Family, household and environmental risk factors for hospital admissions in childhood

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

Objective for processing

University College London (UCL) requires access to a birth cohort, which includes HES admitted patient care (APC) data already disseminated under NIC-10094-P6P4B-v4.2. NIC-10094-P6P4B. The data for Agreement NIC-10094-P6P4B-v4.2 Data is held by the Office of National Statistics (ONS) as part of a birth cohort within their Secure Research Service (SRS). The following is a summary of the aims of the research project provided by UCL. [18 paragraphs unchanged] UCL as the research sponsor, and City City, St George's University of London (CUL) as the main collaborator, are joint controllers as the organisations responsible for ensuring that the data will only be processed for the purpose described above. Although CUL City, St George's University of London have been listed as a joint controller, CUL City, St George's University of London are not accessing the data under this agreement. Although City City, St George's University of London (CUL) have been listed as a joint controller, CUL City, St George's University of London are not accessing the data under this agreement. [10 paragraphs unchanged] University of Edinburgh and University of Tampere are part of the study advisory group. City City, St George's University of London is joint controller of the data. [4 paragraphs unchanged]

Processing activities

[4 paragraphs unchanged] The data will be accessed onsite at the premises of UCL. Additionally Additionally, the data will be accessed by authorised personnel via remote access. [46 paragraphs unchanged]

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

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

Objective for processing

In this project, researchers at University College London (UCL) will examine environmental and household risk factors for preventable hospital admissions in children, and whether children whose parents were born abroad face barriers to accessing preventive primary and community health services, which in turn leads to the need for hospital admission. UCL plan to base the analyses on the same data disseminated for the related project NIC-10094-P6P4B-v4.2. The data disseminated under DARS-NIC-10094-P6P4B was for the use of the data for a project led by City, University of London entitled ‘Births and their outcome: analysing the daily, weekly and yearly cycles and their implications for the NHS.’ The aims of this project were to develop a birth cohort database based on administrative health data and compare daily, weekly and yearly variations in numbers of spontaneous and other births by time, day and season of birth and to compare variations in rates of adverse outcomes. The project involved birth registration records already linked by ONS to birth notification records, previously known as the NHS Numbers for Babies (NN4B) records being linked to HES data. University College London (UCL) requires access to a birth cohort, which includes HES admitted patient care (APC) data already disseminated under NIC-10094-P6P4B-v4.2. The data for Agreement NIC-10094-P6P4B-v4.2 is held by the Office of National Statistics (ONS) as part of a birth cohort within their Secure Research Service (SRS). Agreement DARS-NIC-10094-P6P4B-v4 states : ‘Collaborators (including UCL) cannot access the data provided under this agreement until the signed collaborative agreement has been provided to NHS Digital.’ The collaborative agreement has been signed and provided to NHS Digital. Therefore, UCL has met the conditions set out under Agreement DARS-NIC-10094-P6P4B-v4 which permits them to receive data. The following is a summary of the aims of the research project provided UCL. UCL researchers will use an existing birth cohort of all babies born in England between 2005 and 2014 for this project. This birth cohort database is held in the Office for National Statistics (ONS) Secure Research Service (SRS) as a result of linkage of: UCL will examine environmental and household risk factors for preventable hospital admissions in children, and whether children whose parents were born abroad face barriers to accessing preventive primary and community health services, which in turn leads to the need for hospital admission. -ONS birth registration data UCL require access to the existing linked database to extend the analyses carried out for project NIC-10094-P6P4B-v4.2, which focused on the role of birth hospital and obstetric practices on child health outcomes, including mortality and hospital admissions during infancy. UCL researchers want to build on these results to examine further risk factors for potentially preventable hospital admissions in young children. -ONS death registration data The research objectives are to: -ONS stillbirth registration data -NHS birth notification data -HES APC data for mothers and babies (NHS Digital data) - Demographics data These datasets were linked by researchers at City, University of London for a previous project (Data Access Agreement NIC-10094-P6P4B-v4.2). City, University of London and UCL are collaborators for the current project (DARS-NIC-234656-C3J1D). This is why City, University of London is also the joint data controller for this project, together with UCL. The processing for this project will be carried out by ONS, who will link the existing birth cohorts to further datasets, and UCL, who will carry out the analyses on the linked, pseudonymised data. This is why UCL and ONS are the data processors. The identifiers (maternal and child dates of birth, NHS numbers and maternal postcodes at delivery) are kept in a separate database, accessible only by ONS staff members in the Health Analysis and Life Events Team. UCL are asking NHS Digital for permission to use the existing linked database to extend the analyses carried out for project NIC-10094-P6P4B-v4.2, which focused on the role of birth hospital and obstetric practices on child health outcomes, including mortality and hospital admissions during infancy. UCL researchers want to build on these results to examine further risk factors for potentially preventable hospital admissions in young children. The research objectives are to: [1 paragraph unchanged] The research questions for this objective are: [1 paragraph unchanged] 1b) What is the association between adverse housing conditions (for example overcrowding, indoor air pollution exposure) and risk of admission for respiratory tract infections or injuries admissions in children? 2. determine the impact of family factors i.e. country of birth, year [8 words unchanged] to the risk of hospital admissions in children, focusing on hospital admissions (such as injury admissions) which are considered preventable via timely access to preventive primary and community health services. services or related to housing characteristics. The specific research questions for this objective are: [2 paragraphs unchanged] To carry out these two projects, UCL are proposing to link the existing birth cohort, via the ONS birth registration records, to: The following NHS England data will be accessed: a) 2011 Census data - Hospital Episode Statistics Admitted Patient Care – necessary to provide longitudinal information on hospital admissions in mothers and children for respiratory tract infections and hospital admissions (such as injury admissions) which are considered preventable via timely access to preventative primary and community health services or related to housing risk factors. b) Air pollution data: - Demographics data – necessary to link patient to their Census data -Openly available data on annual averages of 8 pollutants modelled at 1x1 km grids across the British Isles: https://uk-air.defra.gov.uk/data/pcm-data, available 2005-2014; Note these data are openly available online The level of the data will be: -distance to A roads for each postcode in England, mapped from Open StreetMap (https://www.openstreetmap.org/) available for 2005-2014; Note these data are openly available