This Report has been produced by the Honest Broker Service, part of the HSC Business Services Organisation.
All queries can be forwarded to: HonestBrokerService@hscni.net
Further background information can be found here.
Previous year’s Annual Reports can be found here.
© Crown Copyright1. Foreword
This has been another highly successful year for the Honest Broker Service (HBS) and the portfolio of research projects supported by the service continues to grow with 12 new research projects approved during the financial year.
Remote access through the Secure e-Research Platform is continuing to allow efficient access with 66 researchers working across 47 active projects.
Of particular note this year saw the first secure linkage of data from the General Practice Intelligence Platform, via the Honest Broker Service, to a range of secondary care data to facilitate a four-nation study on respiratory illness looking at prevalence of Chronic Obstructive Pulmonary Disorder (COPD) and Asthma across the United Kingdom.
In addition to this the Honest Broker Service team has supported the pseudonymisation of Patient Level Costing data which is an important HSC project lead by the Department of Health.
The Data Warehouse and HBS team have also been involved in data provision to aid the establishment of a new Early Life Cohort Study known as Generation New Era. This work fell out of scope for normal HBS projects however support from BSO was commissioned by the Department of Health and the HBS team worked closely with the HSC Trusts and the research team in establishing information governance arrangements to allow the sharing of data. The HBS team and BSO Family Practitioner Services colleagues then played a key role in the sampling and managing the Opt out process prior to sharing data with the fieldwork team.
The HBS received a small grant from Health Data Research UK in 2023/24 to assist with improving transparency of health data access processors for researchers and the public and this Annual Report report has been produced in a new accessible format as a result of this, along with significant improvements to the public facing content on the HBS website.
Looking ahead the HBS have been reviewing the Terms of Reference for
the Honest Broker Service Governance Board this year and in 2024/25 the
board will be re-convened as the HSC Data Access Committee, through
which we will pilot a new scrutiny process for reviewing and approving
access requests.
This process along with the new Terms of Reference will be published on
the HBS website in 2024/25.
It is pleasing to note the continued investment in the service and excellent customer satisfaction results.
Dr Aaron Peace
Chair of HBGB
Consultant Cardiologist WHSCT
Chief Executive of the Clinical Translational Research and Innovation
Centre (C-TRIC)
2. Introduction
Background
Patient and service user data is critical for improving Health and Social Care in Northern Ireland (HSC) services, whether for individual care or for improving population health, planning, innovation and research.
However, concerns about privacy are significant when we talk about using our personal medical data and the health service has to have robust legal, privacy and security controls in place at all times.
The layers of governance processes as well as the general complexity of the health service systems and data can be intimidating for researchers or analysts wishing to access data.
The Honest Broker Service (HBS) was established in 2014, to allow non-identifiable HSC data to be shared or accessed safely in order to maximise the uses and benefits which can be gained from it to the health service and population of Northern Ireland.
The Honest Broker Service acts as the regional Secure Data Environment (SDE) for Health and Social Care in Northern Ireland.
Also commonly referred to as a Trusted Research Environment or Data Safe Haven, a Secure Data Environment is a highly secure computing environment where accredited researchers can access data which has been de-personalised to protect confidentiality.
The HBS operates under the Five Safes Framework which is a globally recognised set of principles operated by similar secure data research services.
In addition to providing access to data for research, the HBS also provides de-personalised, data to Department of Health (DoH) and HSC organisations to support a range of purposes including clinical audit and service evaluation.
Governance
- A Memorandum of Understanding has been established between all HSC Northern Ireland Organisations which outlines how the HBS operates
- The HSC Business Services Organisation is the responsible organisation for the delivery of the Honest Broker Service, and hosts the HSC Regional Data Warehouse
- The Honest Broker Service Governance Board are responsible for scrutiny of applications and oversight of the service. This Board is representative of the HSC organisations who are signatories to the Memorandum of Understanding. The Terms of Reference for this Board are currently under review and the Board is being re-convened as the HSC Data Access Committee. 2023/24 membership of the Governance Board is detailed in Appendix 1
- The Northern Ireland Trusted Research Environment (NITRE) Strategic Board provides strategic direction, support and investment, developing HSC infrastructure to enable utility of HSC data assets for evaluation and research, within safe guards of best practice. It acts as a commissioner and funder of the HBS.
Further detail on the services provided, processes, data uses and how to apply to access the service are available on the HBS website.
3. Research Activity
Twelve new HBS research projects were approved in 2023/24. In addition to this the members of the HBS team have been involved in supporting recruitment into a new Early Life Cohort Study which is detailed in the HBS Activities in Focus Section. Summaries for each project can be viewed by clicking the links in the table below.
