found on the website - https://www.ucl.ac.uk/child-health/research/population-policy-and-practice-research-and-teaching-department/cenb-clinical-22

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

Datasets: + HES-ID to MPS-ID HES Admitted Patient Care

Objective for processing

Version 1 of this agreement is for the same purpose as version 0 but will be for all young people born in England on or after 01.09.1984 (rather than 01.09.1995). The following detail outlines further detail for this change: To enhance information on vulnerable children, UCL request an amendment to transfer to ONS SRS a pseudonymised mother and baby/babies linkage flag, attached to the relevant pseudonymised HES record for mother and baby/babies (this is pre-linked by NHS Digital). To enable inclusion of maternal characteristics (such as age), and to follow up the health of young people born on or after 01.09.1984; UCL request to extend the age range to all individuals in England born on or after 01.09.1984, from the current age limit of 01.09.1995. This project aims to improve understanding of the effects of the COVID-19 lockdown and restricted access to schools and health care on outcomes for children and young people. UCL require data over the child and adult life course because they will be taking a longitudinal perspective across life. UCL will compare how exposure to vulnerability during childhood and youth, in relation to health conditions, school factors (e.g. special needs, exclusion) and social care influence health outcomes during childhood and adulthood. These relationships need to be assessed in cohorts followed through the child to adult life course before the pandemic, and compare with similar cohorts who experienced deficits in health and social care and education, during the pandemic. In this way, UCL can estimate potential impacts of these deficits in care that might be attributable to the pandemic. Extension of the age range and the mum-baby link will allow evaluation of exposures and outcomes in adulthood, including of children who themselves become parents. Extension of the age range will enable linkage of adolescents with education records to their health records as adults up to the oldest age. Researchers will examine vulnerability indicators in the mother (e.g., maternal history of social care, SEND support, or school exclusion in her childhood, and characteristics of a child’s mother before or after delivery, such as maternal age at delivery, chronic mental or physical conditions, previous teenage motherhood) on child health outcomes. These linkages will be re-run for all children and extended to incorporate linkage to social care data held by DfE, and to the Emergency Care Data Set (ECDS) held by NHS Digital. The mum-baby link is already approved for use in NIC-393510-D6H1D. and is pre-linked by NHS Digital. UCL require administrative data for all children in England who appear in the specified NPD and HES datasets to create longitudinal cohorts of children born on or after 1.9.1984 - 95. UCL request a transfer of identifying variables from NPD datasets to NHS Digital to enable linkage to HES records. Initially, linkage between names, date of birth and postcodes will be via the Personal Demographic Service (PDS) and then, using NHS number from PDS to hospital episode statistics (HES). The output will be pseudonymised linkage keys (following the process used in the ECHILD project). UCL have previously demonstrated high quality linkage of 92% of NPD records to HES for individuals born in 1990/01 (NIC-27404-D5Z3F). Researchers will evaluate linkage bias, but anticipate acceptable linkage rates from 1984/5. HES histories will be included for any mother linked through HES to a child with an anonymised PMR, even if the mother is not included in NPD. Clarifications of funding sources and outputs to better describe the development of vulnerability cohorts before and after COVID are also included in this amendment. The following detail from version 0 has been updated to reflect the above change: [1 paragraph unchanged] CPRU is one of 15 NIHR Policy Research Units formed to undertake [35 words unchanged] departments and arms-length bodies, such as NHS England and Public Health England. Additional funding support is provided by Administrative Data Research UK, Health Data Research UK, and NIHR. For this programme of research, UCL are the sole Data Controller who [15 words unchanged] National Statistics (ONS) and The Institute for Fiscal Studies (IFS) are also listed as data processors. The study is looking at the impact of COVID-19 and lockdown on Children and young people and whether there are any differences in individuals under the health and social effects age of household confinement on vulnerable children and young people when compared to other children and young people. 