online • Identifiable, Identifiable data was used for the linkage under a previous version of this DSA (NIC-234656v2.12) although this linkage has been completed and no further identifiable data processing will be required. The identifying details will be stored in a separate database to the linked dataset used for analysis. -daily exposures for ozone, PM2.5 and Nitrogen dioxide modelled at 100m x 100m grid across London only for 2010-2014). These data are not openly available; they have been provided for this project by Cambridge Environmental Research Consultants (http://cerc.co.uk/). The data will be minimised as follows: c) building characteristics data, available at address (dwelling) level from Energy Performance Certificates: available 2005-2014. Note these data are openly available online from: https://epc.opendatacommunities.org/ The EPCs will be aggregated up to postcodes for linkage purposes (see below). - Only babies born between 2005 and 2014 will be included, d) Modelled tobacco expenditure data at output area level for 2010 and 2012. Note these data are not openly available, they have been provided for this project by CACI Ltd (https://www.caci.co.uk/). - Additionally, mothers’ episodes for the same period, but only those episodes associated with the delivery. NOTE: the air pollution data, building characteristics data and modelled tobacco expenditure data do not relate to an individual. They are not personal data. Identifiable data (NHS number) was only retained for a limited time period for the purpose of data quality assurance and then securely destroyed. The Census data will be linked to the existing cohort via the mothers' date of birth, forename and surname and postcode at the time of delivery (as recorded on ONS birth registration records). Only births occurring up to 12 months before and until 12 months after the Census (that is, births between March 2010 and March 2012) will be linked to 2011 Census data. The air pollution, buildings and tobacco expenditure data will be linked via the maternal and child postcodes on the Personal Demographic Service (see below). Therefore, although the HES APC or NHS birth notification data in the cohort will not be used for linkage to Census, buildings or air pollution data, the resulting linked dataset will include birth and death registration data, NHS birth notification data, and HES APC data for mothers and babies, Census 2011 data, and postcode-level air pollution and building characteristics data and output area level tobacco expenditure data.. UCL as the research sponsor, and City University of London (CUL) as the main collaborator, are joint controllers as the organisations responsible for ensuring that the data will only be processed for the purpose described above. Although CUL have been listed as a joint controller, CUL are not accessing the data under this agreement. Although City University of London (CUL) have been listed as a joint controller, CUL are not accessing the data under this agreement. The purpose of this application falls under Article 6 (1) (e) of the GDPR and the lawful basis for using information collected routinely for administrative purposes for research is the ‘public task’. This is part of the University’s commitment to ‘integrate research and innovation for the long-term benefit of humanity’. The application also falls under Article 9 (2) (j), as scientific research. The lawful basis for processing personal data under the UK GDPR is: Article 6(1)(e) - processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller. To allow NHS Digital to apply patient objections to the data. ONS will send the encrypted HESIDs (the patient identifier in HES) disseminated under DARS-NIC-10094-P6P4B to NHS Digital. NHS Digital would decrypt this data, run it through the patient objections process to remove those who have objected. NHS Digital would then encrypt the HES IDs before flowing the data back to ONS. The data processor as outlined in this agreement would only be permitted to access this ‘new’ data. The lawful basis for processing special category data under the UK GDPR is: Amendment Request: Article 9(2)(j) - processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject. UCL request to make the following amendment: The funding comes from multiple sources. Current funders include: • Link the mothers and babies in the birth cohort to longitudinal environmental exposure (buildings and air pollution data) via linkage to maternal and child postcode histories in the Personal Demographic Service (PDS), held by NHS Digital. ONS will transfer identifiers for mothers and babies in the birth cohort (NHS number, baby sex, date of birth, postcode at birth/delivery) and a study ID to NHS Digital. NHS Digital will link the mother & baby identifiers to PDS and extract longitudinal postcode histories and time stamps (date of change of postcode on PDS). NHS Digital will then transfer the postcode histories to ONS. UCL will link the longitudinal postcode histories to postcode level environmental data on air pollution, distance to road, tobacco expenditure and buildings data in the ONS SRS. These postcode-level environmental datasets will be uploaded to the ONS SRS by UCL. Following linkage, only Study IDs, longitudinal air pollution and buildings exposure histories, and time stamps will be retained for analysis in the ONS SRS. • National Institute for Health Research (NIHR) – Funding was in place until September 2022. The aim of this amendment to longitudinal environmental exposure data is to improve the accuracy of measurement of exposures to potential environmental hazards during pregnancy and childhood. • Medical Research Council (MRC) – Funding is in place until March 2024 • Health Data Research-UK – Funding is in place until March 2028. Funding to continue the work described will be sought on an ongoing basis. The funder(s) will have no ability to suppress or otherwise limit the publication of findings. The Office of National Statistics Secure Research Service (ONS SRS) provides IT hosting services to UCL and will store the data as contracted by UCL. ONS is a processor acting under the instructions of UCL. ONS will process data as a third-party processor under this agreement. Only ONS will have access to identifiers already held from the birth cohort. University of Edinburgh and University of Tampere are part of the study advisory group. City University is joint controller of the data. The University of Edinburgh are only involved in the Scottish aspect of the study only and any associated analysis of Scottish data, and the London School of Hygiene and Tropical Medicine is undertaking some preparatory mapping of the air pollution data only, to which the birth data will be linked. Data will be accessed by: A PhD student affiliated with UCL. The individual has completed mandatory data protection and confidentiality training and is subject to UCL’s policies on data protection and confidentiality. The individual accessing the data will do so under the supervision of a substantive employee of UCL. UCL would be responsible and liable for any work carried out by the individual. The PhD student would only work on the data for the purposes described in this Agreement. UCL has held meetings with parents in collaboration with two charities (The British Lung Foundation, through their partner organisation The Clean Air Parents’ Network, and Shelter) to update them on the results and discuss how they can best be communicated to the general public. UCL researchers will consult the Great Ormond Street Biomedical Research Centre Young Person's Advisory Group, and the Parents' and Carers' Advisory Group regarding how to communicate results to a broad public of parents and children. UCL staff have visited primary schools to speak to children about environmental epidemiology and how researchers use data for child health research. UCL previously held a meeting with parents who were born abroad but had children whilst living in the UK in December 2021 to discuss the project, including how to interpret and present results.

