FIRST-line support for Assistance in Breathing in Children (FIRST-ABC): A master protocol of two randomised trials to evaluate the non-inferiority of high flow nasal cannula (HFNC) versus continuous positive airway pressure (CPAP) for non-invasive respiratory support in paediatric critical care

FIRST-ABC is a master protocol of two pragmatic randomised clinical trials (RCTs) aiming to evaluate the clinical and cost-effectiveness of the use of High Flow Nasal Cannula (HFNC), as compared with Continuous Positive Airway Pressure (CPAP), when used as the first-line mode of non-invasive respiratory support in two distinct clinical scenarios:

1. in critically ill children requiring non-invasive respiratory support for an acute illness (step-up RCT); and

2. in critically ill children requiring non-invasive respiratory support within 72 hours of extubation following a period of invasive ventilation (step-down RCT).

Both HFNC and CPAP are currently routinely used across the NHS. This research is important to patients and the NHS because there is currently limited high-quality evidence to support whether HFNC or CPAP should be used as the first line mode of non-invasive respiratory support in critically ill children. In addition, previous research has also not studied the use of these interventions in the two distinct clinical scenarios outlined above. High quality evidence from RCTs is therefore urgently needed to help guide clinical decision making and to inform paediatric critical care clinicians and the wider NHS on the clinical and cost-effectiveness of HFNC versus CPAP – evidence which FIRST-ABC aims to provide.

FIRST-ABC is funded by the National Institute for Health Research (NIHR), Health Technology Assessment Programme. NIHR do not determine the purpose or the manner in which the data will be processed. FIRST-ABC is sponsored by Great Ormond Street Hospital for Children NHS Foundation Trust (GOSH) and managed/coordinated by the Intensive Care National Audit & Research Centre (ICNARC). The health economic analysis will be completed by a health economist based at the London School of Hygiene and Tropical Medicine (LSHTM) on behalf of GOSH and ICNARC. FIRST-ABC aims to include a total of 1,200 children (600 in the step-down RCT and 600 in the step-up RCT, randomised to either HFNC or CPAP) from around 25 NHS paediatric critical care units across England, Wales and Scotland. NHS England data is requested for all cohort members who have had health events of interest within England and Wales.

Processing personal data is necessary for ICNARC's legitimate interests which are described in this agreement. The data to which access is requested are proportionate and necessary to achieve those interests. ICNARC has completed a legitimate interests assessment (LIA) and is satisfied that the interests of the data subjects do not override ICNARC’s legitimate interests; that they would reasonably expect the processing and it would not cause unjustified harm. The data subjects interests and fundamental rights are protected through appropriate minimisation of fields and patient records being processed; pseudonymisation to minimise any risk of identifying individuals; protection of the data in a secure environment, and guaranteeing secure destruction at any stage at the request of NHS England or after a defined period on completion of the project.

The primary objective for processing NHS England data in this application is to provide important outcome data for FIRST-ABC. The outcomes for the trial are:

Primary outcome:

• Time to liberation from respiratory support, defined as the start of a 48-hour period during which the child was free of all forms of respiratory support.

Secondary outcomes:

• Mortality at paediatric intensive care unit (PICU) / high dependency unit (HDU) discharge, day 60 and day 180

• Rate of (re)intubation at 48 hours

• Duration of PICU/HDU and hospital stay

• Patient comfort, during randomised treatment and during non-invasive respiratory support (i.e. HFNC and/or CPAP), assessed using the validated COMFORT-B score

• Proportion of patients in whom sedation is used during non-invasive respiratory support

• Parental stress, in hospital at the time of consent at/around 24-48 hours, measured using the Parental Stressor Scale: PICU

• Health-related quality of life (HrQoL) at six months using age-appropriate Pediatric Quality of Life Inventory (Peds-QL) and the Child Health Utility 9D (CHU-9D) questionnaires

Cost effectiveness analysis (CEA) outcomes:

• Total costs at six months

• Quality-adjusted life years (QALYs) at six months

• Incremental net monetary benefit gained at a willingness-to-pay of £20,000 per QALY at six months associated with HFNC versus CPAP

The following NHS England data will be processed: Mortality (date of death), and selected fields from Hospital Episode Statistics (HES) Admitted patient care (APC) and HES Outpatients (OP) datasets, and the Emergency Care Data Set (ECDS) - these data will be unsuppressed, pseudonymised and provided at record level. The data is required to be provided at record level as it will be linked to the individual FIRST-ABC Study participant trial data using the FIRST-ABC trial number.