QUB - Queen’s University of Belfast
UU - Ulster University
UoE - University of Edinburgh
As in previous years, research projects approved this year cover a wide range of topics. This word cloud provides a brief summary of the areas covered by all research projects to date. Dementia, Maternity and Prescribing are the 3 most popular areas of research.
COVID-19 remains the 4th most common area of research in the HBS, with a further two COVID-19 related projects approved in 2023/24. These projects aim to investigate excess mortality in Northern Ireland during the pandemic and rates of self harm and suicidal ideation over the last 15 years, including throughout the pandemic.
Two distinct linkage projects were approved in 2023/24. These require datasets external to the HSC to be linked to datasets within the Regional Data Warehouse, and as such require additional data access agreements to be written and signed by the relevant parties before data transfer can proceed.
The other projects approved cover areas such as obesity and anaemia in pregnancy, overactive bladder medication, prescribing of psychotropic medications in dementia patients, inequalities in breast cancer, antibiotic prescribing, bowel cancer screening and gonorrhoea spread.
A list of journal articles published in 2023/24 relating to Honest Broker Service research projects can be found in Appendix 5.
Datasets Used
The vast majority of projects analyse a range of linked datasets held by BSO and the Regional Data Warehouse.
The Primary care datasets held by the BSO Family Practitioner Services Information Unit are shown in green. The Enhanced Prescribing Database (EPD) and Patient Registrations (NHAIS) are the most commonly used datasets, being used by 65 and 61 projects since the inception of HBS respectively. In 2023/24, 8 of the 12 approved projects requested access to EPD data, and 7 requested NHAIS data.
The Regional Data Warehouse hold the Secondary Care Datasets (shown in blue). The Patient Administration System (PAS) is the most commonly requested of these datasets, and has been used by 48 projects to date (6 projects requested this dataset in 2023/24). The Northern Ireland Maternity Service (NIMATS) dataset is the second most commonly requested dataset from the Regional Data Warehouse, being utilised by 20 projects (2 projects requested this dataset in 2023/24). General Register Office (GRO) death data continues to be requested as frequently as PAS data with 48 projects using it to date. CaPPS (Cancer Patient Pathway System), and Outpatient datasets are less commonly requested. COVID-19 related datasets introduced in 2020/21, including Pillar 1 and Pillar 2 COVID-19 testing and Vaccine Management System data continue to be requested regularly.
Secure e-Research Platform
47 Projects operated under remote access via the Secure e-Research Platform (SeRP) during the course of the year, compared to 35 in 2022/23. 66 researchers named as active users compared to 40 in 2022/23. The majority of users (65%) were working on a single project, 23% working on 2 projects, and the remaining 12% working on between 3 and 8 projects. There were only 4 projects active in the Safe Haven.
Projects operating in the Safe Room in Belfast
Due to the success of the remote access platform, Demand demand for the HBS Safe Haven in Franklin Street continued to decrease in 2023/24, with 54% utilisation of available sessions compared to 70% utilisation in 2022/23. In February 2024, the Safe Haven moved from the ground floor of Franklin Street to the 2nd floor to a smaller space containing two workstations. The move resulted in the closure of the Safe Haven for the majority of February, reducing the total days open in 2023/24 to 136, compared to 148 days open in 2022/23. Since reopening after the COVID-19 pandemic, only two workstations per session have been offered to researchers to allow for social distancing, so the move has not resulted in a decrease in availability. Safe Haven usage is predicted to remain low, with the majority of existing and new projects choosing to work via SeRP, however, the Safe Haven remains an essential part of HBS because some of the more sensitive projects utilise data which is not permitted to be uploaded to the remote platform.
Output Requests
708 files were checked by HBS in 2023/24:
- 77 intermediate files
- 220 final files
- 411 SeRP files
This is an increase from 2022/23, which is expected due to the increase in active users, with an additional 26 researchers accessing data via SeRP this year. The proportion of files checked relating to COVID-19 has increased in 2023/24 compared to 2022/23 (54% in 2023/24 compared to 23% in 2022/23).The majority of outputs are submitted via SeRP. July was the busiest month with regards to output checking, due to a large submission of graphs for publication.
50% of the files were checked and returned to the researchers the
same day they were submitted. 34% of the files were returned within one
day of submission and the remainder took two or more days. Current
guidelines state that intermediate Outputs should be cleared within five
working days, and final outputs within 20 days. This target was met in
all cases.
Project Modifications
Project modifications form a significant part of the HBS workload every year. They can include:
- Research team personnel changes
- Addition of new datasets to the existing project data
- Addition of new variables from datasets already approved for use by the project team.
- Other changes to the originally approved project application.
All of these changes require a significant amount of work from the HBS team to implement , and can be vital to ensure a project can meet it’s desired objectives.