Processing activities

There are two proposed linkages with the existing birth cohort data: No data will flow to NHS England for the purposes of this Agreement. NHS England will link the mother & baby identifiers (supplied by the ONS in question 1) to PDS and extract longitudinal postcode histories and time stamps (date of change of postcode on PDS). NHS England will then transfer the postcode histories and study IDs to ONS. The data will contain a directly identifying data item (postcodes) which is required for linkage to postcode level environmental data held on the Office for National Statistics (ONS) Secure Research Service (SRS), and will be deleted once linkage has been completed. UCL will link the longitudinal postcode histories to postcode level environmental data on air pollution, distance to road, tobacco expenditure and buildings data in the ONS SRS. These postcode-level environmental datasets will be uploaded to the ONS SRS by UCL. Following linkage, only Study IDs, longitudinal air pollution and buildings exposure histories, and time stamps will be retained for analysis in the ONS SRS. The data will be stored on servers at the ONS SRS. The data will be accessed onsite at the premises of UCL. Additionally the data will be accessed by authorised personnel via remote access. The data will remain on the servers of the ONS SRS at all times. Personnel are not technically capable of downloading or copying data to local devices. The data will not leave England/Wales at any time. Access is restricted to substantive employees or PhD students working under the supervision of substantive employees of UCL who have authorisation from the Principal Investigator. All personnel accessing the data have been appropriately trained in data protection and confidentiality. This agreement is an extension request for UCL to continue using the linked data. [2 paragraphs unchanged] These are described in detail below. The existing birth cohort data is the result of linkage of Hospital Episode Statistics (HES) admitted patient care (APC) data (data previously disseminated by NHS Digital, England, and to which this agreement relates) and the following disseminated by ONS: -ONS - ONS birth registration data -ONS - ONS death registration data -ONS - ONS stillbirth registration data -NHS - NHS birth notification data. data Note that UCL are not asking NHS Digital to resupply any HES data, nor the birth, stillbirth, death registrations data or birth notification data. The existing database held in the ONS SRS has been cleaned and validated and so UCL does not wish to have any of the HES or civil registration or birth notification data resupplied. This is why UCL are proposing ONS supply the identifiers for the existing cohort for [8 words unchanged] ‘2- Proposed linkage between birth cohort and postcode-level air pollution and buildings data’. data. [1 paragraph unchanged] NOTE (DECEMBER 2020): THIS PART OF THE LINKAGE HAS ALREADY BEEN COMPLETED [1 paragraph unchanged] Births between 27th March 2010 and 27th March 2012 (that is, births [18 words unchanged] 2011 Census questionnaires completed by the mothers in the cohort and their partners(or partners (or by the household reference person in the household in which the mothers live if this is not the mother or her partner). [11 paragraphs unchanged] 3) The ONS Data processing team will clean, hash and encrypt the [102 words unchanged] the original identifiers are deleted. The encryption is done by creating 11 matchkeys match keys for each individual, based on combinations of the variables forename, surname, initial, [10 words unchanged] variations in spelling discrepancies, forename/surname transpositions or moving between postcodes. The 11 matchkeys match keys for each individual are then encrypted (hashed). The original identifiers are deleted. It is the encrypted matchkeys match keys that are linked. This is done hierarchically so that if exact agreement is not reached between two records in the two datasets, (using the hashkey hash key for forename, surname, full date of birth, sex and postcode), the algorithm moves on to matching on less exact matchkeys. match keys. Once the datasets are linked, the hashed identifiers are also deleted. [4 paragraphs unchanged] NOTE: This linkage has not yet been completed (by December 2020) and UCL are seeking an amendment for this part of the linkage [2 paragraphs unchanged] 1) A member of the Health Analysis and Life Events Team at ONS will extract identifiers (NHS number, baby sex, date of birth, postcode at birth/delivery) history) and Study ID from the birth cohort identifier file. 2) ONS will securely transfer these identifiers for mothers and babies in the birth cohort to NHS Digital. England. 3) NHS Digital England will link the mother & baby identifiers to PDS and extract longitudinal postcode histories and time stamps (date of change of address on PDS). NHS Digital England will also extract some further variables, including whether a mother or baby has deregistered from PDS (and if so when). 4) NHS Digital England will return cohort Study ID, postcode histories and the other PDS variables to ONS, who will put the data in the ONS SRS. [5 paragraphs unchanged] Research staff working on analysing the birth cohort with linked Census and post-code level data will require training in data security and information governance, and accreditation as Approved Researchers by the ONS. They will also require substantial contract with UCL. There will also be one UCL-registered PhD student, who is supervised by the lead UCL researcher, working on the data. All researchers working on the data will therefore have a duty of confidentiality enshrined in their UCL contract and are required to follow the UCL data protection policy. NHS Digital draws to the attention of the University College London that NHS Digital regards the University College London as being responsible for the actions and omissions of the PhD student. The identifying details will be stored in a separate database to the linked dataset used for analysis. All analyses will use the pseudonymised dataset. There will be no requirement and no attempt to reidentify individuals when using the pseudonymised dataset. Once linked, UCL researchers Researchers from the University College London (UCL) will check, clean and validate the linked Census and postcode-level air pollution, [149 words unchanged] history, family structure and potentially preventable admissions (project 2) in children, whilst taking into account considering other important risk factors for hospital admissions including preterm birth, and birth weight.

Expected output

