Children's Surgery Outcome Reporting (CSOR) Research Database - Clinical Data

The University of Oxford and Oxford University Hospitals NHS Foundation Trust requires access to record-level identifiable health Data from NHS England for the purpose of the following research programme: The Children’s Surgery Outcome Reporting (CSOR) programme.

The following is a summary of the aims of the CSOR research programme provided by the study team:

At present, significant variation exists in the way children with surgical conditions are managed. Some of this variation is expected and unimportant, but some is unwarranted and associated with variation in outcome. Due to multiple limitations in the paediatric surgical data that are available for analysis (both in terms of research data, and real-time centre specific outcomes data), it is not possible to differentiate the two. There are therefore children being treated for surgical conditions whose outcomes are worse than they would be if better data were available for analysis.

The overall purpose of the Children’s Surgery Outcome Reporting (CSOR) programme is to investigate whether it is possible to collect paediatric surgical outcomes data using a system that links routinely collected data and parent reported outcome data and provides a platform for centre specific feedback of outcomes in order to reduce unwarranted outcome variation.

The CSOR research database will be established to collect and link the data that are required to identify unwarranted variation between hospitals in management and outcomes of children with surgical conditions. Three sources of data will be linked within the CSOR research database.

1) Data collected in hospitals’ electronic patient record systems

2) National routine sources of data available from NHS England (this Data Sharing Agreement)

3) Data collected annually from the child or their parent/guardian about the child’s quality of life

The collected data will primarily be used to determine whether hospitals observed outcomes differ to the outcomes expected based upon the case-mix of the children they have treated. This analysis will inform a facilitated feedback process through which participating sites are helped to understand why the outcomes they have achieved for children are better or worse than would be expected. Results of analysis will also be shared via a public facing web-based dashboard.

On a quarterly basis the datasets listed below are required for all infants/young people in the cohort in order to predict how successfully they would be expected to be treated, and determine how successfully they are observed to have been treated. The level of data will be Identifiable to facilitate linkage with other data sources and maintain ongoing contact with parents for the purposes of colleting quality of life data. The combination of NHS Number and Date of Birth provides the minimum level of identifiable information required for robust linkage. The Data are required to enable case-mix adjustment and determination of outcomes.

The following NHS England Data will be accessed:

• Hospital Episode Statistics

o Admitted Patient Care

o Critical Care

• Person Demographics Service (PDS)

• Civil Registration of Deaths

The rationale for specific data categories is described below:

1) The infant’s demographics and identifiers – Required to ensure accurate linkage with other collected study data.

2) Maternal and infant health characteristics – required to take account of co-morbidities that may impact how successful an infant’s treatment is.

3) Infant health outcomes – required to calculate an infant’s treatment success score, the primary measure that will be used to identify unwarranted variation in practice.

The Data will be minimised as follows:

• Limited to only data for infants whose eligibility according to diagnosis (oesophageal atresia, posterior urethral valves, congenital diaphragmatic hernia, necrotising enterocolitis, congenital diaphragmatic hernia and gastroschisis) and treatment in a participating site after launch of the CSOR research database has been confirmed.

• The CSOR pilot programme will involve 10 participating hospitals from across England and Scotland.

• The geographical area is restricted to patients treated in the participating hospitals.

All analyses will be carried out on pseudonymised datasets.

For every child in the CSOR research database, the collected data will be utilised to calculate an observed CSOR treatment success score and to predict an expected CSOR treatment success score based upon the child’s underlying characteristics. The CSOR Treatment Success Score is a composite outcome that is calculated based upon the numbers and types of operations a child has undergone, the numbers of times they have been admitted for an infection related to their surgical condition, their quality of life, and the duration for which they survive. The difference between a child’s observed and expected CSOR treatment success score will be calculated. Each participating hospital’s mean difference between observed and expected CSOR treatment success scores will be calculated. If this value is greater than 0 it suggests the hospital’s outcomes are better than expected, and if less than 0 it suggests their outcomes are worse than would be expected. The analyses will be used to inform a facilitated feedback process to help participating hospitals understand why their outcomes may be better or worse than expected.

Secondarily, the data contained within the CSOR research database will be used to conduct studies within the scope of improving the health and wellbeing of children with surgical conditions.

The CSOR Steering Committee provide overall supervision, management, strategic direction and governance of the activities of the CSOR Research, including monitoring and supervising the progress of data collection and analysis. The steering committee will review all proposals to utilise data within the CSOR research database.

The University of Oxford is the sponsor for this study. The University of Oxford and Oxford University Hospitals NHS Foundation Trust are joint Controllers responsible for ensuring that the data will only be processed for the purpose described above. Both organisations will also process the data in this agreement.

The following organisations have collaborated on devising the programme and will provide electronic patient record data to the CSOR research database:

• Alder Hey Children’s Hospital

• Birmingham Children’s Hospital

• Cambridge University Hospitals NHS Foundation Trust

• Chelsea & Westminster Hospital

• Evelina Children’s Hospital

• Great Ormond Street Hospital

• Oxford University Hospitals NHS Foundation Trust

• Royal Hospital for Children

• Royal Manchester Children’s Hospital

• Southampton General Hospital

The CSOR study is funded by the National Institute for Health Research. The current funding period is 01/03/2020 to 28/02/2025. The NIHR are involved in monitoring the progress of the study and reviewing study protocols, but do not decide the purpose and means of processing the data in the CSOR study, and are therefore not considered to be a Data Controller in this agreement. NIHR have no ability to control or suppress the outcomes published under this programme.

UK GDPR LEGAL BASIS FOR THE PROCESSING OF DATA

The University of Oxford and Oxford University Hospitals NHS Foundations Trust are joint controllers as the organisations responsible for ensuring that the Data will only be processed for the purpose described above.

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;

The lawful basis for processing special category data under the UK GDPR is:

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.

