The clinical and cost-effectiveness of testing for Group B Streptococcus: a cluster randomised trial with economic and acceptability evaluations (GBS3)

The University of Nottingham requires access to NHS England Data for the purpose of medical research which aims to evaluate routine testing of Group B Streptococcus (GBS) colonisation in pregnant women via the GBS3 trial.

Group B Streptococcus (GBS) is a bacterium present in the vagina in 25% of pregnant women. Giving women antibiotics in labour reduces the risk of their babies developing GBS infection. Current UK practice is to offer antibiotics when the baby is at higher risk of developing the infection based on maternal risk factors. This “risk factor” screening is imperfect: some babies born to mothers without risk factors still develop an infection and many women with risk factors do not carry GBS but receive antibiotics unnecessarily. A potential solution is “routine testing” of every pregnant woman and offering antibiotics in labour to those who are carrying GBS. Overall, 71 maternity units (68 of which are in England, 3 of which are in Wales) will be randomly allocated to the “risk factor” or the “routine testing” approach. Those allocated to the “routine testing” approach will be further randomly divided into testing women using a vaginal-rectal swab either a) at 35-37 weeks of pregnancy, or b) in labour, using a rapid test. Women with a positive test result will be offered antibiotics in labour. The number of babies who develop serious infection will be compared. As infections are relatively rare, information on 320,000 women needs to be collected to be able to see a difference between the approaches. Routinely collected data will be used from NHS England systems and other health national data systems to answer the principal research question:

"Does routine testing of women for Group B Streptococcus (GBS) colonisation either in late pregnancy or during labour reduce the occurrence of early-onset neonatal (new-born baby) sepsis (an extreme response to an infection), compared to the current risk factor-based strategy?"

NHS England Data will be used to assess two methods of routine testing which are available against the current policy of administration of antibiotics based on maternal risk factors.

An economic evaluation, which makes use of NHS England Data, will also be performed with the aim of identifying, measuring, and valuing the costs and consequences of alternative testing strategies for GBS in pregnancy or labour, and to synthesise the evidence using metrics amenable to cost-effectiveness-based decision-making.

All eligible women at the same hospital will receive the same treatment. All women giving birth ≥24 weeks’ pregnancy within their site’s study period, regardless of mode of delivery, and all their live born babies will be included in the dataset. Women who experience a stillbirth at labour and delivery will be included as they may have had testing for GBS and GBS may be implicated in the aetiology of their stillbirth.

GBS3 comprises of three studies:

- a randomised controlled trial

- an economic evaluation

- a qualitative exploration of the acceptability of testing

The following NHS England Data will be accessed:

• Maternity Services Data Set (MSDS) v2

Maternity data is required to estimate potential missing information and to perform descriptive analysis of the population, and to extract obstetric information from the mothers’ records and birth information from the children’s records. This information extracted from MSDS will be used to assess the secondary outcomes for this trial for both mothers and babies.

• HES APC and critical care:

Inpatients admissions data will be used to extract details of all birth and delivery admissions in order to recover missing information from the MSDS data and later admissions will be used for any other evidence of sepsis diagnosed or suspected for the mother and children, through diagnosis and procedure codes. These datasets will also be used for the economic analysis.

• HES OP and ECDS:

HES outpatient appointments and accident and emergency attendances at NHS hospitals in England will then be used for the economic assessment.

• Mortality:

Mortality data will be used to assess the 3 outcomes: maternal death, cause of maternal death, baby death before discharge.

The level of the Data will be:

• Identifiable – For mother and child, NHS number, Date of Birth (DOB) and full postcode are required to perform the linkage to non-NHS England Data.

The Data will be minimised as follows:

- Limited to data for a study cohort identified by NHS England as meeting the following criteria:

> Limited to patients from 68 sites within England identified by the Nottingham Clinical Trial Unit (NCTU) team based at the University of Nottingham.

> The trial cohort will consist of mothers giving birth between May 2021 and March 2024 and who delivered or intended to deliver (according to the site chosen by the woman at booking appointment; this may be different to the ‘actual delivery site’, should the woman later change her mind) in the specific maternity units taking part in the GBS3 trial, or who delivered at home but the delivery care was provided by one of GBS3 Site. From these records, data will also be requested on new-born babies.

> Limited to Data between August 2020 and June 2024. For each individual mother, data will only be provided from 9 months before the delivery date and until up to 3.5 years after the delivery date.

> From the data provided by NHS England and the data provider for Wales, GBS3 is hoping to establish a sample size of approx. 320,000 women and their children.

There are a few outcomes that will be studied and are not available from NHS England routine data, therefore they will need to be collected manually during the trial and linked to the final dataset. Using their consent, 100 women per site will have extra data extracted locally for the process outcomes study. These outcomes will be linked to the NHS England routine data to be able to conduct the analysis.