During this year, the HBS team received 63 project modification requests, compared to 43 last year. These included 35 requests to extend the project end date to allow analysis to continue beyond the planned end date, 15 requests to add additional team members, and 11 requests to change project data—for additional datasets or variables to be added, or to extend the date range covered by the existing data provided to the team.
Registry Updates
The Honest Broker Service provide vital patient healthcare updates, including prescription data and General Registry Office Mortality Data to a number of organisations (including Queen’s University Belfast, QUB) who maintain important disease registries both in NI, and across the UK as a whole. These include the NI Cancer Registry (NICR) , the NI Cerebral Palsy Register (NICPR), the NI Cohort for the longitudinal study of ageing (NICOLA), and the National Joint Registry (NJR).
More information is available here:
NILS
The Honest Broker Service has a dedicated statistician who works with the Northern Ireland Longitudinal Study (NILS) in the NI Statistics and Research Agency (NISRA), acting as both an advisor and data processor for distinct data linkage projects involving healthcare data.
The NILS is proportionally the largest longitudinal study in the UK covering 28% of the Northern Ireland population and accounting for approximately 50% of Northern Ireland households. The NILS ‘core’ data are drawn from the Health Card Registration data and then linked to data from Census returns (1981, 1991, 2001 & 2011), Vital Events (births, deaths and marriages), Migration and Property data. The result is over 30 years of longitudinal data which is regularly being updated. In addition to this rich resource the HBS supply further Health and Social Care data via Distinct Linkage Projects which undergo an approvals process involving input not only from both NILS and HBS statisticians, but other industry experts and academics. Due to the nature of these projects, they are extremely complex and follow strict procedures ensuring patient confidentiality is protected at all times, and data is pseudonymised prior to being transferred to NILS.
During 2023/24 the Research Support Unit in NISRA have been working on testing of the 2021 Census linkages and NILS 2021 is now available to access, extending the research data coverage to 40 years.
4. Non-Research Activity
Requesting Organisations, Purpose & Themes
There were two new non-research projects approved in 2023/24 (compared to three in 2022/23). These projects all supported the Department of Health, and were composed of one service evaluation project and one business intelligence project. One project focused on patient level costing in the Health Service and the other was to update the NI Health and Social Care Inequalities Monitoring system with up to date prescriptions data (relating to Anxiety, Antihypertensive drugs and Statins).
This brings the total number of non-research projects completed
through the HBS to 65. 31 of these projects supported the DoH, 12 were
for Health Trusts, and nine for the Health and Social Care Board.
Approximately half of projects (32) were for service evaluation, 14 for
performance management, and 11 for Clinical Audit. Topics covered (shown
in this word cloud, right, have focused mostly on areas including
prescribing patterns, hospital admissions and mortality rates.
Datasets Used
The dataset most commonly used by non-research projects in 2023/24 was the Enhanced Prescribing Database (EPD) - this was also the most commonly requested in the previous two years). Considering the Honest Broker Service receive a high proportion of research projects looking at prescribing trends, this is to be expected.
Since the Honest Broker Service began in 2014, EPD has been used for 31 non-research projects and the Patient Administration System (PAS) has been used for 25 non-research projects. The A&E datasets (NIRAES / eEMS from the South & South Eastern Trusts, and Symphony from the Belfast, Northern & Western Trusts) would be the third most commonly requested after these, being utilised by 11 projects in total.
5. Research in Focus
HBS Project 084: HDRUK Inflammation and Immunity Driver Programme
Inflammation-mediated diseases present a global healthcare challenge, causing a significant adverse impact on individuals’ quality of life and a substantial economic strain on healthcare systems.
The HDRUK Inflammation and Immunity Research Driver Programme is a UK-wide research programme designed to explore inflammation and immunity as general underpinning mechanisms, initially focusing on widespread respiratory and allergic diseases. These conditions can be worsened by acute inflammatory episodes due to viral infections and environmental factors such as pollution, tobacco, pollen, weather, drugs, foods, stinging insects etc.
As part of this driver programme, Dr John Busby and his team from Queen’s University Belfast have been working with the Honest Broker Service on a pilot project using GP Intelligence Platform Data (GPIP). A cohort of patients diagnosed with asthma and chronic obstructive pulmonary disease, identified by GPIP data, was linked to other healthcare datasets from the Regional Data Warehouse including inpatient data, accident and emergency attendances, prescription data and laboratory tests.
This project is the first time that GPIP data has been linked as part of an HBS project. The project aims to facilitate future respiratory research within Northern Ireland by investigating key issues such as socioeconomic and geographical disparities in healthcare and investigating variations in care processes and outcomes. Prior to this project, respiratory research in NI was substantially impaired by an inability to accurately identify patients with airways disease. This project is allowing work in Northern Ireland to proceed alongside projects in England, Scotland and Wales.