PATIENT AND PUBLIC INVOLVEMENT AND ENGAGEMENT

The need for the CSOR programme was first identified and developed through work with the Parental Advisory Group (PAG) set up by the study team at the University of Oxford. The PAG, consisting of over 100 parents/families of children with surgical conditions, charities and support group representatives from across the UK, remain actively involved throughout the course of the programme. As a minimum, annual meetings have been held to update, discuss and gather feedback on the aims and methods of establishing the research database. Feedback from the PAG has been key to the development of the proposed parent consent and data collection process. There will be two parent or patient representatives on the CSOR Research Database Steering Committee to ensure that the parent/patient voice is maintained in the functioning of the CSOR Research Database. The CSOR Research Database will continue to be reviewed at the annual PAG meeting. The members of the CSOR Research Database Steering Committee will have no access to the NHS England Data described in this Data Sharing Agreement.

The dissemination for this Data Sharing Agreement will require sharing of identifiable data items about eligible children and their parents without the consent of either party. There is the potential that some parents may object to this, and subsequently finding out that their/their child's data have been used for research purposes without consent may cause them distress. Instead, the study has support under section 251 of the NHS Act 2006 to enable the common law duty of confidentiality to be temporarily lifted so that confidential patient information can be processed without consent. This is an area that has been explored in depth with a parent advisory group consisting of a large number of parents of children with surgical conditions, as well as representatives of charities and support groups. Feedback from this group has been overwhelmingly positive. Most parents assume that data are already shared for the purposes of research, and as long as data are appropriately protected, parents did not have any objections to these data being shared. The potential benefit from sharing of the data in terms of developing a far greater understanding of the models of care and treatments that are best for children with surgical conditions, as well as the potential to reduce unwarranted variation in management and outcomes for these children therefore far outweighs the potential harms of the data collection.

Previous attempts to set up condition-specific databases for key paediatric surgical conditions have been limited by significant issues with case ascertainment, as they relied on voluntary case reporting by surgical staff. These databases have shown that reliance on site staff to supply contact details results in approximately 30% of the eligible population not being approached for consent, as site staff have been unable to supply their contact details. Where contact details are received though, and parents are approached for consent without prior knowledge that their contact details have been shared, this results in a relatively high consent rate (~60%). In a recent (2017-2019) paediatric surgical study in which approximately 190 parents were approached by a research team without prior consent, no complaints were received about the fact that their contact details had been shared outside of their usual clinical team without prior consent.

Currently, children with the conditions included in this Programme have poor outcomes and significant long-term health needs. These include 25-30% one-year mortality rates for infants with necrotising enterocolitis and congenital diaphragmatic hernia, 60% rates of primary school aged faecal incontinence for children with Hirschsprung’s disease, and the need for multiple operations in the first 28 days of life, for infants with gastroschisis. Management of these children in the UK currently varies widely. No mechanisms exist to determine which elements of this variation are acceptable, and which are unwarranted and associated with variation in outcome. The Data disseminated under this DSA hopes to provide the information that is needed to calculate the number and types of operations that children have undergone, as well as the number of times they have been admitted and treated for an infection related to their underlying surgical condition.

These outcomes contribute towards calculating a child’s CSOR treatment success score. When used in conjunction with the additional data held in the CSOR research database, knowledge of a child’s CSOR treatment success score will create for the first time the ability to understand which elements of management are inconsequential, and which are unwarranted. The ability to detect unwarranted variation in management hopes to help standardise delivery of care across different surgeons and hospitals, with the expectation that the overall delivery of care for children with these conditions will improve. This hopes to, in the long run, improve the health and wellbeing of children born with these conditions in the future, as well as allowing evidence-based restructuring of the way care is delivered for these children.

The following outputs hope to be developed:

1) A Research Database (the CSOR Research Database) which can be utilised for identifying unwarranted variation in the management and outcomes of children with surgical conditions, and for the conduct of approved research to improve outcomes for children with surgical conditions. Data collection is anticipated to begin in quarter 3 of 2023.

2) Peer reviewed publications and presentations relating to development of the CSOR Research Database and CSOR Research Database methodology. Publication is anticipated in quarter 3 of 2024.

3) A ‘case mix adjustment model’ used to predict how successful an infant’s treatment is expected to be based upon their underlying characteristics. The initial model will be developed using data collected in the first year of the CSOR Research Database, and will be revised initially on a yearly, and then likely five-yearly, basis as additional data are added to the database.

4) Peer reviewed publications and presentations describing the development of the case-mix adjustment model. Publication of the initial model is anticipated in quarter 2 of 2025.

5) Peer reviewed publications and presentations relating to the results of analyses conducted on the pseudonymised dataset. These will include, but not be limited to, analyses describing management strategies, infant outcomes, and factors affecting outcomes. The date of these publications will be influenced by case incidence and ascertainment but are not anticipated until quarter 4 of 2025 at the earliest.

6) Live dashboards accessed by hospitals and the public describing the activity in participating hospitals and comparisons of the hospital’s observed and expected outcomes. Development is anticipated to be complete by quarter 4 of 2023.

7) Facilitated feedback and peer review processes, the aims of which are to help hospitals understand their own performance, will be informed by analysis of data contained in the final linked, pseudonymised dataset. Development of these processes is anticipated throughout 2023, with implementation in quarter 3 or 4 of 2024.

8) Professional and public facing annual reports describing the activities of the CSOR Research Database, and including, amongst other things, information relating to case ascertainment, national management strategies, and national benchmarks for infant outcomes. Anticipated yearly from quarter 3 2024 onwards.

9) Podcasts, posters, information leaflets and videos explaining to parents, patients and the public, the results of analyses conducted using data contained in the CSOR Research Database. The date of these outputs will be influenced by case incidence and ascertainment but are not anticipated until quarter 4 of 2025 at the earliest.

10) Peer reviewed publications and presentations describing the periodic process evaluation conducted to determine whether implementation of the CSOR Research Database and facilitated feedback programme has improved the health and wellbeing of children with surgical conditions. The results of the initial process evaluation are anticipated to be published in quarter 3 of 2025.