University of Nottingham is the controller as the organisation responsible for ensuring that the Data will only be processed for the purpose described above.

University of London conduct transcription and analysis of qualitative study data from interviews to a small proportion of the cohort to assess acceptability, barriers and facilitators to implementation for GBS3. Interviews are transcribed and anonymised before analysis. The University of London researchers who are responsible for the qualitative study played no part in the design of the clinical trial or economic evaluation, and their organisation will not receive or use routine Data from NHS England.

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.

Currently, at least 10% of all babies born in the UK are treated with injected antibiotics, even though the incidence of early onset GBS sepsis is only 0.57 per 1000 births. This is costly and many babies receive antibiotics that may be unnecessary. As well as increasing resistance against infectious diseases, interference with the new-born baby microbiome (collection of microorganisms) may have long-term health implications. GBS3 is examining the effectiveness of routine GBS screening compared to the current approach, which is limited to high-risk groups. Therefore, this research is carried out for achieving purposes in the area of public health and for the public interest.

National Institute for Health and Care Research (NIHR) fund the trial. The funding is specifically for the trial described. Funding is in place until 31/05/2024.The NIHR Health Technology Assessment has no role in the data collection and analysis and does not have any influence in determining the purpose or means of data processing of the NHS England Data.

University of Oxford is a processor acting under the instructions of University of Nottingham. University of Oxford’s role is limited to the economic evaluation to compare the overall costs of each strategy to ascertain which represents the best value for money for the NHS.

University of Dundee provides Trusted Research Environment (TRE) hosting services to University of Nottingham and will store the data as contracted by University of Nottingham. University of Dundee manage the data flow between each data provider and the TRE and carry out the pseudonymisation of the data. NHS England Data will be stored in the TRE located at the Health Informatics Centre (HIC) at University of Dundee but will be processed by the NCTU team based at the University of Nottingham who will be granted remote and safe access to the TRE. University of Dundee are only responsible for the TRE where the data is stored and pseudonymised and are therefore a processor. They do not have any other role within the trial study.

There are co-investigators who are speciality advisors from the following organisations:

• Birmingham Women’s and Children’s NHS Foundation Trust,

• University Hospital Southampton NHS Trust,

• Group B Strep Support,

• University of Central Lancashire,

• Grace Midwifery Ltd.,

• NHS England

These organisations offer advice but they do not have input into determining the purposes for data processing and their organisations will not receive any data. Decisions regarding project design reside with University of Nottingham only.

Data processing is only carried out by substantive employees of the University of Nottingham, University of Dundee and University of Oxford who have been appropriately trained in data protection and confidentiality.

The University of Nottingham had detailed, sustained, and invaluable input into all aspects of the NIHR grant application and trial protocol, including review of the plain English summary and all public facing documents, from their Patient and Public Involvement (PPI) co-investigators, who represent two leading charities, supporting the research:

•Group B Strep Support (www.gbss.org.uk), the UK’s leading charity working to stop GBS infections in babies. Their Chief Executive was a member of the Department of Health research prioritisation panel and is the co-vice-chair of the Royal College of Obstetricians and Gynaecologists (RCOG)’s Women’s Network

•National Childbirth Trust (www.nct.org.uk), the UK’s leading charity for parents, represented by their Midwife and Parent Educator.

A PPI group has been formed, to provide ongoing advice and support to the trial. The group will meet periodically in person with the CI, Deputy CI, Senior Trial Manager or Trial Manager, the first meeting took place on the 10th June 2019. They plan to also form a dispersed group that will be linked via a closed Facebook Group. The Facebook page will be used as a forum and document-sharing repository. The PPI group’s tasks will include:

• Review and provide feedback on all public facing information, both for the cluster randomised trial and the qualitative study.

• Review and provide feedback on all information provided to healthcare practitioners.

• Help the qualitative researchers develop the interview schedules.

• Engage in workshops to develop training packages for midwives in the testing hospitals

• Develop and potentially participate in video clips, for posting online or showing in antenatal clinic waiting rooms that supplement written information.

• Help the co-investigators respond to queries about testing policies.

• Advise the co-investigators on the interpretation of the results of the qualitative study.

• Create plain language summaries of the results of the project.

• Help with the dissemination of the results.

The helpline of both charities will be provided with structured advice regarding the trial and testing strategies, so that they can directly respond to women’s queries. Both charities will aid in the publicity of the study throughout its duration, via their respective websites, social media channels and newsletters. They will be instrumental in the dissemination of the trial results and will update their own information resources with the results and the implications of the GBS3 trial.

