Modelling the transition from neonatal to paediatric care: a data linkage study

Following birth, around one in seven babies are admitted for specialist neonatal care in the UK. Admission rates to neonatal care have increased in the last few years, partly due to improved survival of the most vulnerable babies, particularly those born very prematurely or those with serious health problems. More and more of these babies now survive, but the impact of their health and the care received immediately after birth can be lifelong. There has also been an increase in admissions to paediatric intensive care units (PICU) in the last ten to fifteen years. Many admissions may relate to children who received neonatal care immediately after birth, although the exact number is not known.

Very little is known about what happens between neonatal and paediatric care including which children are likely to experience both types of care, and how clinical services, parents and professionals manage the transition. This research will link together neonatal and paediatric care records for the first time to allow investigation of the first two years of the lives of these children.

This project is part of a larger study funded by the National Institute for Health Research Advanced Fellowship programme. The University of Leicester relies on the GDPR Article 6(1)(e) for the legal basis for processing data (processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller). Additionally, as health data is a special category of Personal Data, the Data Controller relies on 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).

The University of Leicester require linked pseudonymised record-level data for use in the data linkage workstream of this project (IRAS ID: 283808). The University of Leicester will not release any data to any third-party organisations. Any results from this study will be published in aggregate form, with small numbers suppressed as per the HES analysis guidance.

Pseudonymised data related to England (Hospital Episode Statistics (HES) Admitted Patient Care (APC) and Accident and Emergency (A&E) / Emergency Care Data Set (ECDS)), and England and Wales (Civil Registrations (Deaths) data is requested. This will allow for a rich population-based cohort, which is not possible to achieve via any other data sources.

The study team require access to the Civil Registrations (Deaths) data to investigate deaths which occur outside of neonatal or paediatric intensive care (e.g. deaths at home or in hospices). The study team require access to HES ACP/ A&E/ECDS to investigate and understand the healthcare resources used by these children. Without access to this data the study team will only know about intensive care, when the other types of care (e.g. ward and Accident and Emergency) are likely to be more commonly accessed.

This project forms part of a larger study which has three workstreams:

(1) data linkage of neonatal and paediatric data to investigate outcomes in the first two years of life;

(2) exploration of neonatal discharge practices and

(3) understanding the experiences of parents who have had a critically ill child.

This data request forms the entirety of workstream (1) and the results will inform aspects of workstream (2). Workstream (1) will investigate the needs of clinical subgroups of babies (e.g. preterm babies, babies with heart problems) and will include two PhD projects. One will be looking at the outcomes of preterm born children. The other will be investigating the outcomes of children who experience a brain injury around the time of birth. Both PhD student will focus their section of the research on neonatal care, intensive care (data from PICANet) and deaths data. The student investigating preterm born children will not require access to Hospital Episode Statistics (HES) Admitted Patient Care (APC) and Accident and Emergency (A&E) data sets. The student investigating brain injuries may require access to these datasets but will only be given a subset of variables needed for their analysis.

Ethics approvals are in place for this study.

The Data Controller for this agreement is the University of Leicester as they determine the purpose and means of processing the data. The University of Leicester will also be the sole processor of NHS England data. The NNRD (based at Imperial College, London) and PICANet (based at the University of Leeds and is run in collaboration with the University of Leicester) will provide data for this project. The project is funded by the National Institute for Health Research, but they have no involvement in the design, planning or running of this study, and as such they are not considered Data Controller.

This study is supported by a Study Expert Advisory Group and a Parent Advisory Group. The Study Expert Advisory Group contains a paediatric and neonatal doctor, an epidemiologist, a social statistician, a nurse and a parent representative The Parent Advisory Group contains parents who have all had experience of children in Paediatric Intensive Care Units (PICU) and Neo-natal care and the group is supported by a representative from the charity, Bliss, who is the leading UK charity supporting families with a baby born premature or sick. Both groups provide regular input into the study for Patient and Public Involvement (PPI). The project has been designed with parents and families at the heart of it and developed with parents who had children who experienced neonatal care, and all suggestions and ideas were discussed with a diverse group. The Parent Advisory Group was established for the purposes of this study and the initial idea for this project was conceived during a PPI meeting of a previous research study. The parents and families will continue to be involved throughout, including the co-production of materials for parents and healthcare professionals at the end of this project

NHS England is requested to link together two data sources: NNRD (National Neonatal Research Database) and PICANet (Paediatric Intensive Care Audit Network) containing identifiable data (NHS Number, Date of Birth and Postcode) along with a unique Study ID. NHS England will link the NNRD and PICANet and provide information about common records and also about records only in the NNRD or only in PICANet.