Expected outputs: The expected outputs of the processing will be: 1) an enhanced administrative birth cohort including all children born in England between 2005-2014, linked to HES APC records for themselves and their mothers, longitudinal air pollution data and building characteristics data, and Census 2011 data for mothers and their resident partners for a subset of children born between 2010 and 2012 1) Submissions to peer reviewed journals (expected target dates: November 2022 – paper submitted to Lancet Regional Health Europe; March 2024, paper submitted to Lancet Planetary Health; May 2024; paper submitted to Epidemiology; September 2024 paper submitted to BMJ; May 2025 – paper submitted to Environment International; September 2025 -paper submitted to European Respiratory Journal); 2) description of the exposure of pregnant women and young children to indoor and outdoor pollutants over time and by age 2) Presentations at scientific conferences (expected target dates: Health Services UK conference July 2022; British Society for Population Studies – September 2022; International Population Data Linkage Conference – September 2024; International Society for Environmental Epidemiology, September 2024; European Respiratory Society Conference – September 2025) 3) estimates of the association between ambient air pollution exposure during pregnancy and the risk of hospital admission with respiratory tract infections during early childhood The outputs will not contain NHS England data and will only contain aggregated information with small numbers suppressed as appropriate in line with the relevant disclosure rules for the dataset(s) from which the information was derived. 4) estimates of the association between adverse housing characteristics (eg mould, indoor air pollution, overcrowding) and the risk of hospital admission with respiratory tract infections during early childhood The outputs will be communicated to relevant recipients through the following dissemination channels: 5) estimates of the association between maternal and paternal country of birth, family structure, and admission to hospital for potentially preventable causes in children • Journals 6) estimates of the association between maternal and paternal knowledge of English/English proficiency (self-assessed from the Census data) and length of stay in the UK and admission to hospital for potentially preventable causes in children • Workshops involving parents and children Output 1), creation of the birth cohort, including linkage to environmental data, is expected by the end of March 2021. UCL will describe the success of linkage between Census 2011 data and the existing birth cohort, and conduct analyses to validate the linkage, including examining agreement among key variables such as maternal ethnic group and country of birth, that are available both in the existing birth cohort and on the Census records. The resulting Census-linked birth cohort database will be described in a journal publication, such as a Data Resource Profile in the International Journal of Epidemiology. • Webinars open to academics and policy makers Output 2) It is expected these descriptive analyses to be available in the summer of 2021 • Social media Output 3), an estimate of the strength of association between air pollution exposure during pregnancy and the risk of hospital admissions for respiratory tract infections: these results will be published in an academic journal. UCL will target high impact journals including BMJ, Lancet or Environmental Health Perspectives. UCL will also present the results at academic conferences, such as Society for Social Medicine or the Lancet Public Health Science Conference. UCL expect that these results would not be available until the end of the period of this three-year data sharing Agreement period due to the complexity of analyses. • Public reports available from the PICNIC study website Output 4) an estimate of the association between adverse housing characteristics and the risk of hospital admission with respiratory tract infections during early childhood: these results will be published in an academic journal. UCL will target high impact journals such as the Lancet Child and Adolescent Health or Environmental International. UCL will also present the results at academic conferences, such as Society for Social Medicine or the Lancet Public Health Science Conference. UCL expect that these results would not be available until the end of the period of this three-year data sharing Agreement period due to the complexity of analyses. • Direct bilateral engagement with policy makers and arm’s length bodies (including the UK Health Security Agency and Office for Health Improvement and Disparities) via meetings organised by , and the NIHR Children's Policy Research Unit and meetings with UKHSA Output 5), an estimate of the association between mother's and father's country of birth and the risk of potentially preventable hospital admissions: these results will be presented in a scientific journal; UCL will target the Lancet or the Lancet Child and Adolescent Health. UCL will also present scientific conferences, such as the Migration, Ethnicity, Race and Health conference, and the NIHR School for Public Health Research Annual Scientific Meeting. UCL expect these results to be available by June 2021. • Press/media engagement by working with the UCL press office and social media Output 6), estimates of the association between maternal and paternal knowledge of English/English proficiency (self-assessed from the Census data) and length of stay in the UK and admission to hospital for potentially preventable causes in children: these results will be presented in a scientific journal; UCL will target the Lancet Global Health or the BMJ. UCL will also present scientific conferences, such as the Migration, Ethnicity, Race and Health conference, and the NIHR School for Public Health Research Annual Scientific Meeting. UCL expect these results to be available by June 2021. • Working with charities to raise awareness about the work: via charities, including Mums for Lungs, Clean Air Parents Network, British Lung Foundation and Shelter. UCL researchers will communicate the results to policy makers and arms length bodies, including Public Health England, via meetings organised by the NIHR School for Public Health Research, and the NIHR Children's Policy Research Unit. UCL Researchers will use several avenues to communicate results to parents. Updates from the project will be posted on a website hosted by UCL: https://www.ucl.ac.uk/child-health/research/population-policy-and-practice/child-healthinformatics-group. UCL researchers will also work with the UCL press office to communicate results from the studies via press, television and radio outlets, and via social media. UCL researchers have worked with two charities who support this work: Shelter and The British Lung Foundation. The British Lung Foundation, through their partner organisation The Clean Air Parents' Network have agreed that updates about this study can be placed on their website and Facebook page. UCL expect to publish multiple academic papers between October 2023 and December 2025, and present results at academic conferences throughout 2024 and 2025. UCL will also hold meetings with parents in collaboration with these two charities to update them on the results and discuss how they can best be communicated to the general public. The new Policy Research Unit for Children and Families (CPRU), funded by the NIHR, will be setting up a new young people's advisory group who are specifically trained in research using administrative data sources. The principal investigator of this project is also a co-investigator on CPRU. Project 2 (on parents' country of birth and children's access to preventive health services) in particular is highly relevant to the work of the CPRU. UCL researchers will also be able to consult this group of children and young people about the work carried out, and how best to communicate results to parents. Further, UCL researchers will consult the Great Ormond Street Biomedical Research Centre Young Person's Advisory Group, and the Parents' and Carers' Advisory Group regarding how to communicate results to a broad public of parents and children.