All outputs will contain only aggregated data with small numbers suppressed as per the HES Analysis guide. As per funder requirements (NIHR) all publications will be open access. All data and knowledge are owned by the University of Oxford.

A wide range of routes will be used to disseminate the results of the programme to the beneficiaries, including children with one of the included conditions, adults who were previously treated for one of the included conditions, parents of children with one of the included conditions, professionals and policy-makers.

Dissemination strategies to parents and the public will be guided by one of the CSOR co-investigators who is an expert by experience (a parent of a child with a surgical condition), as well as by an established parental advisory group consisting of parents of children with surgical conditions, and representatives of relevant support groups and charities. Leaflets, videos, infographics and social media will be used as appropriate to aid dissemination.

The programme has direct links with NHS England and the Department of Health and Social Care and will disseminate directly to relevant policy-makers. that the CSOR programme team aim to ensure the outputs are appropriately used to commission services in England, and that full dissemination of outcomes to professional colleagues and integration with professional standards and guidelines. The study team have agreement to specifically hold a dissemination event at the British Association of Paediatric Surgeons Annual Conference to ensure the results reach the paediatric surgical audience directly. The study team also hope to disseminate through journal publications and presentations at other key conferences. Dissemination to the wider public aims to be via regular newsletters available on the CSOR website, website bulletins and social media communications via the CSOR Twitter account.

The key route of dissemination to the involved professionals/stakeholders will be through the use of the live dashboards, and facilitated feedback and peer review processes. Through these, professionals involved in the care of infants whose data are included in the CSOR Research Database will be guided through the interpretation of the data for their own hospital in order to help identify any ways in which practice could be changed to improve the care that is being delivered. Data collected by the CSOR system will be used to develop benchmarked, auditable standards for delivery of paediatric surgical services, against which individual units can compare themselves, and be compared.

Agreement has been forthcoming from the National Consultant Information Programme for this database, once successfully developed to reside within NHS England in order to facilitate individual surgeon feedback via the The National Council of Integrative Psychotherapists (NCIP) portal (http://gettingitrightfirsttime.co.uk/ncip/), and the study team have also had an indication that in Scotland it could be hosted within the Information Services Division. This important surgeon-level ‘building block’ will allow the data to be built from surgeon to unit to national level. Use of the data to identify gold-standard practice hopes to allow clinicians, managers and commissioners to review and modify the way their services are organised and delivered.

The CSOR team anticipate process evaluation will begin between 2025 Q4 and 2026 Q1, therefore there are no yielded benefits to date.

The University of Oxford and Oxford University Hospitals NHS Foundation Trust requires access to record-level identifiable health Data from NHS England for the purpose of the following research programme: The Children’s Surgery Outcome Reporting (CSOR) programme.

The following is a summary of the aims of the CSOR research programme provided by the study team:

At present, significant variation exists in the way children with surgical conditions are managed. Some of this variation is expected and unimportant, but some is unwarranted and associated with variation in outcome. Due to multiple limitations in the paediatric surgical data that are available for analysis (both in terms of research data, and real-time centre specific outcomes data), it is not possible to differentiate the two. There are therefore children being treated for surgical conditions whose outcomes are worse than they would be if better data were available for analysis.

The overall purpose of the Children’s Surgery Outcome Reporting (CSOR) programme is to investigate whether it is possible to collect paediatric surgical outcomes data using a system that links routinely collected data and parent reported outcome data and provides a platform for centre specific feedback of outcomes in order to reduce unwarranted outcome variation.

The CSOR research database will be established to collect and link the data that are required to identify unwarranted variation between hospitals in management and outcomes of children with surgical conditions. Three sources of data will be linked within the CSOR research database.

1) Data collected in hospitals’ electronic patient record systems

2) National routine sources of data available from NHS England (this Data Sharing Agreement)

3) Data collected annually from the child or their parent/guardian about the child’s quality of life

The collected data will primarily be used to determine whether hospitals observed outcomes differ to the outcomes expected based upon the case-mix of the children they have treated. This analysis will inform a facilitated feedback process through which participating sites are helped to understand why the outcomes they have achieved for children are better or worse than would be expected. Results of analysis will also be shared via a public facing web-based dashboard.

On a quarterly basis the datasets listed below are required for all infants/young people in the cohort in order to predict how successfully they would be expected to be treated, and determine how successfully they are observed to have been treated. The level of data will be Identifiable to facilitate linkage with other data sources and maintain ongoing contact with parents for the purposes of colleting quality of life data. The combination of NHS Number and Date of Birth provides the minimum level of identifiable information required for robust linkage. The data requested are required to enable case-mix adjustment and determination of outcomes.

For this DSA, the following NHS England Data will be accessed:

• Hospital Episode Statistics (HES) Admitted Patient Care (APC)

• HES Critical Care (CC)

• Person Demographics Service (PDS)

• Civil Registration of Deaths

The rationale for specific data categories is described below:

1) The infant’s demographics and identifiers – Required to ensure accurate linkage with other collected study data.

2) Maternal and infant health characteristics – required to take account of co-morbidities that may impact how successful an infant’s treatment is.

3) Infant health outcomes – required to calculate an infant’s treatment success score, the primary measure that will be used to identify unwarranted variation in practice.

Data will only be requested for infants whose eligibility according to diagnosis (oesophageal atresia, posterior urethral valves, congenital diaphragmatic hernia, necrotising enterocolitis, congenital diaphragmatic hernia and gastroschisis) and treatment in a participating site after launch of the CSOR research database has been confirmed. The CSOR pilot programme will involve 10 participating hospitals from across England and Scotland. The requested geographical area has been restricted to patients treated in the participating hospitals. The occurrence of the six conditions is not limited by specific demographics, and therefore the dataset cannot be limited by demography.

The data requested are the minimum required in order to enable linkage of the three sources of information and achieve the objectives described above. A standing item is retained on the agenda of the CSOR steering committee to review the programme data collected and requested to ensure that it remains the minimum possible to achieve the aims of the CSOR programme.