35 years at the start time of the COVID-19 pandemic. Children and young people (CYP) or individuals who are vulnerable due to social welfare or chronic health needs are expected to experience more adverse health and social effects of the COVID-19 lockdown than other CYP. their peers. Key concerns for services are the effects of household confinement during the [25 words unchanged] infection and related public health responses (such as lockdown) have had on CYP, children and individuals, including those that are deemed vulnerable, to inform strategies for the current wave of infection, and any future waves. This study; Department of Health and Social Care - Education and Child Health Insight Insights from Linked Data - COVID (DHSC-ECHILD-COVID) (ECHILD-COVID) builds on the Education and Child Health Insight Linked Data (ECHILD) project (DARS-NIC-27404-D5Z3F - approved), (DARS-NIC-27404-D5Z3F) which uses linked education and HES data for four one-year cohorts amounting to two million CYP children and young people in England. The linkage under this application agreement will be extended urgently to address the impact of COVID-19 on all CYP children and individuals who are aged between 0 and 34 years in the COVID-pandemic year (linkage involving an expected 18 of 15 million CYP) and in particular vulnerable CYP as this is individuals). The subject of the group most likely to be impacted by lockdown. The researchers wish to research will include all children and young people (CYP) individuals appearing in HES records from (the latest of) birth or April 1997 onwards, who are aged between 0 and 24 34 years in the COVID-pandemic year (hence start date for birth is the start of school year 1.9.1995). 1.9.1984). PURPOSE ECHILD-COVID addresses four priority areas raised by the Department of Health and Social Care (DHSC) with the Children’s Policy Research Unit (CPRU) team relating to the secondary impacts of infection and lockdown on: DHSC-ECHILD-COVID addresses four priority areas raised by the Department of Health and Social Care (DHSC) with the Children’s Policy Research Unit (CPRU) team relating to the secondary impacts of infection and lockdown on: ~ Children who need safeguarding ~ CYP who need safeguarding ~ low income families ~ poorer families ~ Children with special educational needs ~ CYP with special educational needs [1 paragraph unchanged] These vulnerable groups can only be reliably identified through linkage of longitudinal health, education and social care data. The researchers will draw on the published DfE definition for vulnerable individuals. This relates to children assessed as being in need under section 17 of the Children Act 1989 (i.e. have records indicating contact with social care services, or are being looked after) special educational needs or additional needs (such as prolonged absences or school exclusion). Children or individuals assessed as high risk by educational providers or local authorities will also be considered vulnerable (e.g. children on the edge of receiving support or those at risk of becoming not in employment, education or training). For the purpose of this application 'vulnerable' can be defined as: The vulnerable cohort is defined by their exposure during childhood and early adulthood (up to the age 24). ECHILD is focused on the exposure to vulnerability for children and youths and the outcomes related to that exposure. The age groups of those aged 25-37 will allow follow up on vulnerable and non-vulnerable young people into adulthood to assess differences in health outcomes across these groups. Outcomes include health, education and social care outcomes in children and young people and health outcomes measured in adulthood. The period from age 25 onwards will examine outcomes for these groups in adulthood. The researchers will draw on the published DfE definition for vulnerable children and young people. This relates to children and young people aged 0-25 years who are assessed as being in need under section 17 of the Children Act 1989 (i.e. have a child in need plan, child protection plan, or are a looked-after child), have an education, health and care (EHC) plan or have been assessed as otherwise vulnerable by educational providers or local authorities (e.g. children on the edge of receiving support or those at risk of becoming not in employment, education or training). UCL draw on work on definitions of vulnerability by DfE (as above) and Public Health England, and consider three broad groups: - clinically vulnerable children and young adults (those with chronic mental or physical health conditions) - socially vulnerable (those receiving statutory support from social care as a child in need, or from education services, for example as special educational needs support or pupil referral unit) - those at high risk of being vulnerable due to social circumstances but not known to be receiving state support (eg: referred to social care but not receiving services (i.e. not a child in need); receiving free school meals, living in a deprived neighbourhood, high risk of becoming not employed, in education or training (NEET)). [1 paragraph unchanged] Children with indicators of vulnerability can only reliably be identified through linkage of health, education and social care data. [1 paragraph unchanged] RQ1: What are the differences in emergency hospital contacts during the COVID-19 pandemic for vulnerable CYP children and individuals compared with other CYP? children and individuals? Is there any evidence that differences are related to COVID-19 infection or the secondary effects of lockdown? [1 paragraph unchanged] The researcher