The University of Nottingham will provide a number of codes to NHS England, one for each of the NHS English sites that are taking part in the trial and two dates attached to each code which represent the start and end date of data collection.

NHS England Data will provide the relevant records from the MSDS, HES, mortality and ECDS datasets to the University of Nottingham. The Data will:

• contain directly identifying data items including NHS number, Date of Birth and postcode which are required to link the data at record level with data already held by the University of Nottingham.

The research team will then link the NHS England dataset to the rest of the datasets needed in the project using the three identifiers:

(For England):

1. National Neonatal Research (NNRD),

2. Paediatric Intensive Care Audit Network (PicaNet),

3. BadgerNet Neonatal and Maternity,

4. UK Health Security Agency (UKHSA) laboratory test data.

Linkage of the NHS England Data for England with the above datasets occurs within the TRE. NHS England raw data will never leave the TRE.

Once the linkage has been completed, the data will be pseudonymised by the HIC TRE team. The personal identifiers will be kept, however they will be separated from the main data and securely stored in a separate TRE environment as a backup, only accessible by the approved managers based at Nottingham Clinical Trial Unit (NCTU), University of Nottingham. Data will be stored in electronic format and all files, folders and back up files will be securely destroyed at the end of the retention period with the use of an appropriate software.

Other data analysed but not linked to the NHS England Data include:

For Wales:

• NHS Wales Informatics Service maternity, hospital and mortality data

• Health Protection Wales laboratory test data

The research team will also link a subset of the NHS England to three small datasets that have been created by manually collecting data in the trial sites to conduct three different studies: maternal anaphylaxis (a severe reaction to a trigger such as an allergy) outcome assessment, process outcomes assessment and accuracy of routine data to select cases of new born sepsis assessment. The identifiers will be then removed from the analysis dataset and stored in a safe place within the TRE, not accessible by the analysts. The final pseudonymised dataset will be then created for the trial and economic analyses.

The Data will be stored on the TRE at the University of Dundee.

The analysists/statisticians from University of Nottingham and from the University of Oxford will be able to safely access the Data stored at the HIC TRE to conduct the analysis.

The Controller(s) must confirm and provide evidence upon audit by NHS England that access via any remote device complies with the data security obligations within this DSA and the Data Sharing Framework Contract.

For remote access:

- Remote access will only be from secure locations situated within the territory of use (as further restricted elsewhere within the DSA if so done) stated within this DSA;

- Access controls granting users the minimum level of access required are in place;

- Remote access is only via secure connections (e.g., VPNs or secure protocols) to protect data;

- Multifactor authentication (MFA) is required for remote access;

- Device security, including up-to-date software and operating systems, antivirus software, and enabled firewalls are utilised for the remote access;

- All remote access is undertaken within the scope of the organisation’s DSPT (or other security arrangements as per this DSA) and complies with the organisation’s remote access policy.

The above applies in addition to any condition set out elsewhere within the DSA (e.g. who may carry out processing, and for what purpose

The Data will not leave the UK at any time.

Access is restricted to substantive employees of the University of Nottingham, University of Dundee and the University of Oxford.

All personnel accessing the data have been appropriately trained in data protection and confidentiality.

The outcomes of the subset cohort (100 women per site) detailed in section 5a will be linked to the NHS England routine data to be able to conduct the analysis.

Another study will be conducted using hospital notes extracted from the trial sites (not from NHS England routine data), whereby a central new-born baby adjudication will be conducted on a sample of clinical suspected sepsis cases to assess the robustness and accuracy of the algorithm developed to extract the primary outcome from routine data. Within this study, a list of new-borns identified by a panel of neonatologists, who diagnose and treat new-borns, assessing hospital notes will be linked to the NHS England routine data for validation purposes.

The raw routine data received from each data provider will not reach the Nottingham Clinical Trial Unit (NCTU). However, all research data collected manually and stored at NCTU will be encrypted and sent to Dundee TRE to be ‘merged’ with the routine data to create the full final datasets.

Identifiers will be stored in a separate protected folder within the TRE for the length of the Trial. However, at the end of the trial, the final datasets for the analysis and the identifiers will be sent to the NCTU and safely retained for 7 years as per the university policy. Clinical trial data as per the clinical trials regulations should be kept for a minimum of 5 years for an IMP trial. The university policy is 7 years from the date of first publication.

Where data has been pseudonymised, no attempt will be made by any organisation party to this Agreement to re-identify individuals included in the study population.

Access to identifiable data will be restricted only to the senior research fellow, data manager and IT support team to carry out the pseudonymisation and linkage of the datasets within the University of Dundee’s TRE. The statisticians will carry out the final analysis using only pseudonymised datasets.

Researchers from the University of Nottingham and University of Oxford will analyse the Data for the purposes described above.