The data subjects will be all children treated in any of the participating sites during the Data Sharing Agreement period who have a diagnosis of one of six conditions: necrotising enterocolitis (NEC), Hirschsprung’s disease (HD), gastroschisis, posterior urethral valves (PUV), congenital diaphragmatic hernia (CDH) and oesophageal atresia (OA).

There are no control subjects.

All analyses will be carried out on pseudonymised datasets.

For every child in the CSOR research database, the collected data will be utilised to calculate an observed CSOR treatment success score and to predict an expected CSOR treatment success score based upon the child’s underlying characteristics. The CSOR Treatment Success Score is a composite outcome that is calculated based upon the numbers and types of operations a child has undergone, the numbers of times they have been admitted for an infection related to their surgical condition, their quality of life, and the duration for which they survive. The difference between a child’s observed and expected CSOR treatment success score will be calculated. Each participating hospital’s mean difference between observed and expected CSOR treatment success scores will be calculated. If this value is greater than 0 it suggests the hospital’s outcomes are better than expected, and if less than 0 it suggests their outcomes are worse than would be expected. The analyses will be used to inform a facilitated feedback process to help participating hospitals understand why their outcomes may be better or worse than expected.

Secondarily, the data contained within the CSOR research database will be used to conduct studies within the scope of improving the health and wellbeing of children with surgical conditions.

The CSOR Steering Committee provide overall supervision, management, strategic direction and governance of the activities of the CSOR Research, including monitoring and supervising the progress of data collection and analysis. The steering committee will review all proposals to utilise data within the CSOR research database.

The University of Oxford is the sponsor for this study. The University of Oxford and Oxford University Hospitals NHS Foundation Trust are joint Controllers responsible for ensuring that the data will only be processed for the purpose described above. Both organisations will also process the data in this agreement.

The following organisations have collaborated on devising the programme and will provide electronic patient record data to the CSOR research database:

• Alder Hey Children’s Hospital

• Birmingham Children’s Hospital

• Cambridge University Hospitals NHS Foundation Trust

• Chelsea & Westminster Hospital

• Evelina Children’s Hospital

• Great Ormond Street Hospital

• Oxford University Hospitals NHS Foundation Trust

• Royal Hospital for Children

• Royal Manchester Children’s Hospital

• Southampton General Hospital

These organisations in the bulleted list above are not involved in any way in determining the means and purpose of processing personal Data received from NHS England and will have no access to record level Data from NHS England.

The CSOR study is funded by the National Institute for Health Research. The current funding period is 01/03/2020 to 28/02/2025. The NIHR are involved in monitoring the progress of the study and reviewing study protocols, but do not decide the purpose and means of processing the data in the CSOR study, and are therefore not considered to be a Data Controller in this agreement. NIHR have no ability to control or suppress the outcomes published under this programme.

UK GDPR LEGAL BASIS FOR THE PROCESSING OF DATA

The University of Oxford, as a joint Controller who is also processing the Data will process Personal Data under UK GDPR 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. As a higher education establishment, the University of Oxford conducts research to improve health care and services, and the data requested is necessary for the performance of a task carried out in the public interest.

Oxford University Hospitals NHS Foundations Trust, as a joint Controller who is also processing the Data will process Personal Data under UK GDPR 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. Oxford University Hospital NHS Foundation Trust is a public authority. The Data Protection Act 2018 s7(1)(a) defines ‘public bodies’ for the purpose of the UK GDPR as “a public authority as defined by the Freedom of Information Act 2000”. The FOI Act 2000 Part 1, section 3 (1)(a)(i) specifies that a public authority means any body which is listed in Schedule 1. Schedule 1 Part 3 (40A) of the FOI Act 2000 stipulates “An NHS foundation trust” is a public authority.

The NHS Act 2006 section 43(5), which describes the functions of authorised NHS Foundation Trusts, states that ‘The authorisation must authorise and may require the NHS foundation trust— (a) to carry out research in connection with the provision of health care, (b) to make facilities and staff available for the purposes of education, training or research carried on by others'.

Additionally, under GDPR Article 9(2)(j) processing of Special Category Personal Data (of which Health data is one) is necessary for archiving for research purposes. Data minimisation processes are being followed and only data that is specifically required for the purposes of this study have been requested, to protect the rights of the data subjects. The Controllers have satisfied themselves that this request is appropriate, necessary and proportionate for the performance of the task described in the Purpose statement and that there is no other reasonable and less intrusive means to achieve their purpose.

PATIENT AND PUBLIC INVOLVEMENT AND ENGAGEMENT

The need for the CSOR programme was first identified and developed through work with the Parental Advisory Group (PAG) set up by the study team at the University of Oxford. The PAG, consisting of over 100 parents/families of children with surgical conditions, charities and support group representatives from across the UK, remain actively involved throughout the course of the programme. As a minimum, annual meetings have been held to update, discuss and gather feedback on the aims and methods of establishing the research database. Feedback from the PAG has been key to the development of the proposed parent consent and data collection process. There will be two parent or patient representatives on the CSOR Research Database Steering Committee to ensure that the parent/patient voice is maintained in the functioning of the CSOR Research Database. The CSOR Research Database will continue to be reviewed at the annual PAG meeting. The members of the CSOR Research Database Steering Committee will have no access to the NHS England Data described in this Data Sharing Agreement.

The dissemination for this Data Shairng Agreement will require sharing of identifiable data items about eligible children and their parents without the consent of either party. There is the potential that some parents may object to this, and subsequently finding out that their/their child's data have been used for research purposes without consent may cause them distress. Instead, the study has support under section 251 of the NHS Act 2006 to enable the common law duty of confidentiality to be temporarily lifted so that confidential patient information can be processed without consent. This is an area that has been explored in depth with a parent advisory group consisting of a large number of parents of children with surgical conditions, as well as representatives of charities and support groups. Feedback from this group has been overwhelmingly positive. Most parents assume that data are already shared for the purposes of research, and as long as data are appropriately protected, parents did not have any objections to these data being shared. The potential benefit from sharing of the data in terms of developing a far greater understanding of the models of care and treatments that are best for children with surgical conditions, as well as the potential to reduce unwarranted variation in management and outcomes for these children therefore far outweighs the potential harms of the data collection.