Linking the FIRST-ABC data with HES and ECDS data from NHS England will contribute to the integrated economic evaluation to assess the cost-effectiveness of the first line use of HFNC versus CPAP. ICNARC have carefully reviewed the HES data dictionary to ensure that the data requested is the minimum required to meet the study objectives. Only those which are essential to meet the objectives of the research have been chosen. Specifically, HES OP, HES APC and ECDS data will be used to understand the economic impact of HFNC versus CPAP. Selected fields are those anticipated to drive potential differences in costs and resource utilisation between the groups and will be important in assessing the cost effectiveness of HFNC versus CPAP.

Patient dates of death (if applicable) are requested from NHS England to directly contribute to the clinical evaluation secondary outcome data (mortality at various time-points) (primary purpose). In addition, parents of trial participants are actively followed-up and sent a questionnaire by the FIRST-ABC trial team at ICNARC at six months post-randomisation. Given the nature of critical illness, unfortunately some participants will pass away during the trial follow-up period. Where the trial team learns that a participant has passed away, no contact will be made with parents - helping to ensure parents and relatives are not caused undue distress by no longer appropriate contact (secondary purpose).

In line with the timescales stated in the study outcomes, only six months of NHS England data will be requested for each study participant from their randomisation date (180 days from the date of randomisation). Both the step-down RCT and step-up RCT commenced patient recruitment in August 2019.

Recruitment for the step-down RCT finished in May 2020 and the final participant will have been in the trial for six months as of November 2020 (the final data collection timepoint for this cohort). Therefore, only data from years 2019/2020 and 2020/2021 is required for the step-down RCT cohort.

The recruitment period for the step-up RCT is planned to complete in January 2022 (given no delays) and subsequently the final patient in this cohort will have been in the trial for six months at the end of July 2022. Therefore, data from the following years will be needed for the step-up RCT cohort: 2019/2020, 2020/2021, 2021/2022 and 2022/2023.

The decision to initiate non-invasive respiratory support is most often made during a time-sensitive emergency situation, where any delay in commencing treatment could be detrimental to the patient and the scientific validity of the trial. This makes attempts to obtain fully informed prior consent from parents/legal guardians during such an emergency situation inappropriate and could cause additional stress to families who are already very distressed by their child’s critical illness. In addition, both modes of non-invasive respiratory support evaluated as first-line treatment in this study (CPAP and HFNC) are relatively safe, commonly used and in current clinical practice - only determined by individual clinician preferences. Considering these reasons, a deferred consent model (‘research without prior consent’) has been utilised in the FIRST-ABC Study, a model that has been found to be acceptable to parents/guardians as well as clinicians in several previous RCTs conducted in the paediatric critical care setting. This consent model has been informed and refined by extensive patient and public involvement (PPI) work and is based on the CONseNt methods in paediatric Emergency and urgent Care Trials (CONNECT) study guidance. The East of England – Cambridge South Research Ethics Committee - who are specifically flagged and expertly positioned to review research involving children in the NHS - has granted the use of the research without prior consent model to FIRST-ABC. All of the reasons outlined above demonstrate that there are no alternative, less intrusive ways of achieving the study purposes than linking FIRST-ABC data with NHS England data.

ICNARC is an independent health research charity which aims to help improve the quality of critical care through audit, research and education with, and in the interests of, patients and those who care for them. ICNARC requires data from NHS England for the purposes of these legitimate interests.

ICNARC process data for this study under the legitimate interest legal basis. This is because ICNARC is a registered charity and the data processing described here is to support scientific and statistical research. Specifically, data is processed under the following articles of the Data Protection Act 2018:

Article 6 (1) (f) Legitimate interests: the processing is necessary for your legitimate interests or the legitimate interests of a third party unless there is a good reason to protect the individual’s personal data which overrides those legitimate interests.

The data is required for a research project - and therefore ICNARC process the special category data (health data) under the following article of GDPR - 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.

Great Ormond Street Hospital for Children NHS Trust rely on Articles 6 (1) (e) – “public interest” and 9 (2) (j) – “scientific research” as the legal bases under GDPR.