Expected measurable benefits

The aim of this project is to identify household, parental and environmental risk factors for hospital admissions in children. UCL will also calculate population attributable risks for each of these factors, meaning that the relative contribution of, for example, overcrowded housing to emergency hospital admissions can be estimated. A key strength of this project is that UCL will be able to measure longitudinal exposures to environmental risk factors across pregnancy and early childhood. The results will allow the design and targeting of interventions, either through the NHS or other organisations (such as local authorities) to reduce the need for children to be admitted to hospital. This project therefore aims to improve the evidence base in order to reduce stress for families, improve children's access to health services and health outcomes both in the short and long term and to reduce costs for the NHS. The expected measurable benefits of the project will be: 1) help the system to better understand the health and care needs of populations – for example, whether children whose parents are born abroad are more or less likely to use emergency secondary care services. 2) lead to the identification or improvement of treatments or interventions, or health and care system design to improve health and care outcomes or experience – for example whether improvements in housing or reductions in air pollution are likely to lead to reduced need for children to be admitted to hospital with respiratory infections. 3) advance understanding of regional and national trends in health and social care needs, including to what extend variation in exposures to outdoor air pollution explains differences in admission rates for respiratory infections across England. 4) inform planning health services and programmes, for example to improve equity of access, experience, and outcomes. – for example, whether children whose parents were born abroad, or living in different family structures have equal access to health services. [2 paragraphs unchanged] 2) The National Health Service (Clinical Commissioning Groups; CCGs): Part 2 of the project will examine whether all children have equal access to healthcare, independent of their parentscountry parents’ country of birth. If inequalities are found, this will provide evidence for CCGs [8 words unchanged] services (this could include extending interpreting services or setting up outreach clinics). 3) Parents and children using the NHS: This project will provide parents with information about the relative importance of [25 words unchanged] to push for improvements in local environments, housing quality, or healthcare access. It is hoped that through publication of findings in appropriate media, the findings of this research will add to the body of evidence that is considered by the bodies, organisations and individual care practitioners charged with making policy decisions for or within the NHS or treatment decisions in relation to specific patients. This project has been supported by charities, including Mums for Lungs, Clean Air Parents Network, British Lung Foundation and Shelter. UCL have visited schools to talk about how epidemiologists use data, including on air pollution and health, to help suggest healthy changes. Some of this work has been in collaboration with Mums For Lungs. UCL have produced a website for the part of the project that covered by MRC funding: https://www.ucl.ac.uk/child-health/research/population-policy-and-practice-research-and-teaching-department/cenb-clinical-31

Benefits reported

No There has been some delay to realising the benefits have yet been yielded. DARS-NIC-234656-C3J1D was originally approved in April 2019; however however, UCL were not able to access the data until an amendment was approved in November 2019. This amendment involved removing opt outs from the cohort. UCL gained access to the data in the ONS SRS in February 2020; however, when COVID lockdown happened UCL lost access to the data for another 2 months, before ONS agreed that UCL could access the SRS from home. UCL were able to access the data 10 months after original approval, since accessing the data, UCL have cleaned and validated the birth cohort. The first outputs are expected to be ready by summer/autumn 2021. UCL gained access to the data in the ONS SRS in February 2020; however, when COVID lockdown happened UCL lost access to the data for another 2 months, before ONS agreed that UCL could access the SRS from home. Due to the COVID pandemic, data flows for this project also had to be amended during 2021. UCL have now yielded the following benefits to date: - Presentation of findings to the Health Services Research UK conference in July 2022 (UCL won an award for Innovation in inclusion) - Presentation of findings to British Society for Population Studies Conference, September 2022 - Paper submitted to Lancet Regional Health Europe