The NNRD will provide NHS England with a Unique study ID for each record in their cohort and PICANet will also provide a Unique study ID for each record in their cohort. The unique study IDs they provide will be unique to each child in their separate cohorts, and as neither NNRD nor PICANet know what records they are each sending to NHS England, there may be a NNRD AND PICANET Study ID for one child. NHS England will link the NNRD and PICANet cohorts and provide information about common records and also about records only in the NNRD or only in PICANet.

Health data from Hospital Episode Statistics (HES) Admitted Patient Care (APC)and Accident and Emergency (A&E) / Emergency Care Data Set (ECDS) and Civil Registrations (Deaths) data will then be linked to the combined NNRD/PICANet cohort. Only data related to the first two years of life of the child is required to understand the potential short-term impact of care received in the neonatal period immediately after birth.

The team at the University of Leicester only require access to pseudonymised record level data which will allow for additional data from the NNRD and PICANet to be added in. The NNRD will provide data from 1 January 2013 to 31 December 2018 and PICANet will provide data from 1 January 2013 to 31 December 2020 (this includes 2 years follow up data). This allows for all children in the cohort to have reached two years of age. Hospital Episode Statistics (HES) / Emergency Care Data Set (ECDS) and Civil Registrations (Deaths) data is required from 1 January 2013 to 31 December 2020 for children aged 2 years and below. This will allow observation of two years of follow up for all children in the cohort.

METHODOLOGY

(1) The NNRD (Imperial College London) and PICANet (University of Leeds) will upload a cohort each via the Secure Electronic File Transfer Service (SEFT), made up of approximately 480,000 individual records in total. The cohort will each contain a Unique Study ID, and the personal identifiers: NHS Number, Date of Birth, and Postcode.

(2) NHS England will identify common records between the two cohorts to create one main cohort link to NHS England HES/ECDS and Mortality data. NHS England will also highlight records which are unique to the NNRD or PICANet via an Encrypted HES_ID.

(3) NHS England will use the single cohort to extract HES and Mortality data for children in the first two years of life. (i.e. only provide data of children aged 2 years and under from date of birth 1 January 2013 to 31 December 2018).

(4) NHS England will then remove identifiable fields from the HES/ECDS and Mortality extracts, leaving the unique study ID.

(5) NHS England will return the pseudonymised HES/ECDS and Mortality extracts to the Data Recipient (University of Leicester) via SEFT.

(6) Data flowing to the University of Leicester from the NNRD/PICANet - The NNRD and PICANet will provide requested clinical data for all records provided to NHS England with the pseudonymised identifier (Pseudo-ID) which can be used by the team at Leicester to link the NNRD/PICANet data with that provided by NHS England via the Study ID only. Approval has been granted by HQIP for the flow of PICANet data. DSAs are in place for data flow between NNRD/PICANet and the University of Leicester.

All statistical data analyses will be undertaken at the University of Leicester. Only substantive employees or PhD students of the University of Leicester will be allowed to access the data. All staff and students are required to undertake the University annual data security training plus prescribed annual NHS security training. Students work under the direct supervision of SES. Any data breaches or unethical behaviour are investigated promptly and can result in disciplinary action or student expulsion.

No identifiers are received or stored at the University of Leicester. There will be no linkage of NHS England data to other data sets other than those state in this agreement.

Statistical data analysis will be carried out on University of Leicester managed machines connected to the secure university server either directly in person or remotely via a Virtual Private Network (VPN), using an appropriate statistical package. Only substantive employees of the University of Leicester can access this secure server. Students can also access the server if granted appropriate permissions. All access is restricted to members of the project team (staff and students). Areas of the secure server dedicated to research projects are restricted to members of the research team. To remotely access the secure server, a University of Leicester managed machine is required alongside 2-step factor authenticator and connection via the VPN. Users are then able to securely access the secure server on the University’s IT framework. All data analysis will be conducted within the confines of the University’s secure server and will not be downloaded to remote devices for storage or processing.

HES and ECDS DISCLOSURE CONTROL / SMALL NUMBER SUPPRESSION

In order to protect patient confidentiality, when presenting results calculated from HES record level data, outputs will contain only aggregate level data with small numbers suppressed in line with HES Analysis Guide. When publishing HES data, you must make sure that:

· cell values from 1 to 7 are suppressed at a local level to prevent possible identification of individuals from small counts within the table.

· Zeros (0) do not need to be suppressed.

· All other counts will be rounded to the nearest 5.

Data will not be made available to any third parties other than those specified except in the form of aggregated outputs with small numbers suppressed in line with the HES Analysis Guide.

The results of this study are hoped to be disseminated widely. The research team has strong links with professional organisations in this area including: the British Association of Perinatal Medicine; the Royal College of Paediatrics and Child Health and the Neonatal Nurses Association.