will use longitudinal linked data from hospital episodes statistics (HES), linked to education and social care data (held by DfE) to assess the impact of the COVID-19 pandemic on CYP and in particular vulnerable CYP. As vulnerable CYP are hard to identify in healthcare records, the researcher will identify these CYP through administrative data histories of ever being a Child in Need (CiN), having special educational needs (SEN), a chronic health condition requiring hospitalisation, or combinations of these exposures. The researcher will derive these vulnerability indicators from a linked longitudinal dataset comprising social care, education and hospital records (HES) for all CYP in England. Examining health data from the time of birth to current age (up to, but not including, age 25 years) is critical for identifying markers of vulnerability in administrative data. For example, previous work completed by UCL has shown that chronic underlying conditions, or congenital disorders associated with special education needs may not be recorded at every admission (e.g., asthma may not be recorded when a child is admitted for an operation) and UCL have demonstrated the added value of using the whole longitudinal record. Addressing these questions requires understanding of the causal effects of vulnerability status on outcomes pre- and post- COVID-19. To enable the analyses to address these research questions, the researcher will link HES data (i.e. HES APC, outpatient, critical care, A&E and ECDS data, plus death registration data) to administrative data contained in the datasets collectively supplied within the National Pupil Dataset (NPD), provided by DfE (the researcher refers to NPD data as education, CiN, and children looked after (CLA)). These datasets (HES-NPD) will be linked by NHS Digital for children and young people in England using pseudonymised linkage keys. The researcher will therefore: develop phenotypes and coding clusters to define vulnerable groups; evaluate relationships between vulnerability groups across the life course; and evaluate health, education and social care outcomes in comparator cohorts before and after the onset of COVID-19. Vulnerable cohorts are defined by exposure during childhood and youth. Health, education and social care outcomes can be measured before age 25. The period from age 25 onwards will examine health outcomes in adulthood for cohorts with and without exposure to vulnerability during childhood and youth. The mid- to long-term impact of delays, or withdrawal of healthcare, education and social care during the COVID pandemic (the deficit in care) require information on expected outcomes without such delays from similar cohorts followed up pre-COVID (i.e. what would have happened had COVID not occurred?). The legal basis for processing personal data for this purpose data at UCL falls under Article 6(1)(e) of the General Data Protection Regulations (GDPR), i.e. “a task carried out in the public interest”. It also falls under Article 9(2)(j), “processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes”. The processing of data for this study is a task of public interest as it will provide evidence on the effect of the COVID-19 pandemic on health outcomes and use of healthcare services among vulnerable children. This will benefit and inform policy makers, service providers, vulnerable children and their families. As vulnerable children or individuals are hard to identify in healthcare records, the researcher will use administrative data histories of ever being a Child in Need (CiN), having special educational needs (SEN), frequent absences, a chronic health condition requiring hospitalisation, or combinations of these exposures. The researcher will derive these vulnerability indicators from the linked longitudinal ECHILD dataset. Examining health data across the life course is critical for identifying markers of vulnerability and health outcomes in childhood and adulthood. For example, previous work completed by UCL has shown that chronic underlying conditions, or congenital disorders associated with special education needs may not be recorded at every admission (e.g., asthma may not be recorded when a child is admitted for an operation) and UCL have demonstrated the added value of using the whole longitudinal record. All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract ie: employees, agents and contractors of the Data Recipient who may have access to that data). Previous birth contributes to vulnerability status because teenage motherhood at first live-birth is recognised as a social risk factor, which influences a child’s health outcomes, even for subsequent children born to older mothers. Ethnic background will not be categorised as a component of any vulnerability indicator. UCL consider ethnic background as potentially influencing associations between vulnerability indicators and outcomes (i.e. ethnic background will be analysed as a confounder). For example, UCL will explore whether the proportion of children with vulnerability indicators or with adverse health outcomes, vary according to ethnic background, and if so, include ethnic background in statistical models that seek to determine the influence of vulnerability