The organisations involved in this application as follows: ICNARC – joint data controller, manages the study (as the Clinical Trials Unit), holds the study data and will conduct the primary analyses (i.e. a data controller who are also processing the data); Exponential-E are a contractor of ICNARC’s and provide the servers (including back-ups) on which data will be processed by ICNARC staff; London School of Hygiene and Tropical Medicine (LSHTM) is where the study health economist is based who will conduct the health economic analyses; Great Ormond Street Hospital for Children NHS Trust, as trial sponsor, is joint data controller but will not receive or process any trial data (including any data from NHS England). The trial management group includes members from the following organisations: Great Ormond Street Hospital for Children NHS Trust, University College London, Great Ormond Street Institute of Child Health, Birmingham Women's and Children's NHS Foundation Trust, University of Salford, University Hospitals Bristol NHS Foundation Trust, University of Leeds, LSHTM, and ICNARC. However, only ICNARC and LSHTM will receive and process study data.

For consented participants discharged alive from hospital for the admission from which they were recruited into the FIRST-ABC study, data from NHS England was requested to contribute to the analysis of the study outcomes, alongside separate trial data collected from hospital records at participating sites, and reported outcomes from the participants themselves.

To obtain pseudonymised record level HES APC, HES OP, ECDS and Civil Registration (Deaths) data from NHS England, an authorised member of the FIRST-ABC Study team at the ICNARC CTU provided identifiable record level data on a spreadsheet to NHS England via Secure Electronic File Transfer on two separate occasions. These spreadsheets will contain the following data fields: FIRST-ABC Trial number, date of randomisation (start date for calculating the six-month follow-up period), full name, date of birth, NHS number and postcode.

Cohort data retrieval from NHS England occurred at two time points. Cohort data retrieval number one was for the step-down RCT cohort. This occurred upon the sign off of the data sharing agreement. This cohort contained the maximum of 600 patients (participants who passed away in hospital, declined or withdrawn consent will not be submitted). Patients in this cohort were recruited between August 2019 and May 2020. The final participant will have been in the trial for six months as of November 2020 (the final data collection timepoint for this cohort). Therefore, to ensure each participant has six months of HES/ECDS data, data from years 2019/2020 and 2020/2021 are required for the step-down RCT cohort. Civil Registrations (Deaths) data were provided at the same time as HES/ECDS data.

Cohort data retrieval number two was for the step-up RCT cohort. This occurred once the last patient recruited in this cohort will have been in the trial for six months. This cohort contained the maximum of 600 patients (participants who passed away in hospital, declined or withdrawn consent will not be submitted). The recruitment commenced in August 2019 and is anticipated to complete in January 2022 (given no delays). This will mean that the final patient in this cohort will have been in the trial for six months at the end of July 2022. Therefore, to ensure each participant has six months of HES/ECDS data, data from years 2019/2020, 2020/2021, 2021/2022 and 2022/2023 are required for the step-up RCT cohort. Civil Registrations (Deaths) was provided approximately 6 weeks after the final patient follow-up date.

Members of the FIRST-ABC Study team at ICNARC will access the trial data and NHS England data via secure ICNARC databases, including using a VPN connection when working remotely. Remote access is required in line with national UK Government guidance during the COVID-19 pandemic to work from home wherever feasible. All members of the ICNARC FIRST-ABC Study team work on ICNARC-provided devices that are password protected and encrypted to the secure standard required by the UK regulator for data protection. All study data are stored in databases or folders where access is restricted to authorised essential members of the FIRST-ABC study team only. The data is held on secure servers provided by Exponential-E. ICNARC will also regularly back-up the data through Exponential-E. Exponential-E is an IT contractor authorised by ICNARC with ISO 27001 security assurances in place.

A small number of named members in the FIRST-ABC study team at ICNARC will link up the identifiable trial data with the mortality status and date of death (where applicable) from the Civil Registrations (Deaths) file provided by NHS England. This is in line with reasonable expectations of the parents of trial participants, as outlined in ‘Objectives for Processing’.

The FIRST-ABC study team at ICNARC will additionally create a fully pseudonymised trial dataset which will be linked to both the pseudonymised deaths data, and the pseudonymised HES/ ECDS data from NHS England using the unique trial number field. Linking the FIRST-ABC trial data with the NHS England data will allow ICNARC to complete analysis of mortality at various time points – one of the secondary outcomes of the study. The pseudonymised linked record level dataset will also be made available to the health economist, who is part of the FIRST-ABC study team, based at the London School of Hygiene and Tropical Medicine (LSHTM) who will conduct the health economic evaluation.