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

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

Objective for processing

In this project, researchers at University College London (UCL) will examine environmental [26 words unchanged] health services, which in turn leads to the need for hospital admission. These analyses therefore extend UCL plan to base the analyses on the same data disseminated for the related project NIC-10094-P6P4B-v4.2. The data disseminated under DARS-NIC-10094-P6P4B was for the use of the data for a project led by City, University of London entitled ‘Births and their outcome: analysing the daily, weekly and yearly cycles and their implications for the NHS.’ The aims of this project were to develop a birth cohort database based on administrative health data and compare daily, weekly and yearly variations in numbers of spontaneous and other births by time, day and season of birth and to compare variations in rates of adverse outcomes. The project involved birth registration records already linked by ONS to birth notification records, previously known as the NHS Numbers for Babies (NN4B) records being linked to HES data. The data disseminated under DARS-NIC-10094-P6P4B was for the use of the data for a specific project led by City, University of London entitled ‘Births and their outcome: analysing the daily, weekly and yearly cycles and their implications for the NHS’. The aims of this project were to compare daily, weekly and yearly variations in numbers of spontaneous and other births by time, day and season of birth and to compare variations in rates of adverse outcome. The project involved birth registration records already linked by ONS to birth notification records, previously known as the NHS Numbers for Babies (NN4B) records being linked to HES data, this data was then added to a database. This database adds value to the HES data by adding further data items from birth registration and notification and more complete data for important data items, notably birthweight and gestational age where HES data was incomplete. It was created to enable ‘analyses relating to inequalities in the outcome of pregnancy and information for service users about the outcome of midwifery, obstetric and neonatal and related health care’. Agreement DARS-NIC-10094-P6P4B-v4 states : ‘Collaborators (including UCL) cannot access the data provided under this agreement until the signed collaborative agreement has been provided to NHS Digital.’ The collaborative agreement has been signed and provided to NHS Digital. Therefore, UCL has met the conditions set out under Agreement DARS-NIC-10094-P6P4B-v4 which permits them to receive data. UCL researchers will use an existing birth cohort of all babies born in England between 2005 and 2014 for this project. This birth cohort database is available held in the Office for National Statistics (ONS) Secure Research Service (SRS) as a result of linkage of: [5 paragraphs unchanged] These datasets were linked by researchers at City, University of London for a previous project covered by Data Access Agreement NIC-10094-P6P4B-v4.2. The data controller for the existing birth cohort is City, University of London, who agreed that this data could be used for this amended purpose by UCL. This is why City, University of London is also the joint data controller for this project, together with UCL. The processing for this project will be carried out by ONS, who will link the existing birth cohorts to further datasets, and UCL, who will carry out the analyses on the linked, pseudonymised data. This is why UCL and ONS are the data processors. The identifiers (maternal and child dates of birth, NHS numbers and maternal postcodes at delivery) are kept in a separate database, accessible only by ONS staff members in the Health Analysis and Life Events Team. - Demographics data These datasets were linked by researchers at City, University of London for a previous project (Data Access Agreement NIC-10094-P6P4B-v4.2). City, University of London and UCL are collaborators for the current project (DARS-NIC-234656-C3J1D). This is why City, University of London is also the joint data controller for this project, together with UCL. The processing for this project will be carried out by ONS, who will link the existing birth cohorts to further datasets, and UCL, who will carry out the analyses on the linked, pseudonymised data. This is why UCL and ONS are the data processors. The identifiers (maternal and child dates of birth, NHS numbers and maternal postcodes at delivery) are kept in a separate database, accessible only by ONS staff members in the Health Analysis and Life Events Team. [5 paragraphs unchanged] 2. determine the impact of family factors i.e. country of birth, year of migration and migration, knowledge of English and family structure to the risk of hospital admissions in children, focusing on hospital admissions which are considered preventable via timely access to preventive primary and community health services. [1 paragraph unchanged] 2a) what is the association between parental country of birth and family structure and risk of hospital admission for conditions preventable through community or primary care in children? 2b) how do parents' knowledge of English/English proficiency (self-assessed from Census data) [16 words unchanged] risk of hospital admission in children for conditions preventable through community or primary care? primary care? [2 paragraphs unchanged] b) Air pollution data (Freely available data on annual averages of 8 pollutants modelled at 1x1 km grids across the British Isles: https://uk-air.defra.gov.uk/data/pcm-data, available 2005-2014; distance to A roads for each postcode in England, available for 2005-2014; daily exposures for ozone, PM2.5 and Nitrogen dioxide modelled at 100m x 100m grid across London only for 2010-2012) b) Air pollution data: c) building characteristics data, available at postcode level from Energy Performance Certificates: https://epc.opendatacommunities.org/ - available 2005-2014 -Openly available data on annual averages of 8 pollutants modelled at 1x1 km grids across the British Isles: https://uk-air.defra.gov.uk/data/pcm-data, available 2005-2014; Note these data are openly available online The Census data will be linked to the existing cohort via the mothers' date of birth, forename and surname and postcode at the time of delivery (as recorded on ONS birth registration records). Only births occurring up to 12 months before and until 12 months after the Census (that is, births between March 2010 and March 2012) will be linked to 2011 Census data. The air pollution and buildings data will be linked via the maternal postcode at delivery - also recorded on ONS birth registration records. -distance to A roads for each postcode in England, mapped from Open StreetMap (https://www.openstreetmap.org/) available for 2005-2014; Note these data are openly available online Therefore, although the HES APC or NHS birth notification data in the cohort will not be used for linkage to Census, buildings or air pollution data, the resulting linked dataset will include birth and death registration data, NHS birth notification data, and HES APC data for mothers and babies, Census 2011 data, and postcode-level air pollution and building characteristics data. -daily exposures for ozone, PM2.5 and Nitrogen dioxide modelled at 100m x 100m grid across London only for 2010-2014). These data are not openly available; they have been provided for this project by Cambridge Environmental Research Consultants (http://cerc.co.uk/). c) building characteristics data, available at address (dwelling) level from Energy Performance Certificates: available 2005-2014. Note these data are openly available online from: https://epc.opendatacommunities.org/ The EPCs will be aggregated up to postcodes for linkage purposes (see below). d) Modelled tobacco expenditure data at output area level for 2010 and 2012. Note these data are not openly available, they have been provided for this project by CACI Ltd (https://www.caci.co.uk/). NOTE: the air pollution data, building characteristics data and modelled tobacco expenditure data do not relate to an individual. They are not personal data. The Census data will be linked to the existing cohort via the mothers' date of birth, forename and surname and postcode at the time of delivery (as recorded on ONS birth registration records). Only births occurring up to 12 months before and until 12 months after the Census (that is, births between March 2010 and March 2012) will be linked to 2011 Census data. The air pollution, buildings and tobacco expenditure data will be linked via the maternal and child postcodes on the Personal Demographic Service (see below). Therefore, although the HES APC or NHS birth notification data in the cohort will not be used for linkage to Census, buildings or air pollution data, the resulting linked dataset will include birth and death registration data, NHS birth notification data, and HES APC data for mothers and babies, Census 2011 data, and postcode-level air pollution and building characteristics data and output area level tobacco expenditure data.. [1 paragraph unchanged] Amendment: To allow NHS Digital to apply patient objections to the data. ONS will send the encrypted HESIDs (the patient identifier in HES) disseminated under DARS-NIC-10094-P6P4B to NHS Digital. NHS Digital would decrypt this data, run it through the patient objections process to remove those who have objected. NHS Digital would then encrypt the HES IDs before flowing the data back to ONS. The data processor as outlined in this agreement would only be permitted to access this ‘new’ data. To allow NHS Digital to apply patient objections to the data, as approved by IGARD. ONS will send the encrypted HES IDs disseminated under DARS-NIC-110094-P6P4B to NHS Digital. NHS Digital would decrypt this data, run it through the patient objections process to remove those who have objected. NHS Digital would then encrypt the HES IDs before flowing the data back to ONS. The data processor as outlined in this agreement would only be permitted to access this ‘new’ data. Amendment Request: UCL request to make the following amendment: • Link the mothers and babies in the birth cohort to longitudinal environmental exposure (buildings and air pollution data) via linkage to maternal and child postcode histories in the Personal Demographic Service (PDS), held by NHS Digital. ONS will transfer identifiers for mothers and babies in the birth cohort (NHS number, baby sex, date of birth, postcode at birth/delivery) and a study ID to NHS Digital. NHS Digital will link the mother & baby identifiers to PDS and extract longitudinal postcode histories and time stamps (date of change of postcode on PDS). NHS Digital will then transfer the postcode histories to ONS. UCL will link the longitudinal postcode histories to postcode level environmental data on air pollution, distance to road, tobacco expenditure and buildings data in the ONS SRS. These postcode-level environmental datasets will be uploaded to the ONS SRS by UCL. Following linkage, only Study IDs, longitudinal air pollution and buildings exposure histories, and time stamps will be retained for analysis in the ONS SRS. The aim of this amendment to longitudinal environmental exposure data is to improve the accuracy of measurement of exposures to potential environmental hazards during pregnancy and childhood.