on child outcomes, after adjusting for ethnic background. To enable the analyses to address these research questions, the researcher will link HES data (i.e. HES APC, outpatient, critical care, A&E and ECDS data, plus death registration data) to administrative data contained in the datasets collectively supplied within the National Pupil Dataset (NPD), provided by DfE (the researcher refers to NPD data as education, CiN, and children looked after (CLA)). These datasets (HES-NPD) will be linked by NHS Digital for children and individuals in England who are aged between 0 and 34 years at the time of the COVID-19 pandemic using pseudonymised linkage keys. The legal basis for processing personal data for this purpose data at UCL falls under Article 6(1)(e) of the General Data Protection Regulations (GDPR), i.e. “a task carried out in the public interest”. It also falls under Article 9(2)(j), “processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes”. The processing of data for this study is a task of public interest as it will provide evidence on the effect of the COVID-19 pandemic on health outcomes and use of healthcare services among vulnerable children and individuals. This will benefit and inform policy makers, service providers, vulnerable children and their families. All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract ie: employees, agents, PhD students and contractors of the Data Recipient who may have access to that data). Patient and public information groups have been engaged with. o The project has seen 8 engagements so far, with 2 more planned during 2022. o This includes advocacy and representative groups that were brought together for the ECHILD public stakeholder event on 29 April 2021. Amongst others, this included the Children’s Commissioner for England, NSPCC, SCOPE, NASEN, Contact, Council for Disabled Children, Down's Syndrome Association, MENCAP, GenerationR Alliance, Parentkind (PTA UK). o A nationwide survey of parents and carers of disabled children and young people is being finalised with SCOPE. Fieldwork is expected to begin in September.

Processing activities

The application shows data held, this data is what is held under NIC-393510. This agreement (NIC-381972) will allow the data to be linked to that which is already held at UCL. The only data requested under NIC-381972 is a record identifier which allow the customer to link those matched with the ECDS data held under NIC-393510, this is because there are no HES IDs in ECDS. The activities outlined below happened under v0 of this agreement but only for those born on or after 1.9.1995. Version 1 of the agreement will follow the same activities but for those born on or after 1.9.1984. The application shows data held and requested which is the data that is held and requested under NIC-393510-D6H1D. This agreement (NIC-381972-Q5F0V) will allow the data to be linked to that which is already held at UCL. The only data requested under NIC-381972-Q5F0V is a record identifier which allow the customer to link those matched with the ECDS data held under NIC-393510-D6H1D, this is because there are no HES IDs in ECDS. [1 paragraph unchanged] The researcher will analyse outcomes for all CYP, children and individuals and compare differences in health outcomes (e.g. emergency hospital contacts, deaths) between [36 words unchanged] adversity, or acute complications of chronic or complex health conditions for all CYP, children and individuals, and according to whether they had indices for vulnerability or not. The [20 words unchanged] whether hospital contacts are related to underlying chronic conditions. Preliminary results October 2020. 2021. [1 paragraph unchanged] The researcher will model expected healthcare contacts for all CYP children and individuals after the onset of the COVID-19 pandemic, based on observed trajectories of healthcare contact for CYP children and individuals for periods prior to the start of COVID-19. The researcher will assess [81 words unchanged] post-COVID-19 period, and potential long-term outcomes of COVID-19 restrictions. Preliminary results in Jan 2021. January 2022. The researcher will also conduct analyses of all CYP children and individuals to explore associations between inequalities (using index of multiple deprivation), ethnic group, and vulnerable vs other CYP, children and individuals, and outcomes measured in health care, NPD data (i.e. education and CiN/CLA) in periods before, during and after the COVID-19 pandemic. Funding for these analyses is provided through the NIHR Policy Research Programme, NIHR Programme for Applied Research, Health Data Research UK and Administrative Data Research UK (see section 8). The researcher will refresh the linked NPD datasets annually in 2021 and [11 words unchanged] on health, education and social care outcomes for vulnerable, compared with other CYP. children or individuals. To address these research questions, HES (APC, A&E, OPD, ECDS, critical care) [35 words unchanged] pseudonymised linkage key, these datasets will be linked for all children and young people (CYP) individuals born in England on or after 1.9.1995. 