Early Life Cohort - Feasibility Study
In September 2022, the Department of Health commissioned the Business Services Organisation to provide the Northern Ireland sample for the Early Life Cohort Feasibility Study (ELC-FS). The Early Life Cohort (ELC-FS) Feasibility Study is a UK-wide study launched in 2023, collecting data on a new generation of UK-born babies in their first year of life. The study is managed by the UCL Centre for Longitudinal Studies and funded by the Economic and Social Research Council. Data collection for the study will be carried out by IPSOS. The project aims to test the feasibility of sampling and recruitment for an innovative new UK-wide birth cohort study to capture information about babies and their economic and social environments, and their health, wellbeing and development. With the economic and social repercussions of the COVID-19 pandemic, and the impacts of Brexit on our economy and society still unknown, there is a need to understand the challenges facing this generation. The study will provide valuable insight into inequalities in early child development and how they are changing over time, and whether the social and biological factors driving these trajectories are evolving.
The key scientific themes of the study include:
Cognitive, social, and emotional development of infants
Infant-parent relationships, and the early home environment
Infant health, including growth, nutrition and sleep
Mental health of parents and the developing child
Social, environmental and neighbourhood influences on infant and family
Inequality, disadvantage, and social mobility
Genomics, early adversity and biological embedding of stress
The ELC-FS aims to draw a nationally representative sample of babies born in a two-month period from all four UK nations. It aims to recruit over 1,800 families with at least 900 in England and 300 each in Wales, Scotland and Northern Ireland. The HBS team used the National Health Application Infrastructure System (NHAIS) and the Northern Ireland Maternity System (NIMATS) to create sampling frame for the study containing births from June and July 2023. The process involved the extraction of the cohort, matching to the NHAIS system to confirm addresses and any potentially sensitive cases such as bereavement or adoption so these individuals could be removed from the sampling frame before the sample was taken. In England, Wales and Scotland, families were first contacted in July 2023 and fieldwork will finish in May 2024. In Northern Ireland, families were first contacted in February 2024 and interviews are ongoing, with fieldwork expected to finish in August 2024. This process required close collaboration between the HBS team and colleagues in BSO, the ELC team and IPSOS. The HBS team will now assist the ELC-FS team with their follow-up non-response bias analysis which will be carried out through the research project application route.
6. Non-Research in Focus
PLICS
In early 2023, the HBS were contacted by the Strategic Planning & Performance Group of the Department of Health to carry out a new internal project focussing on Patient Centred Costing (PCC). PCC is a detailed costing methodology which costs at individual activity level and builds a full picture of the cost of patients’ healthcare journeys. It has been used effectively in both NHS England and Wales for the best part of a decade, and is now being implemented in NI.
In April 2023, a third-party supplier installed the costing system in each of the Trusts. This system allows the Trusts to import their own data from the BSO Regional Data Warehouse, and it assigns a cost to each health service interaction (based on national methodologies). The result is a patient level dataset where clinical data has been converted into cost data. It is still record level data containing identifiers and is sensitive. HBS Support for the project was required in the processing and transfer of the patient level data from the Trusts to the pseudonymised format required by SPPG.
A pilot was carried out in Summer 2023, which allowed all parties involved to test the process. It was planned that the Trusts would then submit their first year of data in October, however this was delayed to November.
The Trusts were sending 5 different types of files to HBS for processing:
- Admitted Patient Care (Covers Inpatients, Daycases and Acute Daycare – Source PAS)
- Non-Admitted Patient Care (Covers Outpatients and Ward Attenders – Source -PAS)
- Emergency Care (covers ED and Minor Injury Units – Source(s) Symphony and E-ems)
- Community Care Contacts (Covers Community AHP and Nursing - Various sources e.g. LCID, Paris)
- Care Homes (Covers Residential and Nursing Homes clients)
HBS were required to perform a large amount of processing on these files. This included matching postcode to Super Output Area (SOA), converting dates of birth to age bands, encryption of Health and Care Number and deletion of any other identifiable data columns so that the data was de-identified prior to sending to SPPG, output of data into the desired format for SPPG, and then the final transfer to the NIHAP System for SPPG to then complete their analysis.
The HBS team developed R code which aimed to automate the process as much as possible, however it was still a very labour-intensive task, with 1.09 billion rows of data passing through HBS (original estimates during the project planning phase had expected approximately 206 files, with around 6 million rows!).
The project is now complete, and work is underway to prepare for receipt of the 2023/24 financial year data.