Processing activities

[2 paragraphs unchanged] 2) Between the existing birth cohort and longitudinal postcode-level air pollution and buildings data. data via PDS. These are described in detail below. The existing birth cohort data is [18 words unchanged] NHS Digital, and to which this agreement relates) and the following disseminated from elsewhere i.e. not NHS Digital): by ONS: [4 paragraphs unchanged] It should be noted that, whilst NHS Digital only disseminated the HES data, NHS Digital is now the data controller for all of the above data. Note that UCL are not asking NHS Digital to resupply any HES data, nor the birth, stillbirth, death registrations data or birth notification data. The existing database held in the ONS SRS has been cleaned and validated and so UCL does not wish to have any of the HES or civil registration or birth notification data resupplied. This is why UCL are proposing ONS supply the identifiers for the existing cohort for the PDS linkage outlined below under the heading ‘2- Proposed linkage between birth cohort and postcode-level air pollution and buildings data’. [1 paragraph unchanged] NOTE (DECEMBER 2020): THIS PART OF THE LINKAGE HAS ALREADY BEEN COMPLETED [1 paragraph unchanged] Births between 27th March 2010 and 27th March 2012 (that is, births [18 words unchanged] 2011 Census questionnaires completed by the mothers in the cohort and their partners (or partners(or by the household reference person in the household in which the mothers live if this is not the mother or her partner). The proposed linkage can be summarised as follows: The proposed linkage can be summarised as follows: [10 paragraphs unchanged] 3) The ONS Data processing team will clean, hash and encrypt the [8 words unchanged] the birth registration file, using the encryption algorithm described in the document ‘ONS ’ONS policy for safeguarding data whilst managing Admin Data Research Network projects'. This [16 words unchanged] be processed separately. The original (unencrypted identifiers) will be deleted after encryption. The encryption algorithm is described in detail in the "ONS Policy for Safeguarding Data Whilst Managing ADRN Projects" (https://www.ons.gov.uk/file?uri=/aboutus/whatwedo/programmesandprojects/theadministrativedataresearchnetworkcollaboration/adrcsafeguardingpaper_tcm77-404473.pdf). In brief, the fields forename, surname, initial, sex, date of birth and postcode are cleaned, then encrypted (hashed) in each dataset before the original identifiers are deleted. The encryption is done by creating 11 matchkeys for each individual, based on combinations of the variables forename, surname, initial, date of birth, sex and postcode, to allow for potential variations in spelling discrepancies, forename/surname transpositions or moving between postcodes. The 11 matchkeys for each individual are then encrypted (hashed). The original identifiers are deleted. It is the encrypted matchkeys that are linked. This is done hierarchically so that if exact agreement is not reached between two records in the two datasets, (using the hashkey for forename, surname, full date of birth, sex and postcode), the algorithm moves on to matching on less exact matchkeys. Once the datasets are linked, the hashed identifiers are also deleted. The encryption algorithm is described in detail in the "ONS Policy for Safeguarding Data Whilst Managing ADRN Projects" (file:///C:/Users/rafa1/Downloads/adrcsafeguardingpaper_tcm77-404473.pdf). In brief, the fields forename, surname, initial, sex, date of birth and postcode are cleaned, then encrypted (hashed) in each dataset before the original identifiers are deleted. The encryption is done by creating 11 matchkeys for each individual, based on combinations of the variables forename, surname, initial, date of birth, sex and postcode, to allow for potential variations in spelling discrepancies, forename/surname transpositions or moving between postcodes. The 11 matchkeys for each individual are then encrypted (hashed). The original identifiers are deleted. It is the encrypted matchkeys that are linked. This is done hierarchically so that if exact agreement is not reached between two records in the two datasets, (using the hashkey for forename, surname, full date of birth, sex and postcode), the algorithm moves on to matching on less exact matchkeys. Once the datasets are linked, the hashed identifiers are also deleted. [3 paragraphs unchanged] 2 - 2- Proposed linkage between birth cohort and postcode-level air pollution and buildings data: data The air pollution/buildings datasets only contain postcode-level data which cannot be linked specifically to one individual. The linkage will be carried out in the following steps: NOTE: This linkage has not yet been completed (by December 2020) and UCL are seeking an amendment for this part of the linkage 1) A member of the Health Analysis and Life Events Team at ONS will extract mother’s full postcode at delivery and Study ID from birth cohort identifier file. The aim of this part of the linkage is to link the mothers and babies in the existing cohort to small-area level environmental data via the PDS to obtain longitudinal environmental exposure histories for mothers and babies in the cohort. 2) Postcode level datasets will be sent by UCL to the Health Analysis and Life Events Team at ONS. The air pollution/buildings datasets only contain address, postcode or output area-level data. These data cannot be linked specifically to one individual. The linkage will be carried out in the following steps: 3) 1) A member of the Health Analysis and Life Events Team at ONS will link the extract identifiers (NHS number, baby sex, date of birth, postcode level data to at birth/delivery) and Study ID from the birth cohort identifier file using the full postcode at delivery. file. 4) Mother’s postcode at deliver will be removed from linked file 2) ONS will securely transfer these identifiers for mothers and babies in the birth cohort to NHS Digital. 5) The air pollution and buildings data, together with the unique mother study IDs, will then be transferred to the area in the SRS where the clinical birth cohort data are held, so that the postcode level air pollution data and building characteristics data can be linked to the clinical data via the mother's study ID. 3) NHS Digital will link the mother & baby identifiers to PDS and extract longitudinal postcode histories and time stamps (date of change of address on PDS). NHS Digital will also extract some further variables, including whether a mother or baby has deregistered from PDS (and if so when). The researchers working on the birth cohort will therefore not be able to see the clinical data, the full postcode and the air pollution /buildings data simultaneously. 4) NHS Digital will return cohort Study ID, postcode histories and the other PDS variables to ONS, who will put the data in the ONS SRS. These linkage steps will be carried out separately, ie the birth cohort will be linked to air pollution and building characteristics data first, then to Census data. The intended outcome is a database which integrates the clinical and birth registration information in the existing cohort with individual-level Census 2011 data from mothers and resident partners, and the postcode level data on air pollution and building standards. 5) UCL will upload the postcode level data on environmental exposures (air pollution, building characteristics and tobacco expenditure) to ONS SRS. All the datasets will be pseudonymised for analysis, that is, all mother and baby identifiers (date of birth, date of death, NHS numbers and full postcodes) will be kept separately to the clinical and Census data at all times. The identifiers will not be available to the researchers working on analysing the birth cohort data (the separation principle). Linkage between Census and the birth cohort data will be carried out using privacy preserving methods based on hashing and encrypting the identifiers before they are moved to the ONS Data Integration server for linkage. The clinical data, Census responses and identifiers cannot be simultaneously viewed by the ONS staff linking the Census and birth cohort data. Similarly, postcode level data will be linked to the birth cohort and clinical data using a method that ensures that full postcodes and clinical data cannot be viewed concurrently. All data analyses will take place in the Secure Research Service, an ONS data safe haven. All outputs from the data analyses (tables and figures) will be checked for potential individual disclosure risk by trained staff at the ONS. 6) UCL will link the Mother-baby postcode histories to the postcode-level environmental exposure data using postcodes in the SRS. Research staff working on analysing the birth cohort with linked Census and post-code level data will require training in data security and information governance, and accreditation as Approved Researchers by the ONS. They will also require substantial contract with UCL. There will also be one UCL-registered PhD student, who is supervised by the lead UCL reseacher, working on the data. All researchers working on the data will therefore have a duty of confidentiality enshrined in their UCL contract and are required to follow the UCL data protection policy. NHS Digital draws to the attention of the University College London that NHS Digital regards the University College London as being responsible for the actions and omissions of the PhD student. 7) Following linkage, UCL will delete the actual postcode histories for mothers and babies. Only Study IDs, longitudinal air pollution, tobacco expenditure and buildings exposure histories, the additional variables from PDS (eg. date of de-registration) and time stamps will be retained and linked to the main mother baby cohort using the study ID. Once linked, UCL researchers will check, clean and validate the linked Census and postcode-level air pollution and building characteristics data to ensure linkage is reliable. UCL researchers will derive variables indicating pollution and housing quality, and variables indicating parents' country of birth, length of stay in the UK and knowledge of English/English proficiency (self-assessed from Census data) from the linked data. Other important risk factors for hospital admissions in children have already been derived in the available cohort database, that is in the linked ONS birth and registration and HES APC data, including birth weight, gestational age, mode of delivery, and maternal age. These linkage steps will be carried out separately, i.e. the birth cohort will be linked to Census data then to the air pollution, building characteristics and tobacco expenditure data. The intended outcome is a database which integrates the clinical and birth registration information in the existing cohort with individual-level Census 2011 data from mothers and resident partners, and the postcode level data on air pollution and building standards. UCL researchers will use the linked birth cohort, that is, linked ONS birth/death registration-HES APC-Census-air pollution-buildings data to calculate admission rates according to each of these risk factors as events per child-years at risk. They will use a statistical method that will measure the association between air pollution, housing and admissions for respiratory tract infections (project 1) and parental migration history and potentially preventable admissions (project 2) in children, whilst taking into account other important risk factors for hospital admissions including preterm birth, and birth weight. Linkage between Census and the birth cohort data will be carried out using privacy preserving methods based on hashing and encrypting the identifiers before they are moved to the ONS Data Integration server for linkage. The clinical data, Census responses and identifiers cannot be simultaneously viewed by the ONS staff linking the Census and birth cohort data. All data analyses will take place in the Secure Research Service, an ONS data safe haven. All outputs from the data analyses (tables and figures) will be checked for potential individual disclosure risk by trained staff at the ONS. Research staff working on analysing the birth cohort with linked Census and post-code level data will require training in data security and information governance, and accreditation as Approved Researchers by the ONS. They will also require substantial contract with UCL. There will also be one UCL-registered PhD student, who is supervised by the lead UCL researcher, working on the data. All researchers working on the data will therefore have a duty of confidentiality enshrined in their UCL contract and are required to follow the UCL data protection policy. NHS Digital draws to the attention of the University College London that NHS Digital regards the University College London as being responsible for the actions and omissions of the PhD student. Once linked, UCL researchers will check, clean and validate the linked Census and postcode-level air pollution, tobacco exposure and building characteristics data to ensure linkage is reliable. UCL researchers will derive variables indicating pollution and housing quality, and variables indicating parents' country of birth, length of stay in the UK and knowledge of English/English proficiency (self-assessed from Census data) from the linked data. Other important risk factors for hospital admissions in children have already been derived in the available cohort database, that is in the linked ONS birth and registration and HES APC data, including birth weight, gestational age, mode of delivery, and maternal age. UCL researchers will use the linked birth cohort, that is, linked ONS birth/death registration-HES APC-Census-air pollution-buildings data to calculate admission rates according to each of these risk factors as events per child-years at risk. They will use a statistical method that will measure the association between air pollution, housing and admissions for respiratory tract infections (project 1) and parental migration history, family structure and potentially preventable admissions (project 2) in children, whilst taking into account other important risk factors for hospital admissions including preterm birth, and birth weight.