Previous attempts to set up condition-specific databases for key paediatric surgical conditions have been limited by significant issues with case ascertainment, as they relied on voluntary case reporting by surgical staff. These databases have shown that reliance on site staff to supply contact details results in approximately 30% of the eligible population not being approached for consent, as site staff have been unable to supply their contact details. Where contact details are received though, and parents are approached for consent without prior knowledge that their contact details have been shared, this results in a relatively high consent rate (~60%). In a recent (2017-2019) paediatric surgical study in which approximately 190 parents were approached by a research team without prior consent, no complaints were received about the fact that their contact details had been shared outside of their usual clinical team without prior consent.

Currently, children with the conditions included in this Programme have poor outcomes and significant long-term health needs. These include 25-30% one-year mortality rates for infants with necrotising enterocolitis and congenital diaphragmatic hernia, 60% rates of primary school aged faecal incontinence for children with Hirschsprung’s disease, and the need for multiple operations in the first 28 days of life, for infants with gastroschisis. Management of these children in the UK currently varies widely. No mechanisms exist to determine which elements of this variation are acceptable, and which are unwarranted and associated with variation in outcome. The Data disseminated under this DSA hopes to provide the information that is needed to calculate the number and types of operations that children have undergone, as well as the number of times they have been admitted and treated for an infection related to their underlying surgical condition.

These outcomes contribute towards calculating a child’s CSOR treatment success score. When used in conjunction with the additional data held in the CSOR research database, knowledge of a child’s CSOR treatment success score will create for the first time the ability to understand which elements of management are inconsequential, and which are unwarranted. The ability to detect unwarranted variation in management hopes to help standardise delivery of care across different surgeons and hospitals, with the expectation that the overall delivery of care for children with these conditions will improve. This hopes to, in the long run, improve the health and wellbeing of children born with these conditions in the future, as well as allowing evidence-based restructuring of the way care is delivered for these children.

The following outputs hope to be developed:

1) A Research Database (the CSOR Research Database) which can be utilised for identifying unwarranted variation in the management and outcomes of children with surgical conditions, and for the conduct of approved research to improve outcomes for children with surgical conditions. Data collection is anticipated to begin in quarter 3 of 2023.

2) Peer reviewed publications and presentations relating to development of the CSOR Research Database and CSOR Research Database methodology. Publication is anticipated in quarter 3 of 2024.

3) A ‘case mix adjustment model’ used to predict how successful an infant’s treatment is expected to be based upon their underlying characteristics. The initial model will be developed using data collected in the first year of the CSOR Research Database, and will be revised initially on a yearly, and then likely five-yearly, basis as additional data are added to the database.

4) Peer reviewed publications and presentations describing the development of the case-mix adjustment model. Publication of the initial model is anticipated in quarter 2 of 2025.

5) Peer reviewed publications and presentations relating to the results of analyses conducted on the pseudonymised dataset. These will include, but not be limited to, analyses describing management strategies, infant outcomes, and factors affecting outcomes. The date of these publications will be influenced by case incidence and ascertainment but are not anticipated until quarter 4 of 2025 at the earliest.

6) Live dashboards accessed by hospitals and the public describing the activity in participating hospitals and comparisons of the hospital’s observed and expected outcomes. Development is anticipated to be complete by quarter 4 of 2023.

7) Facilitated feedback and peer review processes, the aims of which are to help hospitals understand their own performance, will be informed by analysis of data contained in the final linked, pseudonymised dataset. Development of these processes is anticipated throughout 2023, with implementation in quarter 3 or 4 of 2024.

8) Professional and public facing annual reports describing the activities of the CSOR Research Database, and including, amongst other things, information relating to case ascertainment, national management strategies, and national benchmarks for infant outcomes. Anticipated yearly from quarter 3 2024 onwards.

9) Podcasts, posters, information leaflets and videos explaining to parents, patients and the public, the results of analyses conducted using data contained in the CSOR Research Database. The date of these outputs will be influenced by case incidence and ascertainment but are not anticipated until quarter 4 of 2025 at the earliest.

10) Peer reviewed publications and presentations describing the periodic process evaluation conducted to determine whether implementation of the CSOR Research Database and facilitated feedback programme has improved the health and wellbeing of children with surgical conditions. The results of the initial process evaluation are anticipated to be published in quarter 3 of 2025.

All outputs will contain only aggregated data with small numbers suppressed as per the HES Analysis guide. As per funder requirements (NIHR) all publications will be open access. All data and knowledge are owned by the University of Oxford.

A wide range of routes will be used to disseminate the results of the programme to the beneficiaries, including children with one of the included conditions, adults who were previously treated for one of the included conditions, parents of children with one of the included conditions, professionals and policy-makers.

Dissemination strategies to parents and the public will be guided by one of the CSOR co-investigators who is an expert by experience (a parent of a child with a surgical condition), as well as by an established parental advisory group consisting of parents of children with surgical conditions, and representatives of relevant support groups and charities. Leaflets, videos, infographics and social media will be used as appropriate to aid dissemination.

The programme has direct links with NHS England and the Department of Health and Social Care and will disseminate directly to relevant policy-makers. that the CSOR programme team aim to ensure the outputs are appropriately used to commission services in England, and that full dissemination of outcomes to professional colleagues and integration with professional standards and guidelines. The study team have agreement to specifically hold a dissemination event at the British Association of Paediatric Surgeons Annual Conference to ensure the results reach the paediatric surgical audience directly. The study team also hope to disseminate through journal publications and presentations at other key conferences. Dissemination to the wider public aims to be via regular newsletters available on our website, website bulletins and social media communications via the CSOR Twitter account.