7. HBS Activities in Focus
Health Data Research UK - Transparency Standards Funding Award
In October 2023, the HBS was awarded funding from Health Data Research UK (HDRUK) to as part of their new initiative to improve the transparency of processes for accessing health data for research purposes. The aim of the project is to improve the clarity and accessibility of information about data access processes for both the public and for researchers, in line with the Alliance Transparency Standards:
Standard 1: Open access application form and guidance
Custodians should have an accessible copy of the data access application form and accompanying guidance notes on their website.
Standard 2: Transparent application process and criteria
Custodians should provide clear information on every step of the data access process, including the criteria for granting access and how members of the public are involved in the process.
Standard 3: Clear website navigation
Information about data access processes should be easily discoverable and comprehensive.
Standard 4: Consider target audience
Website should use appropriate language for the audience and should include content specifically developed for members of the public in readily understandable terms.
Standard 5: Regular review of website content
Website content should be periodically reviewed and updated as appropriate.
Standard 6: Transparency of data use and auditing
Custodians should be transparent about how data is used for research.
BSO launched a new WordPress site in October, which prompted a thorough review of HBS public facing content. The new site allows more freedom to edit the layout and accessibility of the content. This funding has allowed the HBS team to dedicate additional time and effort to improve the website content, working with the BSO Communications team. One of the key outputs of the project is the development of a set of explainer videos and graphics to inform prospective researchers about our processes but also to improve information for the general public about how the HBS works. These materials have been produced in-house, in collaboration with the BSO Communications team.
The outputs from the project will be:
Review of other Secure Data Environment websites across UK to look for best practice
Review of Honest Broker Service application form & accompanying guidance
Additional web content explaining the service, including tailored content for different stakeholders
The public and patients
Prospective and current researchers
Health and Social Care/Policy Users
Web page enhancement to make the page mores interactive and engaging
Development of a Frequently Asked Questions section
Production of explainer videos describing the services provided, application process and public benefit
Production of an interactive brochure
This content will be launched in early 2024/25. Alan and Rachel will attend the award showcase in London in May 2024 to present the work carried out by HBS and hear from the other award holders.
Output Checking Retreat
In March, Alan & Alison from the HBS attended the first Trusted Research Environment (TRE) Output Checking Retreat, organised by the Data Research Access and Governance Network (DRAGoN) at UWE Bristol. The team is led by Professor Felix Ritchie, who has vast experience in the area of data management, confidentiality, security and access, statistical disclosure control, privacy and identification. During his time at ONS, Prof. Ritchie developed the “5 Safes of Data Governance Framework” which we, and many other TREs around the world, now use as the framework for data sharing. The HBS team have already completed the Output Checking Training & Statistical Disclosure Control Course offered by the DRAGoN team, so when we were offered the opportunity to attend their newest offering, we knew it would be an interesting few days.
Output checking is just one of the 5 Safes. All TREs do it; it can be costly and resource-intensive, and is a task that the HBS team spend a large amount of our time on; both reviewing the outputs themselves, and then in discussions with researchers about the outputs and any issues we have encountered. We review hundreds of output files each year (These typically include frequency tables, graphs, statistical models and abstracts / journal articles), and it requires a significant amount of staff time and expertise in order to effectively review files and provide feedback to teams where necessary.
The team had organised a 3-day event for TREs to learn about best practice and recent developments, review their organisational capability with professional and peer input, devise development plans, and contribute to a ‘best practice’ guideline for TREs. It was facilitated by experts in data governance, statistics, organisational psychology and operations management. It was a great opportunity to get to discuss our work with others and find areas where we could improve how we operate. We found that although we are a relatively small TRE in comparison to some of those who attended (with a team of just 4 statisticians at present) we are operating at a very high efficiency and already have a lot of the recommended processes in place. We did however identify some areas of improvement, and will be trying to implement these into our work in the coming months.
8. User Feedback Survey
Annual Feedback Surveys 2023/24
HBS runs 2 annual feedback surveys – a Post Application Survey for those who have been through the HBS Application Process during the year, and an Active User Survey for those with a current HBS project (regardless of start date).
The 2023/24 surveys were launched on 21/03/24, and closed on 12/04/24. HBS used Citizenspace to host the surveys for the first time this year (previous iterations have utilised word documents being emailed out to all researchers). Citizenspace is the NI Government online consultation tool widely used across the NICS. An email was sent to all researchers on the launch date which contained the link to each survey as well as a QR code, and 2 further reminder emails were issued a week prior to, and 2 days prior to survey close.
| Survey | 2023/24 | 2022/23 | 2021/22 |
|---|---|---|---|
| Post Application | 4 | 3 | 2 |
| Active User | 9 | 6 | 9 |
Overall response numbers have improved for both surveys since last year.
12 new applications were approved by the HBGB during the year, so 1/3 of research teams submitted feedback on their experience which is roughly in line with previous years (approx. 10 applications approved annually).