To minimise the risk of re-identification of study participants, the pseudonymisation process for the trial data will include the following: allocated trial numbers and site codes will be replaced by new identifiers, the link between actual trial numbers/site code will be held securely by ICNARC and not shared with LSHTM; Date of birth will be used to calculate calendar age (in months and/or years) at randomisation, only calculated age and not date of birth will be shared with LSHTM; All dates of any events post-randomisation (including hospital discharge and death) will be removed from the dataset and replaced by calculated fields indicating the elapsed time from randomisation; Date of randomisation will not be included. The same method will be used to remove dates of any pre-randomization events. All free text fields will either be removed from the dataset before transfer, or individual entries will be checked and any identifiable information (including patient, staff or hospital names) will be removed. The subsequent analyses will directly answer the trial research questions and achieve the purpose of finding out whether HFNC is non-inferior to CPAP for the benefit of future NHS patients.

The data is held on secure servers at LSHTM. Data will be stored in an on-site data safe and data will only be accessed by authorised individuals within LSHTM. There will be no requirement or attempt by the team at LSHTM to reidentify individuals. All data processing will only be carried out by substantive employees of named data controllers and processors who have been appropriately trained in data protection and confidentiality. The essential members of the study team at LSHTM will access the data through a secure link via designated LSHTM password protected computer whilst working from home. The data will be stored in a folder where access is restricted to essential members of the study team only. Remote access is required in line with national UK Government guidance during the COVID-19 pandemic to work from home wherever feasible.

The results of FIRST-ABC will be both widely and actively disseminated. The research team has strong links with the PICU community via the Paediatric Intensive Care Society (PICS), PICS Study Group (PICS-SG), and the NIHR CRN: Children Clinical Studies Group (CSG) in Anaesthesia, Intensive Care and Cardiology, and similarly with the nursing community through the British Association of Critical Care Nurses (BACCN), the Royal College of Nursing Critical Care and In-flight Nursing Forum (RCN CCINF) and the European Society of Paediatric and Neonatal Intensive Care (ESPNIC). The research team also has links with the Healthcare Quality Improvement Partnership national audit programme through the Paediatric Intensive Care Audit Network (PICANet).

The findings from our work will be presented at national and international conferences, potentially including the Annual Conference of the Royal College of Paediatrics and Child Health, the World Congress of Pediatric Intensive Care, PICS Annual Scientific Meeting, American Association of Pediatrics Conference, ESPNIC Annual Meeting, and British Association of Critical Care Nurses (BACCN).

A comprehensive report will be submitted to the NIHR for publication in the peer reviewed, open access Health Technology Assessment journal and will include recommendations for future policy, practice and research. The results will also be submitted for publication in in a high-impact, widely-read, open-access (where possible), general medical journal, such as the New England Journal of Medicine or the Journal of the American Medical Association.

The primary results of each RCT will be submitted for publication following the completion of recruitment and follow-up (anticipated to be January 2022 for the step-down RCT and January 2023 for the step-up RCT). The NIHR report will be submitted for publication in January 2023. We would anticipate publication of results within six months of these submission dates.

Evidence to inform future clinical guidelines arising out of the research will be published and disseminated to professional societies concerned with the care of children presenting with acute illness, including PICS and the Royal College of Paediatrics and Child Health. Presentation slides will be prepared for use by the study team or others in disseminating the research findings.

The results of the study will be disseminated to patients and their families, facilitated by the co-applicants, members of the research team who have links with PICS and the NIHR CSG, and via Family Groups we have liaised with already. The FIRST-ABC research team has worked closely with patient representatives throughout the study conduct. Two parents of children who received breathing support are co-applicants on FIRST-ABC grant application and have actively contributed to the study design and procedures, including the use of deferred consent as well as the development of study documents (for example, Patient Information Sheets). The Trial Management Group and the Trial Steering Committee include Patient and Public Involvement representatives as members. These collaborations will continue to ensure dissemination of the results to patients and the public.

A study website and posts on ICNARC’s social media accounts (ICNARC has more than 7,000 followers on Twitter) will also be utilised to actively publicise progress with the research and disseminate our findings.

In addition to disseminating the results to patients and their families - this dissemination plan will ensure that the results of FIRST-ABC are fed back to those delivering and organising care (e.g. nurses, doctors, managers) in the NHS (and across the world), allowing for any learning from FIRST-ABC to influence clinical practice for the benefit of critically ill patients.

All data presented/reported will be aggregated at a national level, with small numbers suppressed in line with HES analysis guide. It will not be possible to identify any individual participating patient in any reports, articles or presentations.

The FIRST-ABC protocol has been published in British Medical Journal (BMJ) Open and the statistical analysis plan has been published in Trials (please see the references below).