The key route of dissemination to the involved professionals/stakeholders will be through the use of the live dashboards, and facilitated feedback and peer review processes. Through these, professionals involved in the care of infants whose data are included in the CSOR Research Database will be guided through the interpretation of the data for their own hospital in order to help identify any ways in which practice could be changed to improve the care that is being delivered. Data collected by the CSOR system will be used to develop benchmarked, auditable standards for delivery of paediatric surgical services, against which individual units can compare themselves, and be compared.

Agreement has been forthcoming from the National Consultant Information Programme for this database, once successfully developed to reside within NHS England in order to facilitate individual surgeon feedback via the The National Council of Integrative Psychotherapists (NCIP) portal (http://gettingitrightfirsttime.co.uk/ncip/), and the study team have also had an indication that in Scotland it could be hosted within the Information Services Division. This important surgeon-level ‘building block’ will allow the data to be built from surgeon to unit to national level. Use of the data to identify gold-standard practice hopes to allow clinicians, managers and commissioners to review and modify the way their services are organised and delivered.

The University of Oxford and Oxford University Hospitals NHS Foundation Trust requires access to record-level identifiable health Data from NHS England for the purpose of the following research programme: The Children’s Surgery Outcome Reporting (CSOR) programme.

The CSOR programme is a five-year, NIHR-funded pilot programme which is working to improve the health and wellbeing of children with surgical conditions. There are three linked Data Sharing Agreements being made to NHS England for the CSOR Programme. This Data Sharing Agreement, DARS-NIC-674822-S2K9T and DARS-NIC-717299-R5H5N .

The following is a summary of the aims of the CSOR research programme provided by the study team:

At present, significant variation exists in the way children with surgical conditions are managed. Some of this variation is expected and unimportant, but some is unwarranted and associated with variation in outcome. Due to multiple limitations in the paediatric surgical data that are available for analysis (both in terms of research data, and real-time centre specific outcomes data), it is not possible to differentiate the two. There are therefore children being treated for surgical conditions whose outcomes are worse than they would be if better data were available for analysis.

The overall purpose of the Children’s Surgery Outcome Reporting (CSOR) programme is to investigate whether it is possible to collect paediatric surgical outcomes data using a system that links routinely collected data and parent reported outcome data and provides a platform for centre specific feedback of outcomes in order to reduce unwarranted outcome variation.

The CSOR research database will be established to collect and link the data that are required to identify unwarranted variation between hospitals in management and outcomes of children with surgical conditions. Three sources of data will be linked within the CSOR research database.

1) Data collected in hospitals’ electronic patient record systems

2) National routine sources of data available from NHS England (this Data Sharing Agreement)

3) Data collected annually from the child or their parent/guardian about the child’s quality of life

The collected data will primarily be used to determine whether hospitals observed outcomes differ to the outcomes expected based upon the case-mix of the children they have treated. This analysis will inform a facilitated feedback process through which participating sites are helped to understand why the outcomes they have achieved for children are better or worse than would be expected. Results of analysis will also be shared via a public facing web-based dashboard.

On a quarterly basis the datasets listed below are required for all infants/young people in the cohort in order to predict how successfully they would be expected to be treated, and determine how successfully they are observed to have been treated. The level of data will be Identifiable to facilitate linkage with other data sources and maintain ongoing contact with parents for the purposes of colleting quality of life data. The combination of NHS Number and Date of Birth provides the minimum level of identifiable information required for robust linkage. The data requested are required to enable case-mix adjustment and determination of outcomes.

For this DSA, the following NHS England Data will be accessed:

• Hospital Episode Statistics (HES) Admitted Patient Care (APC)

• HES Critical Care (CC)

• Person Demographics Service (PDS)

• Civil Registration of Deaths

The rationale for specific data categories is described below:

1) The infant’s demographics and identifiers – Required to ensure accurate linkage with other collected study data.

2) Maternal and infant health characteristics – required to take account of co-morbidities that may impact how successful an infant’s treatment is.

3) Infant health outcomes – required to calculate an infant’s treatment success score, the primary measure that will be used to identify unwarranted variation in practice.

Data will only be requested for infants whose eligibility according to diagnosis (oesophageal atresia, posterior urethral valves, congenital diaphragmatic hernia, necrotising enterocolitis, congenital diaphragmatic hernia and gastroschisis) and treatment in a participating site after launch of the CSOR research database has been confirmed. The CSOR pilot programme will involve 10 participating hospitals from across England and Scotland. The requested geographical area has been restricted to patients treated in the participating hospitals. The occurrence of the six conditions is not limited by specific demographics, and therefore the dataset cannot be limited by demography.

The data requested are the minimum required in order to enable linkage of the three sources of information and achieve the objectives described above. A standing item is retained on the agenda of the CSOR steering committee to review the programme data collected and requested to ensure that it remains the minimum possible to achieve the aims of the CSOR programme.

The data subjects will be all children treated in any of the participating sites during the Data Sharing Agreement period who have a diagnosis of one of six conditions: necrotising enterocolitis (NEC), Hirschsprung’s disease (HD), gastroschisis, posterior urethral valves (PUV), congenital diaphragmatic hernia (CDH) and oesophageal atresia (OA).

There are no control subjects.

All analyses will be carried out on pseudonymised datasets.

For every child in the CSOR research database, the collected data will be utilised to calculate an observed CSOR treatment success score and to predict an expected CSOR treatment success score based upon the child’s underlying characteristics. The CSOR Treatment Success Score is a composite outcome that is calculated based upon the numbers and types of operations a child has undergone, the numbers of times they have been admitted for an infection related to their surgical condition, their quality of life, and the duration for which they survive. The difference between a child’s observed and expected CSOR treatment success score will be calculated. Each participating hospital’s mean difference between observed and expected CSOR treatment success scores will be calculated. If this value is greater than 0 it suggests the hospital’s outcomes are better than expected, and if less than 0 it suggests their outcomes are worse than would be expected. The analyses will be used to inform a facilitated feedback process to help participating hospitals understand why their outcomes may be better or worse than expected.