PUBLICATIONS AND REPORTS

The study team aims to make versions of the study protocol and analysis plans available for anyone to access on the study website. (https://www2.le.ac.uk/departments/health-sciences/research/timms/staff-pages/ses26 or another website to be developed). At least two peer-reviewed publications from this workstream of the research project (publish between 2022 and 2028) are planned. These publications aim to be provided via open access and lay summaries are planned to be provided of all research. Annual reports are planned to be provided to the National Institute for Health Research as part of the University of Leicester's ongoing research project.

CONFERENCES

Healthcare professionals are involved in supporting this research project, and the findings of this research project are planned to be disseminated through their connections. Key findings from this work aim to be presented at national and international conferences (e.g. World Congress of Pediatric Intensive Care) and meetings (e.g. the annual meeting of the Paediatric Intensive Care Audit Network). No individual level data will be presented or included in summaries of findings (e.g. results from statistical models), only aggregated and suppressed outputs as per the HES Analysis guide. Dissemination at conferences are planned throughout the research project (2022-2028). At the end of this project an online meeting open to researchers and healthcare professionals plan to be hosted as well as potentially podcasts and pre-recorded videos to disseminate key findings of the research.

PUBLIC

The results of this research aims to be promoted to parents, families and the public via the website, social media and via charities. Lay summaries or alternative methods (e.g. podcasts) are planned to be provided, written in collaboration with parents and families involved in this project. These aim to be disseminated via social media and other appropriate platforms throughout the project.

PHD THESES

Elements of this work will form part of PhD students work including their theses which will be publicly available upon completion.

The data provided in all outputs will be aggregated with small number suppression as per the HES Analysis Guide.

This study hopes to address an important research question about the outcomes of children following admission to neonatal care after birth. At the moment, no one knows how many of these children require paediatric care in the early years of their lives. This study aims to publish its first results with 18 months of the receipt of data and have published all findings within three years. The study hopes to lead to recommendations and improvements in care in the following ways:

1. Preparation for families being discharged from neonatal care

Currently, parents are provided with limited or no guidance about the likelihood of their child requiring a future admission to paediatric care. For some children, it is likely they will have ongoing healthcare needs throughout their early lives and this study will help the study team identify what child characteristics are associated with admission to paediatric care or paediatric intensive care. This information can be used, as appropriate, to counsel parents at the point of discharge from the neonatal unit.

2. Policy around neonatal discharge

There is no national guidance about when babies may be ready to be discharged from neonatal care. This study hopes to provide insight into what clinical choices (e.g. discharging a baby home on oxygen) may be associated with an increase in the risk of admission to paediatric care. This hopes to provide information to begin the development of policies surrounding discharge from neonatal care.

This work will involve several peer reviewed publications. The study team hopes to also present their work to local and national clinical commissioning groups and organisations (e.g. British Association of Perinatal Medicine, Paediatric Critical Care Society) and at relevant national/international conferences. The study team will work with parent charities and other organisations to provide appropriate outputs for policy makers including NHS England to which the study team will disseminate through links they have within the relevant organisations.

3. Organising healthcare services around the needs of children who require neonatal and paediatric intensive care

In recent years the number of children surviving neonatal care has increased and the number of children living with chronic health conditions is also increasing. Currently, the study team do not understand the demands and workload requirements this will place on the National Health Service. This study aims to allow the study team to investigate current demands and investigate if trends are changing over time to enable the healthcare service to prepare better for the future.

4. Increasing the public conversation around neonatal and paediatric care

There is a lack of awareness about the impact of neonatal care, but in reality nearly 1 in 7 babies experience an admission. Therefore, this research hopes to increase public awareness about the care provided to this cohort of babies and children.

To date The Data Controller has presented work at conferences and The Data Controller has had one paper published (see: https://fn.bmj.com/content/early/2023/11/03/archdischild-2023-325970) with further work forthcoming.

The benefits and updates for 01/2024 are as follows:

1. Improved preparation for families being discharged from neonatal care

The Data Controller have published one paper exploring the risk of PICU admission for babies born very preterm. The Data Controller has also presented this work at multiple academic conferences and future publications and dissemination are planned.

2. Policy around neonatal discharge

National Neonatal Outreach Group is producing suggestions and guidance around the support offered at the point of neonatal discharge, using findings from this research.

3. Organising healthcare services around the needs of children who require neonatal and paediatric intensive care. The Data Controller have a paper currently under review which is exploring the organisation of care for these children in neonatal and paediatric services.

4. Increasing the public conversation around neonatal and paediatric care. The Data Controller are hoping to hold a public event focussing on those children with complex care needs in early life, later this year.