Richards-Belle A, et al. FIRST-line support for assistance in breathing in children (FIRST-ABC): a master protocol of two randomised trials to evaluate the non-inferiority of high-flow nasal cannula (HFNC) versus continuous positive airway pressure (CPAP) for non-invasive respiratory support in paediatric critical care. BMJ Open. 2020 Aug 4;10(8):e038002. doi: 10.1136/bmjopen-2020-038002.

Orzechowska, I., Sadique, M.Z., Thomas, K. et al. First-line support for assistance in breathing in children: statistical and health economic analysis plan for the FIRST-ABC trial. Trials 21, 903 (2020). https://doi.org/10.1186/s13063-020-04818-w

Nearly three-quarters of the 20,000 critically ill children admitted to UK paediatric intensive care units (PICUs) each year receive some form of respiratory support (invasive and/or non-invasive), making it the most common treatment provided in PICU. Both HFNC and CPAP are routinely used in NHS practice, but there is genuine uncertainty as to whether HFNC or CPAP should be used as the first line mode of non-invasive respiratory support in critically ill children in paediatric critical care units. It is not yet known whether HFNC is clinically and cost-effective, as compared to CPAP (this is what FIRST-ABC will tell us) but - it is anticipated that if either mode of non-invasive respiratory support is found to be clinically and cost-effective, that implementation of these outputs into national and international clinical guidelines and subsequently into the NHS will lead to improvements in the delivery of respiratory support in paediatric critical care unit, ensuring first line use of non-invasive respiratory support is informed by robust and high quality randomised evidence. PICU beds are a highly valuable NHS resource, demand for which regularly outstrips supply, leading to delays in the provision of critical care to sick children.

It is anticipated that if HFNC is found to be clinically and cost-effective, implementation of these outputs into clinical guidelines and subsequently into the NHS will help accelerate its adoption in non-critical care areas, improve patient flow into and out of the PICU and promote more efficient utilisation of PICU beds. On the other hand, if HFNC was shown to be less effective and associated with longer PICU stay, this finding could also have profound implications for the optimal use of scarce NHS resources. The FIRST-ABC health economic evaluation will provide information on the cost-effectiveness of HFNC versus CPAP which will be used by policy makers, managers and clinicians in the NHS. All of the above will help to reduce the burden on patients, their carers and the NHS. This will need an effective implementation strategy.

Active and wide dissemination of the results of FIRST-ABC will be an important part of this strategy and will begin upon publication of the primary trial results (see ‘Specific outputs expected, including target date’ for further details). A number of approaches have been identified, including: involving stakeholders; providing evidence in an integrated and graded way; taking account of the context and identifying the elements relevant to decision making, e.g. benefits, harms and costs; making recommendations as specific as possible; and using a multifaceted approach. Dissemination will be lead by the chief investigator working closely with ICNARC CTU and supported by other members of the trial management group. Dissemination of the step-down RCT results will commence in the second half of the 2021. Dissemination of the step-up RCT results will commence in early 2023.

In line with ICNARC’s legitimate interest of processing data for research and statistical purposes - the public (e.g. future critically ill paediatric patients) and the NHS will receive the benefits of the processing. Results of FIRST-ABC will help to guide paediatric critical care clinicians and the wider NHS on the clinical and cost-effectiveness of HFNC versus CPAP.

The FIRST-ABC study database is now locked and primary dissemination of the study results completed. The Final Report to the National Institute for Health Research (NIHR) Health Technology Assessment Programme has been submitted and Editor’s and Peer Reviewer’s comments are currently being reviewed. Yielded benefits to date have included:

Publication of the clinical results for the FIRST-ABC step-up trial in the Journal of the American Medical Association (JAMA)

Publication of the clinical results for the FIRST-ABC step-down trial in the Journal of the American Medical Association (JAMA)

The Trial Protocol and Statistical and Health Economic Analysis Plan have been published for public benefit.

Presentations at national and international conferences have been given to present the research and preliminary results.

Previously there was little research to guide clinical practice of breathing support in acutely ill children, as well as no randomised trials to compare modes of non-invasive respirator support following extubation of critically ill children.

Results from the FIRST-ABC step-up trial reflect that starting an acutely ill child on non-invasive respiratory support with high flow nasal cannula (HFNC) is a reasonable first-line option. Results from the FIRST-ABC step-down trial indicate that post-extubation, starting continuous positive airway pressure (CPAP) is a reasonable non-invasive respiratory option, especially in children under the age of one. For children older than one, HFNC or CPAP are both suitable first-line options.