Secondarily, the data contained within the CSOR research database will be used to conduct studies within the scope of improving the health and wellbeing of children with surgical conditions.

The CSOR Steering Committee provide overall supervision, management, strategic direction and governance of the activities of the CSOR Research, including monitoring and supervising the progress of data collection and analysis. The steering committee will review all proposals to utilise data within the CSOR research database.

The University of Oxford is the sponsor for this study. The University of Oxford and Oxford University Hospitals NHS Foundation Trust are joint Controllers responsible for ensuring that the data will only be processed for the purpose described above. Both organisations will also process the data in this agreement.

The following organisations have collaborated on devising the programme and will provide electronic patient record data to the CSOR research database:

• Alder Hey Children’s Hospital

• Birmingham Children’s Hospital

• Cambridge University Hospitals NHS Foundation Trust

• Chelsea & Westminster Hospital

• Evelina Children’s Hospital

• Great Ormond Street Hospital

• Oxford University Hospitals NHS Foundation Trust

• Royal Hospital for Children

• Royal Manchester Children’s Hospital

• Southampton General Hospital

These organisations in the bulleted list above are not involved in any way in determining the means and purpose of processing personal Data received from NHS England and will have no access to record level Data from NHS England.

The CSOR study is funded by the National Institute for Health Research. The current funding period is 01/03/2020 to 28/02/2025. The NIHR are involved in monitoring the progress of the study and reviewing study protocols, but do not decide the purpose and means of processing the data in the CSOR study, and are therefore not considered to be a Data Controller in this agreement. NIHR have no ability to control or suppress the outcomes published under this programme.

UK GDPR LEGAL BASIS FOR THE PROCESSING OF DATA

The University of Oxford, as a joint Controller who is also processing the Data will process Personal Data under UK GDPR 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. As a higher education establishment, the University of Oxford conducts research to improve health care and services, and the data requested is necessary for the performance of a task carried out in the public interest.

Oxford University Hospitals NHS Foundations Trust, as a joint Controller who is also processing the Data will process Personal Data under UK GDPR 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. Oxford University Hospital NHS Foundation Trust is a public authority. The Data Protection Act 2018 s7(1)(a) defines ‘public bodies’ for the purpose of the UK GDPR as “a public authority as defined by the Freedom of Information Act 2000”. The FOI Act 2000 Part 1, section 3 (1)(a)(i) specifies that a public authority means any body which is listed in Schedule 1. Schedule 1 Part 3 (40A) of the FOI Act 2000 stipulates “An NHS foundation trust” is a public authority.

The NHS Act 2006 section 43(5), which describes the functions of authorised NHS Foundation Trusts, states that ‘The authorisation must authorise and may require the NHS foundation trust— (a) to carry out research in connection with the provision of health care, (b) to make facilities and staff available for the purposes of education, training or research carried on by others'.

Additionally, under GDPR Article 9(2)(j) processing of Special Category Personal Data (of which Health data is one) is necessary for archiving for research purposes. Data minimisation processes are being followed and only data that is specifically required for the purposes of this study have been requested, to protect the rights of the data subjects. The Controllers have satisfied themselves that this request is appropriate, necessary and proportionate for the performance of the task described in the Purpose statement and that there is no other reasonable and less intrusive means to achieve their purpose.

PATIENT AND PUBLIC INVOLVEMENT AND ENGAGEMENT

The need for the CSOR programme was first identified and developed through work with the Parental Advisory Group (PAG) set up by the study team at the University of Oxford. The PAG, consisting of over 100 parents/families of children with surgical conditions, charities and support group representatives from across the UK, remain actively involved throughout the course of the programme. As a minimum, annual meetings have been held to update, discuss and gather feedback on the aims and methods of establishing the research database. Feedback from the PAG has been key to the development of the proposed parent consent and data collection process. There will be two parent or patient representatives on the CSOR Research Database Steering Committee to ensure that the parent/patient voice is maintained in the functioning of the CSOR Research Database. The CSOR Research Database will continue to be reviewed at the annual PAG meeting. The members of the CSOR Research Database Steering Committee will have no access to the NHS England Data described in this Data Sharing Agreement.

The dissemination for this Data Shairng Agreement will require sharing of identifiable data items about eligible children and their parents without the consent of either party. There is the potential that some parents may object to this, and subsequently finding out that their/their child's data have been used for research purposes without consent may cause them distress. Instead, the study has support under section 251 of the NHS Act 2006 to enable the common law duty of confidentiality to be temporarily lifted so that confidential patient information can be processed without consent. This is an area that has been explored in depth with a parent advisory group consisting of a large number of parents of children with surgical conditions, as well as representatives of charities and support groups. Feedback from this group has been overwhelmingly positive. Most parents assume that data are already shared for the purposes of research, and as long as data are appropriately protected, parents did not have any objections to these data being shared. The potential benefit from sharing of the data in terms of developing a far greater understanding of the models of care and treatments that are best for children with surgical conditions, as well as the potential to reduce unwarranted variation in management and outcomes for these children therefore far outweighs the potential harms of the data collection.

Previous attempts to set up condition-specific databases for key paediatric surgical conditions have been limited by significant issues with case ascertainment, as they relied on voluntary case reporting by surgical staff. These databases have shown that reliance on site staff to supply contact details results in approximately 30% of the eligible population not being approached for consent, as site staff have been unable to supply their contact details. Where contact details are received though, and parents are approached for consent without prior knowledge that their contact details have been shared, this results in a relatively high consent rate (~60%). In a recent (2017-2019) paediatric surgical study in which approximately 190 parents were approached by a research team without prior consent, no complaints were received about the fact that their contact details had been shared outside of their usual clinical team without prior consent.