1.9.1984. The Further information with regards to how data minimisation has been applied including why the full cohort is required (longitudinal data for all children and young people individuals born in England) England on or after 1.9.1984) is justified for the following reasons: [1 paragraph unchanged] Examining health data from the time of birth to current age (up to, but not including, age 25 35 years) is critical for identifying markers of vulnerability in administrative data. For [84 words unchanged] requested 60% of available inpatient fields, with no sensitive or identifiable fields). [1 paragraph unchanged] The research aims to draw conclusions that are valid for all children and young people individuals in England. Yet the pandemic has had differential impacts across the country [110 words unchanged] The researchers have requested the minimum granularity possible, for example by requesting MSOA Middle layer Super Output Area rather than LSOA. Lower Super Output Area. [1 paragraph unchanged] The research focuses on the impact of the pandemic and lockdown on vulnerable children and young people individuals (further defined below). Reliable identification of children and individuals meeting this definition is not trivial, requiring longitudinal data from birth across [8 words unchanged] there are relatively few robust estimates of the size of this population. However, there is good evidence that these indicators of vulnerability are common. [8 words unchanged] all children are ever designated a child in need and that 44% (of the 25%) are ever referred to children’s social care before the age of 16 [101 words unchanged] related factors such as local environment, access to schools and healthcare needs. The researchers therefore require data for all children and young people individuals born in England on or after 1.9.1984 as without these data our comparisons would be incomplete, at greater risk of selection bias and not generalisable. [1 paragraph unchanged] 1) DfE will supply the Trusted Third Party (NHS Digital) with a [14 words unchanged] and sex, alongside a study specific pseudonymised linkage key known as the anonymized anonymised Pupil Matching Reference (aPMR). The identifying variables will be used for linkage to the Personal Master Person Service (MPS)/Personal Demographic Service (PDS) (as previously done for NIC 27404). (PDS). DfE will transfer the variables for any CYP born on or after cohort inception (1.9.95). (1.9.84). 2) NHS Digital will match the identifiers from DfE to records held in the PDS MPS/PDS using an algorithm that makes use of the chronology of postcodes in NPD and PDS. MPS/PDS. Matching to PDS MPS/PDS data will be done internally within NHS Digital, no PDS MPS/PDS data will be disseminated to ONS SRS or UCL Data Safe Haven. NHS Digital will link the the NPD pseudonymised linkage key (i.e. anonymised PMR or young person ID) to PDS, MPS/PDS, and then to the ECDS data. 3) For those CYP whose NPD identifiers were matched to PDS, MPS/PDS, onward linkage to HES-mortality data will occur within NHS Digital to link [14 words unchanged] match rank (denoting the step at which the match to HES and PDS MPS/PDS was made) for these linked cases to the UCL Data Safe Haven for linkage to the existing HES-mortality extract held by UCL (NIC-393510). (NIC-393510-D6H1D). 4) UCL will extract the HES-mortality data for all CYP born on or after 1.9.1995, 1.9.1984, using the existing HES extract (NIC 393510), extract, including a pseudonymised mother-baby link and additional HES records of mothers, and link the aPMR and match rank statistics for those CYP that [60 words unchanged] children born in July/August). The attribute data will also include high-level categorical maternal indicators relevant to birth (e.g. parity – 0, 1, 2+ prior births), [19 words unchanged] analysis files, that are covered by this DSA (e.g. diagnosis codes and mother or baby tail) and for which we have the appropriate permissions. 5) DfE will supply ONS SRS with requested de-identified attribute data extracts, with the aPMR for all CYP born on or after 1.9.1995. 1.9.1984. The deidentified attribute NPD and HES data will be linked within the [17 words unchanged] for the project, with strict output controls applied by ONS SRS staff. [1 paragraph unchanged] 7) NHS Digital will retain the identifier file of all individuals linked in NPD-PDS NPD-MPS/PDS and PDS-HES MPS/PDS-HES and all the postcodes used in linkage and postcode dates for 12 24 months to address data queries or potential linkage errors. errors (v1 note: still being retained as 24 months has not yet passed since the linkage has been undertaken). This data set will not contain any attribute data and will be accessible only to NHS Digital staff. At the end of the 12 24 months, NHS Digital will confirm deletion of the data to DfE. NHS Digital will not send confidential data to DfE or UCL DSH. LSHTM and IFS will each have a named researcher and PI investigator who will undertake analyses (on the ONS SRS) relating to a specific [46 words unchanged] conditions. These named researchers will be substantive employees of the respective organisations. These researchers are undertaking this analyses under instruction of UCL. [5 paragraphs unchanged] CLOUD SECURITY AMENDMENT NHS Digital security has provided assurance regarding