We currently have 80 active researchers working on 44 projects (in both the physical Safe Haven & the Remote Access Platform - SeRP); this represents a response rate of 11% of all users.
Post Application Survey Feedback
The feedback received in the Post Application Survey has been extremely positive. All responders reported being Very Satisfied with all aspects of the application process, and all found the application form easy to compete. One user provided an additional comment: “Team were helpful and responsive throughout process”, and another noted “We had excellent assistance from HBS Staff”.
| Question | Very Satisfied / Very Easy | Satisfied / Easy |
|---|---|---|
| How satisfied were you with the HBS/HDR UK website and online guidance materials? | 25 % | 75 % |
| How satisfied were you with the application assistance provided by HBS staff? | 100 % | |
| How easy did you find the application form to complete? | 75 % | 25 % |
| How satisfied were you with the approvals process? | 100 % | |
| How satisfied were you with communication about your application throughout the process? | 100 % | |
| Overall, how satisfied are you with the overall service provided by the Honest Broker Service during the application process? | 100 % |
This is a slight improvement on the previous year – 67% were Very Satisfied with HBS application assistance and overall service provided by the team, and of the 3 users who responded in 2022/23, 1 was dissatisfied with the approvals process, 1 satisfied and 1 very satisfied.
Active User Survey Feedback
Safe Haven
2 of the 9 responders had accessed the Safe Haven during the year. Both were Very Satisfied with the availability and booking process. They also responded to say they were Very Satisfied with the facilities, the software available to them and the output clearance process.
SeRP
Overall satisfaction with SeRP remains high this year.
All responses indicated that they found SeRP Easy or Very Easy to access. We updated our guidance this year to try and address difficulties people were having with this step, and it seems to have helped users; in 2022/23, 3 of 4 responders found this Easy or Very Easy, but 1 reported that they found it difficult. We had received numerous emails relating to difficulties users were having during the log in process, usually due to user error when inputting a passcode followed by password. Although only a small proportion of users responded to the survey, we have noticed fewer emails to the hbs mailbox relating to log in difficulties.
4 of the 9 responders were Very Satisfied with the Availability of desktops on the service, with the remaining 5 being Satisfied – some additional comments noted they had difficulties accessing in the afternoons as this seems to be the busy period, but they usually do manage to log in at some point during the day. We have responded to emails relating to this issue throughout the year (in particular during October/ November 2023) , and did issue reminder emails to all users to correctly log off and exit the platform to ensure that unused desktops remained accessible for those who needed them. This seems to have been less of an issue lately which is positive.
All responders were Satisfied or Very Satisfied with the output clearance process. Only 5 of the 9 responded to the question relating to the handling of technical difficulties by the SeRP team, but all 5 were at least Satisfied by this process. Some of the additional user comments relating to SeRP include:
“I
am so grateful for remote access! It is far more efficient and flexible
for me. Have had a small number of occasions where I have been unable to
access a desktop. It’s frustrating when it happens as analysis time is
precious, but it seemed short term and has not occurred recently.”
“SeRP is such an impressive and valuable
resource. It makes such a big difference to work life not to have to
travel into Franklin Street for analysis purposes (although it was
always a nice place to visit). It’s a really easy system to use and I’m
grateful it is available to researchers. Thanks”
| Overall HBS Satisfaction | 2023/24 | 2022/23 | 2021/22 |
|---|---|---|---|
| Very Satisfied | 6 (67%) | 6 (100%) | 8 (89%) |
| Satisfied | 3 (33%) | 1 (11%) | |
| Total | 9 | 6 | 9 |
Overall, we have received another positive review of the service from our users for the year; 6 of the 9 responders (67%) are Very Satisfied with the service HBS offers, and there were no negative responses to this question. This is similar to last year in which all 6 responders were Very Satisfied. This is great news for us as a team as we constantly strive to provide excellent service to our users, so this positive feedback is welcomed!
Some additional comments from the users include:
“HBS
team super helpful and responsive as always.”
“Staff
always turn outputs around very promptly”
“Generally
quick response and outcome from the service”
“I
think the service offered by HBS is exemplary in terms of advice,
support, timeliness and friendliness. We could not do without this
resource-it is invaluable.”