Currently, children with the conditions included in this Programme have poor outcomes and significant long-term health needs. These include 25-30% one-year mortality rates for infants with necrotising enterocolitis and congenital diaphragmatic hernia, 60% rates of primary school aged faecal incontinence for children with Hirschsprung’s disease, and the need for multiple operations in the first 28 days of life, for infants with gastroschisis. Management of these children in the UK currently varies widely. No mechanisms exist to determine which elements of this variation are acceptable, and which are unwarranted and associated with variation in outcome. The Data disseminated under this DSA hopes to provide the information that is needed to calculate the number and types of operations that children have undergone, as well as the number of times they have been admitted and treated for an infection related to their underlying surgical condition.

These outcomes contribute towards calculating a child’s CSOR treatment success score. When used in conjunction with the additional data held in the CSOR research database, knowledge of a child’s CSOR treatment success score will create for the first time the ability to understand which elements of management are inconsequential, and which are unwarranted. The ability to detect unwarranted variation in management hopes to help standardise delivery of care across different surgeons and hospitals, with the expectation that the overall delivery of care for children with these conditions will improve. This hopes to, in the long run, improve the health and wellbeing of children born with these conditions in the future, as well as allowing evidence-based restructuring of the way care is delivered for these children.

The following outputs hope to be developed:

1) A Research Database (the CSOR Research Database) which can be utilised for identifying unwarranted variation in the management and outcomes of children with surgical conditions, and for the conduct of approved research to improve outcomes for children with surgical conditions. Data collection is anticipated to begin in quarter 3 of 2023.

2) Peer reviewed publications and presentations relating to development of the CSOR Research Database and CSOR Research Database methodology. Publication is anticipated in quarter 3 of 2024.

3) A ‘case mix adjustment model’ used to predict how successful an infant’s treatment is expected to be based upon their underlying characteristics. The initial model will be developed using data collected in the first year of the CSOR Research Database, and will be revised initially on a yearly, and then likely five-yearly, basis as additional data are added to the database.

4) Peer reviewed publications and presentations describing the development of the case-mix adjustment model. Publication of the initial model is anticipated in quarter 2 of 2025.

5) Peer reviewed publications and presentations relating to the results of analyses conducted on the pseudonymised dataset. These will include, but not be limited to, analyses describing management strategies, infant outcomes, and factors affecting outcomes. The date of these publications will be influenced by case incidence and ascertainment but are not anticipated until quarter 4 of 2025 at the earliest.

6) Live dashboards accessed by hospitals and the public describing the activity in participating hospitals and comparisons of the hospital’s observed and expected outcomes. Development is anticipated to be complete by quarter 4 of 2023.

7) Facilitated feedback and peer review processes, the aims of which are to help hospitals understand their own performance, will be informed by analysis of data contained in the final linked, pseudonymised dataset. Development of these processes is anticipated throughout 2023, with implementation in quarter 3 or 4 of 2024.

8) Professional and public facing annual reports describing the activities of the CSOR Research Database, and including, amongst other things, information relating to case ascertainment, national management strategies, and national benchmarks for infant outcomes. Anticipated yearly from quarter 3 2024 onwards.

9) Podcasts, posters, information leaflets and videos explaining to parents, patients and the public, the results of analyses conducted using data contained in the CSOR Research Database. The date of these outputs will be influenced by case incidence and ascertainment but are not anticipated until quarter 4 of 2025 at the earliest.

10) Peer reviewed publications and presentations describing the periodic process evaluation conducted to determine whether implementation of the CSOR Research Database and facilitated feedback programme has improved the health and wellbeing of children with surgical conditions. The results of the initial process evaluation are anticipated to be published in quarter 3 of 2025.

All outputs will contain only aggregated data with small numbers suppressed as per the HES Analysis guide. As per funder requirements (NIHR) all publications will be open access. All data and knowledge are owned by the University of Oxford.

A wide range of routes will be used to disseminate the results of the programme to the beneficiaries, including children with one of the included conditions, adults who were previously treated for one of the included conditions, parents of children with one of the included conditions, professionals and policy-makers.

Dissemination strategies to parents and the public will be guided by one of the CSOR co-investigators who is an expert by experience (a parent of a child with a surgical condition), as well as by an established parental advisory group consisting of parents of children with surgical conditions, and representatives of relevant support groups and charities. Leaflets, videos, infographics and social media will be used as appropriate to aid dissemination.

The programme has direct links with NHS England and the Department of Health and Social Care and will disseminate directly to relevant policy-makers. that the CSOR programme team aim to ensure the outputs are appropriately used to commission services in England, and that full dissemination of outcomes to professional colleagues and integration with professional standards and guidelines. The study team have agreement to specifically hold a dissemination event at the British Association of Paediatric Surgeons Annual Conference to ensure the results reach the paediatric surgical audience directly. The study team also hope to disseminate through journal publications and presentations at other key conferences. Dissemination to the wider public aims to be via regular newsletters available on our website, website bulletins and social media communications via the CSOR Twitter account.

The key route of dissemination to the involved professionals/stakeholders will be through the use of the live dashboards, and facilitated feedback and peer review processes. Through these, professionals involved in the care of infants whose data are included in the CSOR Research Database will be guided through the interpretation of the data for their own hospital in order to help identify any ways in which practice could be changed to improve the care that is being delivered. Data collected by the CSOR system will be used to develop benchmarked, auditable standards for delivery of paediatric surgical services, against which individual units can compare themselves, and be compared.

Agreement has been forthcoming from the National Consultant Information Programme for this database, once successfully developed to reside within NHS England in order to facilitate individual surgeon feedback via the The National Council of Integrative Psychotherapists (NCIP) portal (http://gettingitrightfirsttime.co.uk/ncip/), and the study team have also had an indication that in Scotland it could be hosted within the Information Services Division. This important surgeon-level ‘building block’ will allow the data to be built from surgeon to unit to national level. Use of the data to identify gold-standard practice hopes to allow clinicians, managers and commissioners to review and modify the way their services are organised and delivered.

Yielded Benefits is not a requirement for new applications.