the use of the Office of National Statistics' Secure Research Statistics service (ONS SRS), hosted by CloudUK Ltd in this application. The Office of National Statistics has submitted a selection of security documentation to support the use of cloud storage. NHS Digital Security have reviewed the documentation and provided relevant feedback, where necessary. NHS Digital are satisfied that the documentation demonstrates the level of security and governance in place. This study’s extension of age range to all individuals in England born on or after 01.09.1984 and use of the mother and baby/babies flag will allow evaluation of health outcomes in adulthood that may be influenced by health, education and social care in childhood. UCL will also assess the influence of health, education and social care risk factors among women who give birth on health outcomes in their child. The Office of National Statistics have supplied evidence to support: • The use of the Data Risk Model to assess the Risk Profile Class. • Risk Management of the use of the Cloud for this data, taking into consideration Confidentiality, Integrity and Availability. • The use of Pseudonymisation. • Board level involvement in the Risk Management Process evidenced through Minutes of these meetings. • Understanding of the Shared Responsibility Model The Office of National Statistics have a very good understanding of the security controls available to them to provide the appropriate controls to secure data in the Cloud. Using the Cloud, benefits from the inherited controls that cannot practically be replicated locally such as Physical Controls, Resilience of Systems, Power Supplies, Communications and Geographically dispersed Data Centres within a region. Elasticity in provisioning is also a consideration that benefits organisations in managing workloads. The Cloud provider, CloudUK, will use UK Data Centres only.

Expected output

[1 paragraph unchanged] Preliminary reports will be shared with DHSC, PHE and NHS England, and [6 words unchanged] and a project advisory group. Preliminary results will be produced for October 2020 2021 (RQ1) and January 2021 2022 (RQ2). Reports on linkage evaluation will be published in 2021. An introductory guide to ECHILD has been published on the ECHILD website www.ucl.ac.uk/child-health/echild. Reports describing patterns of vulnerability status, and the impact of vulnerability status on outcomes before and after the pandemic, will be published in 2022. The researcher will submit full reports for fast-track publication in peer reviewed journals and produce briefing reports for DHSC, DfE and other public bodies through the Children and Families Policy Research Unit (CPRU). Findings will also be used in public involvement and engagement events. Study findings will be also disseminated through peer-reviewed academic journals (e.g. BMJ, Lancet Public Health), and social media including lay summaries. Reports on linkage evaluation will be published in 2021. An introductory guide to ECHILD has been will be published on the ECHILD website www.ucl.ac.uk/child-health/echild. Reports describing patterns of vulnerability status, and the impact of vulnerability status on outcomes before and after the pandemic, will be published in 2022-2024. The researcher will submit full reports for publication in peer reviewed journals and produce briefing reports for DHSC, DfE and other public bodies through the Children and Families Policy Research Unit (CPRU). Findings will also be used in public involvement and engagement events. Study findings will be also disseminated through peer-reviewed academic journals (e.g. BMJ, Lancet Public Health), and social media including lay summaries. [4 paragraphs unchanged]

Expected measurable benefits

This project aims to produce urgent results on the impact of COVID-19 infection and lockdown on the health of children and young people individuals and in particular vulnerable children, characterised by education and social care indices from the NPD linked datasets. The It is hoped the study will provide vital understanding of the repercussions of the current response [19 words unchanged] strategies should be developed to better meet the needs of children and young people. individuals in England who are aged between 0 and 34 years at the time of the COVID-19 pandemic. These results (preliminary results in October 2020 2021 and January 2021) 2022) are critical for addressing current health needs arising from COVID-19 infection and responses, and also for informing strategies for future waves of infection. The It is hoped the study will compare different groups of vulnerable and non-vulnerable groups of children and young people, individuals, using indicators of vulnerability drawn from health, social care and education histories in administrative data. The It is hoped the study will estimate impacts of service support (from schools, social care or hospitals) pre-COVID in order to predict detrimental effects of reduced support during COVID. It is hoped the analyses will address a priority for DHSC policy makers, that COVID-19 and lockdown have resulted in disproportionate impacts on health for some vulnerable groups. The analyses aim to explore this question for the