Appendix 1
Membership of the Honest Broker Governance Board 2023/24
Voting Members
| Role | Name | Job Title |
|---|---|---|
| Chair | Dr Aaron Peace | Consultant Cardiologist/Chief Executive of the Clinical Translational Research and Innovation Centre (C-TRIC) and Clinician |
| Vice-Chair | Vacant | |
| Alternate Vice-Chair | Dr Nicola Armstrong | Senior R&D Programme Manager, HSC R&D Division, PHA |
| Belfast Trust | Alison Murphy | R&D Manager |
| Southern Trust | Dr Peter Sharpe | Associate Medical Director for Research, and Clinician. Nominee: Rachelle Moore. |
| South Eastern Trust | Dr Patrick Donnelly | Clinical Director of Research and Innovation and Clinician |
| Northern Trust | Dr Dave Watkins | Executive Director of Medicine and clinician. Nominee: Neil Martin |
| BSO | Cormac O’Brien | BSO ITS Head of Operations |
| SPPG | Vacant | |
| PHA | Laura Moore | HSC R&D Approvals Manager, HSC R&D Division, PHA |
| PCC | Vacant | |
| DoH | Siobhan Morgan | Principal Statistician. Head of Hospital Information Branch (stood down) |
| Non-HSC Lay Member | Dr Hilary Russell | Retired medical geneticist and Health and Social Care Research Ethics Committee Chair |
Non-voting Members
| Organisation | Name |
|---|---|
| BSO | Alan Harbinson |
| BSO | Martin Mayock |
| BSO | Stephen Gibbons |
| ORECNI | Karen Beattie |
| DoH | Charlene McQuillan |
| NITRE | Frances Burns (Guest) |
Appendix 2
Membership of the Honest Broker Working Group 2023/24
| Organisation | Name |
|---|---|
| BSO | Mark Bradley (Chair, now retired) |
| BSO | Alison Afrifa |
| PHA | Dr Nicola Armstrong |
| BSO | Karen Beattie |
| BSO | Alan Harbinson |
| DoH | Charlene McQuillan |
| BSO | Martin Mayock |
| BSO | Naomi Mill |
| BSO | Cormac O’Brien |
| PHA | Eddie Ritson |
| BSO | Stephen Gibbons |
Appendix 3
HBS Memorandum of Understanding
The Memorandum of Understanding (MOU) is established for the purposes of informing the development of an HBS for Health and Social Care. It has been agreed by:
- Health and Social Care Board (HSCB)
- Business Services Organisation (BSO)
- Belfast Health and Social Care Trust (BHSCT)
- Northern Health and Social Care Trust (NHSCT)
- Southern Health and Social Care Trust (SHSCT)
- South Eastern Health and Social Care Trust (SEHSCT)
- Western Health and Social Care Trust (WHSCT)
- The Department of Health (DoH)
- Public Health Agency (PHA)
- Northern Ireland Ambulance Service Health and Social Care Trust (NIAS)
- Northern Ireland Blood Transfusion Service (NIBTS)
- NI Guardian Ad Litem Agency (NIGALA)
- NI Medical & Dental Training Agency (NIMDTA)
- Northern Ireland Practice & Education Council for Nursing and Midwifery (NIPEC)
- NI Fire & Rescue Service (NIFRS)
- Health & Social Care Regulation and Quality Improvement Authority (RQIA)
- NI Social Care Council (NISCC)
- Patient and Client Council (PCC)
A copy of the Memorandum is available here.
Appendix 4
E085: Obesity and Anaemia: The double burden of malnutrition in pregnancy
Malnutrition is a major global health issue causing significant morbidity and mortality across a range of health systems. Globally, weight related chronic diseases cause 4 million deaths annually, up to 800 million people are undernourished and at least 1 billion experience micronutrient deficiencies1. Obesity is now the most common medical condition in women of reproductive age and there is increasing evidence that obesity has a long lasting and severe influence on the growing fetus as well as long lasting health implications for the mother. Pregnant women with overweight/obesity face increased risk for numerous adverse pregnancy and childbirth outcomes, including gestational diabetes, hypertension, congenital malformations, premature birth, stillbirth, and macrosomia. Maternal overweight/obesity is also associated with long-term risk of obesity, T2D, and coronary heart disease among the offspring. Globally, it is estimated that >33% of women of reproductive age are affected by iron deficiency anaemia. Iron deficiency is the leading cause of anaemia and during pregnancy is associated with adverse health risks for both mother and offspring, such as increased risk for preeclampsia, preterm birth, low birth weight, intrauterine growth restriction, as well as adverse effects on neurodevelopment and cardiometabolic outcomes in the offspring. The coexistence of overweight/obesity and anaemia in pregnant women presents a dual health burden, yet research on the short and longer term risk factors and health consequences of this dual burden of malnutrition is sparse. This research aims to examine the influence of maternal obesity on risk of anaemia.
E086: A mixed methods project exploring the use and risks associated with Overactive Bladder Medication in adult women in NI
Anticholinergic (ACh) medication is frequently used to treat women with overactive bladder (OAB) symptoms. This is a bladder problem that means you have an urgent need to pass water that cannot be deferred and can result in a leak of urine. These drugs can have adverse effects and as we age the potential to experience these effects can increase. ACh medication is frequently used to treat overactive bladder symptoms in women as well as alternative surgical treatments such as intravesical onabotulinum toxin injections, sacroneuromodulation or percutaneous tibial nerve stimulation. There is some evidence that other non-OAB medication can have anticholinergic side effects and have been associated with a greater risk of memory issues and mortality.