The expected outputs of the processing will be:

• Reports

• Submissions to peer reviewed journals (such as the Lancet, The new England journal of medicine or the British Medical Journal (BMJ)).

• Presentations

• National and international Conferences

This trial will perform a systematic evaluation of the current and alternative testing strategies, addressing the direct and intended effects and consequences, with the aim to inform and improve medical decision making. The comprehensive project results will be reported in the journal Health Technology Assessment that publishes findings exclusively from NIHR-funded research on the effectiveness, costs and broader impact of health technologies for those who use, manage and provide care in the NHS.

The outputs will not contain NHS England data and will only contain aggregated information with small numbers suppressed as appropriate in line with the relevant disclosure rules for the dataset(s) from which the information was derived.

Any result from the trial analysis, such as risk of new-born sepsis, or from the economic analysis such as increased costs as a result of alternative testing during pregnancy or delivery, will be reported at the level of the randomised allocation and not at the level of the maternity unit. Moreover, appropriate disclosure rules will be followed for each dataset used in the analysis in any publication of the results.

The outputs will be communicated to relevant recipients through the following dissemination channels:

• Reports

• Journals

• Presentations

• Social media

• Dissemination to maternity service users via charity and public organisation e.g., National Childbirth Trust

• Press/media engagement

The individual component studies will be published together or individually in high-impact peer reviewed journals and presented at medical and midwifery conferences locally, nationally, and internationally to disseminate the finding to research, scientific and medical community.

The University of Nottingham will actively engage with the National Screening Committee, UK Health Security Agency (UKHSA) and its equivalents in other devolved countries, national institute for health and care excellence (NICE), the Royal College of Midwives, Royal College of Paediatrics and Child Health, and the Royal College of Obstetricians and Gynaecologists (RCOG) to appraise them of progress and the outcome of the trial. The high-profile nature of this trial will mean the results will be highly anticipated and updates to existing recommendations and policies timetabled to incorporate the results of GBS3.

To reach the relevant groups that could benefit from the results of this study, presentations will be organised to different audiences. The University of Nottingham will be targeting the clinical/health planning sector via conferences that have audiences in obstetrics, midwifery, neonatology, microbiology, public health, health economics and health sociology as these are appropriate interest groups for the specific findings of this project. Results will be presented in national and international conferences, seminars, and press/media engagement. GBSS, the UK’s leading charity working to stop GBS infections in babies, and the NCTU will be invaluable in educating and updating the women and their families throughout the trial and have an extensive network of social and mainstream media connections and parent education classes that will ensure widespread dissemination of the study as it is ongoing, and its results.

The data collection will run from the opening of the first site up to the end of the following 3.5 years followed by the analysis that will be concluded by the end of the project, with the publication of the results anticipated for the end of 2025.

It is envisaged that the results of this trial will provide the UK National Screening Committee (NSC) with the evidence required to decide whether routine GBS testing should be implemented in the UK, addressing criteria 11 and 14 of the NSC’s screening requirements – listed here https://legacyscreening.phe.org.uk/screening-recommendations.php. The results are therefore expected to lead directly to policy decisions around routine testing for the detection of Group B Streptococcus, that will see more immediate patient benefit by identifying high-risk women and supporting the optimization of antenatal care to prevent and reduce the onset of sepsis in new-borns. Should GBS screening prove to be clinically and cost effective, the results will also provide useful data on the acceptability, coverage, cost and impact on health care services of antenatal enriched culture versus intrapartum rapid testing within a UK setting. This will be the first adequately powered RCT of GBS screening in the world and its results may well impact global practice in those countries who have adopted GBS screening without such evidence for its effectiveness.

Further, GBSS and the National Childbirth Trust will use their extensive network of social and mainstream media connections to disseminate these results, in order to increase acceptability and coverage, extremely important for the implementation of new policies to improve the healthcare service, and therefore in the public interest.

All pregnant women in the UK are expected to benefit by receiving an optimal management strategy that provides appropriate, targeted and only necessary treatment for GBS infections. Only women at high risk of passing the infection to their new-born will be treated to decrease the risk for their baby while women who resulted negative to the GBS test will not need to receive antibiotics reducing their risk of intrapartum anaphylaxis due to the treatment. That in turn is also expected to benefit their offspring who will be at a decreased risk of developing early and late onset infections and will not receive unnecessary treatment if their mothers were not at high risk of GBS.

Results from this trial are expected to inform the decision as to whether to change the current clinical practice in the UK with consequent benefit for the whole population. The results of this trial will provide the UK NSC with the evidence required to decide whether routine GBS testing should be implemented in the UK.

The NSC will schedule their next review of their policy to include the results of this trial, available in 2024.

Yielded Benefits is not a requirement for new applications.