Expected output

[1 paragraph unchanged] 1) an enhanced administrative birth cohort including all children born in England between 2005-2014, linked to HES APC records for themselves and their mothers, longitudinal air pollution data and building characteristics data, and Census 2011 data for mothers and their resident partners for a subset of children born between 2010 and 2012 2) estimates of the association between ambient air pollution exposure during pregnancy and the risk of hospital admission with respiratory tract infections during early childhood 2) description of the exposure of pregnant women and young children to indoor and outdoor pollutants over time and by age 3) estimates of the association between adverse housing characteristics (eg mould, indoor ambient air pollution, overcrowding) pollution exposure during pregnancy and the risk of hospital admission with respiratory tract infections during early childhood 4) estimates of the association between maternal adverse housing characteristics (eg mould, indoor air pollution, overcrowding) and paternal country the risk of birth and hospital admission to hospital for potentially preventable causes in children with respiratory tract infections during early childhood 5) estimates of the association between maternal and paternal knowledge country of English/English proficiency (self-assessed from the Census data) and length of stay in the UK birth, family structure, and admission to hospital for potentially preventable causes in children Output 1), creation of the birth cohort, is expected by the end of June 2019. UCL will describe the success of linkage between Census 2011 data and the existing birth cohort, and conduct analyses to validate the linkage, including examining agreement among key variables such as maternal ethnic group and country of birth, that are available both in the existing birth cohort and on the Census records. The resulting Census-linked birth cohort database will be described in a journal publication, such as a Data Resource Profile in the International Journal of Epidemiology. 6) estimates of the association between maternal and paternal knowledge of English/English proficiency (self-assessed from the Census data) and length of stay in the UK and admission to hospital for potentially preventable causes in children Output 2), an estimate of the strength of association between air pollution exposure during pregnancy and the risk of hospital admissions for respiratory tract infections: these results will be published in an academic journal. UCL will target high impact journals including BMJ, Lancet or Environmental Health Perspectives. UCL will also present the results at academic conferences, such as Society for Social Medicine or the Lancet Public Health Science Conference. UCL expect that these results would not be available until the end of the period of this three-year data sharing agreement period due to the complexity of analyses. Output 1), creation of the birth cohort, including linkage to environmental data, is expected by the end of March 2021. UCL will describe the success of linkage between Census 2011 data and the existing birth cohort, and conduct analyses to validate the linkage, including examining agreement among key variables such as maternal ethnic group and country of birth, that are available both in the existing birth cohort and on the Census records. The resulting Census-linked birth cohort database will be described in a journal publication, such as a Data Resource Profile in the International Journal of Epidemiology. Output 3). As for output 2). Output 2) It is expected these descriptive analyses to be available in the summer of 2021 Output 4), 3), an estimate of the strength of association between mother's and father's country of birth air pollution exposure during pregnancy and the risk of potentially preventable hospital admissions: admissions for respiratory tract infections: these results will be presented published in a scientific journal; an academic journal. UCL will target the high impact journals including BMJ, Lancet or the Lancet Child and Adolescent Health. Environmental Health Perspectives. UCL will also present scientific the results at academic conferences, such as Society for Social Medicine or the Migration, Ethnicity, Race and Health conference, and the NIHR School for Lancet Public Health Research Annual Scientific Meeting. Science Conference. UCL expect that these results to would not be available by December 2020. until the end of the period of this three-year data sharing Agreement period due to the complexity of analyses. Output 5), estimates of the association between maternal and paternal knowledge of English/English proficiency (self-assessed from the Census data) and length of stay in the UK and admission to hospital for potentially preventable causes in children (as for output 4). Output 4) an estimate of the association between adverse housing characteristics and the risk of hospital admission with respiratory tract infections during early childhood: these results will be published in an academic journal. UCL will target high impact journals such as the Lancet Child and Adolescent Health or Environmental International. UCL will also present the results at academic conferences, such as Society for Social Medicine or the Lancet Public Health Science Conference. UCL expect that these results would not be available until the end of the period of this three-year data sharing Agreement period due to the complexity of analyses. UCL researchers will communicate the results to policy makers and arms length bodies, including Public Health England, via meetings organised by the NIHR School for Public Health Research, and the NIHR Children's Policy Research Unit. Output 5), an estimate of the association between mother's and father's country of birth and the risk of potentially preventable hospital admissions: these results will be presented in a scientific journal; UCL will target the Lancet or the Lancet Child and Adolescent Health. UCL will also present scientific conferences, such as the Migration, Ethnicity, Race and Health conference, and the NIHR School for Public Health Research Annual Scientific Meeting. UCL expect these results to be available by June 2021. UCL Researchers will use several avenues to communicate results to parents. Updates from the project will be posted on a website hosted by UCL: https://www.ucl.ac.uk/child-health/research/population-policy-and-practice/child-health-informatics-group. UCL researchers will also work with the UCL press office to communicate results from the studies via press, television and radio outlets, and via social media. UCL researchers have worked with two charities who support this work: Shelter and The British Lung Foundation. The British Lung Foundation, through their partner organisation The Clean Air Parents' Network have agreed that updates about this study can be placed on their website and Facebook page. Output 6), estimates of the association between maternal and paternal knowledge of English/English proficiency (self-assessed from the Census data) and length of stay in the UK and admission to hospital for potentially preventable causes in children: these results will be presented in a scientific journal; UCL will target the Lancet Global Health or the BMJ. UCL will also present scientific conferences, such as the Migration, Ethnicity, Race and Health conference, and the NIHR School for Public Health Research Annual Scientific Meeting. UCL expect these results to be available by June 2021. UCL will also hold meetings with parents in collaboration with these two charities to update them on our results and discuss how they can best be communicated to the general public. UCL researchers will communicate the results to policy makers and arms length bodies, including Public Health England, via meetings organised by the NIHR School for Public Health Research, and the NIHR Children's Policy Research Unit. UCL Researchers will use several avenues to communicate results to parents. Updates from the project will be posted on a website hosted by UCL: https://www.ucl.ac.uk/child-health/research/population-policy-and-practice/child-healthinformatics-group. UCL researchers will also work with the UCL press office to communicate results from the studies via press, television and radio outlets, and via social media. UCL researchers have worked with two charities who support this work: Shelter and The British Lung Foundation. The British Lung Foundation, through their partner organisation The Clean Air Parents' Network have agreed that updates about this study can be placed on their website and Facebook page. UCL will also hold meetings with parents in collaboration with these two charities to update them on the results and discuss how they can best be communicated to the general public. The new Policy Research Unit for Children and Families (CPRU), funded by [111 words unchanged] how to communicate results to a broad public of parents and children.

Expected measurable benefits

The aim of this project is to identify household, parental and environmental [24 words unchanged] of, for example, overcrowded housing to emergency hospital admissions can be estimated. A key strength of this project is that UCL will be able to measure longitudinal exposures to environmental risk factors across pregnancy and early childhood. The results will allow the design and targeting of interventions, either through [20 words unchanged] This project therefore aims to improve the evidence base in order to save money for the NHS, reduce stress for families, and improve children’s children's access to health services and health outcomes both in the short and long term. term and to reduce costs for the NHS. [2 paragraphs unchanged] 2) The National Health Service (Clinical Commissioning Groups; CCGs): Part 2 of the project will examine whether all children have equal access to healthcare, independent of their parents’ country parentscountry of birth. If inequalities are found, this will provide evidence for CCGs [8 words unchanged] services (this could include extending interpreting services or setting up outreach clinics). [1 paragraph unchanged]

Benefits reported

Not stated in the previous version; added here.

No benefits have yet been yielded. DARS-NIC-234656-C3J1D was originally approved in April 2019; however UCL were not able to access the data until an amendment was approved in November 2019. This amendment involved removing opt outs from the cohort. UCL gained access to the data in the ONS SRS in February 2020; however, when COVID lockdown happened UCL lost access to the data for another 2 months, before ONS agreed that UCL could access the SRS from home.

UCL were able to access the data 10 months after original approval, since accessing the data, UCL have cleaned and validated the birth cohort. The first outputs are expected to be ready by summer/autumn 2021.

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

Objective for processing

[25 paragraphs unchanged] The purpose of this application falls under Article 6 (1) (e) of the GDPR and the lawful basis for using information collected routinely for administrative purposes for research is the ‘public task’. This is part of the University’s commitment to ‘integrate research and innovation for the long-term benefit of humanity’. The application also falls under Article 9 (2) (j), as scientific research. Amendment: To allow NHS Digital to apply patient objections to the data, as approved by IGARD. ONS will send the encrypted HES IDs disseminated under DARS-NIC-110094-P6P4B to NHS Digital. NHS Digital would decrypt this data, run it through the patient objections process to remove those who have objected. NHS Digital would then encrypt the HES IDs before flowing the data back to ONS. The data processor as outlined in this agreement would only be permitted to access this ‘new’ data.

Benefits reported

Stated in the previous version and removed here.

Yielded Benefits is not a requirement for new applications.

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