whole population, to inform policy to better support children and young people, individuals, and to better understand which types of vulnerability are most affected. The study will examine which groups of children and young people did present to services during the lockdown, whether their problems were directly related to COVID19 or the secondary impact of lockdown, and what underlying health or social risk factors were present. The study also aim to understand unmet need and predict future health needs for the large proportion of CYP who would have been expected to present to services, based on past patterns of care, but did not attend during the pandemic. For example, despite messages to urge patients requiring urgent medical treatment to seek care through the appropriate channels (e.g. A&E), during the pandemic there was a dramatic and unexplained decrease in A&E attendances. Serious concerns have been raised about the impact of the resulting treatment delays, yet much more evidence is required in to quantify the scale of the problem in different population groups and to predict future/ongoing needs. Concerns have been raised about the impact of the resulting treatment delays, yet much more evidence is required in to quantify the scale of the problem in different population groups and to predict future/ongoing needs. The research seeks to help to fill this gap, firstly by evaluating high level differences in impacts for groups of vulnerable (in terms of clinical, socio-demographic socio-demographic, social care and educational needs) versus other children and young people, individuals, which will guide the development of more detailed, in depth research within [35 words unchanged] health and education outcomes (e.g. surgical correction of cleft lip and palate). The It is hoped the results will establish the scale and urgency (e.g. how many children, how extensive were the delays, what are likely the expected unmet healthcare and education needs) of these impacts and guide the development of reactive policies and changes in service provision to mitigate long-term impacts for specific population groups. The It is hoped the research will also add evidence on the impact of COVID-19 on health [12 words unchanged] mechanisms that drive these. The comprehensive geographical coverage and population base of our the research is a real strength of this research and it is hoped it will allow the researchers to draw conclusions for all vulnerable children and young people individuals in England, England who are aged between 0 and 34 years at the time of the COVID-19 pandemic, and to identify groups that are being failed who might be disadvantaged by current policy and or need more support from services. The project is commissioned and funded by DHSC, and findings Findings will be reported directly to DfE and NHS policy makers. The project also addresses two priorities on the impact of COVID19 on vulnerable patients set out by Health Data Research UK. The study will report is reporting preliminary results to DHSC policy makers through our regular 2-monthly meetings, through seminars with wider NHS staff (DHSC, PHE and NHS England – Simon Kenny, NHSE clinical director), England), and through briefing reports and papers published in peer reviewed journals.

Benefits reported

Yielded Benefits is not a requirement for new applications. Since June 2021, when linked ECHILD data could first be accessed, UCL have published the following research findings. i) UCL showed that the quality of data linkage between schools and hospital data was good and improved over time. However, those not linked were disproportionately poor or from certain ethnic groups. This information can be used to reduce biases in linkage and in analyses. (please see the "Linking education and hospital data in England: linkage process and quality" paper published in IJPDS Vol. 6 No. 1 (2021), available: https://doi.org/10.23889/ijpds.v6i1.1671 and "Ethnic bias in data linkage" correspondence in The Lancet Digital Health, Vol 3, Issue 6, E339, (2021), available: https://doi.org/10.1016/S2589-7500(21)00081-9) ii) Recent analyses showed that children who were vulnerable, due to contact with social services or because they received special educational needs support for additional learning needs, had a much greater deficit in hospital care during the COVID pandemic than their peers. Findings indicate a need for targeted ‘catchup’ funding and resources for child health, particularly for vulnerable children who were affected disproportionately. For example, the ring-fenced resource for ‘catch-up’ of NHS care (Health and Social Care Levy) might be further targeted for the vulnerable groups that have disproportionally missed out on hospital contacts. Secondary school pupils receiving special educational needs support or social care services may need to be prioritised for face-to-face outpatient care as it is unclear how effective remote care is for these children (Please see report: “Changes in hospital contacts during the COVID-19 pandemic among vulnerable children and young people”, 4th November 2021. available: https://www.ucl.ac.uk/children-policy-research/projects/assessing-impact-covid-19-pandemic-vulnerable-children).