The study will have four objectives: (1) Explore how commonly these medications are used and prescribing patterns across Northern Ireland. (2) Link healthcare data already collected within the NHS to examine the use of these medications in adult women in Northern Ireland (NI) and explore whether being on these medications is potentially linked to risk of vascular disease, dementia, or death. This will be done by identifying all women in NI over the age of 18 years who have been on these medications over a twelve-year period from 2010-2022 using the Enhanced Prescribing Database, a centralised record of all medications dispensed in community pharmacies across NI. This will be linked to information from emergency departments, hospital admission and discharge datasets and death data to quantify the risk of vascular disease, dementia diagnosis or death. (3) We will explore the value of four different scales that have been developed to look at anticholinergic effects. (4) And finally assess whether any identified risks may be reversible by comparing those who stop medication during the study period with those with continued use.
E087: Investigation of excess mortality in NI
Statistical authorities have reported that more people have died than expected in European countries since spring 2022. There are many possible explanations for this, including the complex effects of the flu season being absent during 2022. There is a need to investigate the situation to understand whether there is currently an increase in deaths in Northern Ireland. Some reasons that have been proposed are: the longer-term direct consequences of COVID-19 infection; delays seeking healthcare, resulting in late diagnosis of serious diseases, such as cancer, which might then be harder to treat; missed opportunities for preventive care such as medicines to lower cholesterol or blood pressure due to changes in healthcare access and use; and pressures in ambulances, emergency departments and hospitals resulting in delayed urgent care. Disrupted or delayed access to screening, diagnosis, and treatment may have changed the trends in mortality that were previously improving. In this study, we aim to investigate the changes in mortality in Northern Ireland, during, compared to before the pandemic and examine the associations between the death and demographic, clinical, and healthcare-related factors.
E088: Prescribing of psychotropic medications among care home residents with dementia before and during the COVID-19 pandemic: an observational study
Care home residents with dementia have been significantly affected by the COVID-19 pandemic because of high numbers of infection and death within this setting. Non-cognitive symptoms of dementia, such as anxiety, depression, and aggression, are reported to have increased during the pandemic due to the introduction of lockdowns, limitations on visits from family members and friends, and infection control measures. Additionally, the way in which healthcare services and care were provided for care home residents changed significantly, and there has been an increase in the use of telephone and video consultations. Several studies have reported a change in the use of psychotropic medications (medications that work on the central nervous system such as antipsychotics, antidepressants, benzodiazepines, anxiolytics, hypnotics, antiepileptics) used to manage non-cognitive symptoms among care home residents with dementia during the first year of the COVID-19 pandemic. These studies were conducted in other parts of the United Kingdom and the world. Therefore, research is needed to provide information about use of these medications for care home residents with dementia in Northern Ireland (NI).
This study aims to explore the use of psychotropic medications over the COVID-19 pandemic for care home residents with dementia in NI compared to their use during the two years before the pandemic. This information will be taken from a database which holds information on all prescriptions that are dispensed by community pharmacists in NI. We will look at the types of psychotropic medications that were prescribed and we will use information from another linked database to assess if prescribing of these medications was affected by residents’ age, gender, or the type of care home they live in. We hope that the findings from this study will improve the quality of information about prescribing of psychotropic medications for care home residents with dementia.
E089: Inequalities in breast cancer presentation, treatment and outcomes in Northern Ireland
Health inequalities are unfair and avoidable differences in health between different groups within society. Women in Northern Ireland may be at particular risk of inequalities because of the high levels of deprivation, high levels of mental illness and long waiting lists. This study will investigate breast cancer presentation, treatment and outcomes in groups of women who may be at risk of experiencing inequalities. In particular, in this analysis within the Honest Broker Service we will focus upon women with mental health disorders, and women from low socio-economic status backgrounds. This study will use routinely collected data to compare breast cancer presentation, treatment and outcomes by a range of potential inequalities. This analysis will be based upon 15,000 breast cancer patients from the Northern Ireland Cancer Registry diagnosed from 2011 to 2021. The study will determine outcomes including stage at diagnosis, pathway to diagnosis, treatments received and mortality up to 12 years after diagnosis. The study will identify mental health disorders from prescribing data and socioeconomic status from house value.Should this study provide evidence that breast cancer patients from specific groups have worse outcomes, this would allow further research into the cause of this inequality and interventions to be developed to improve these outcomes in future.
E090: Independent Review of Licensing System in NI, including the Surrender Principle
This Independent Review of the NI Liquor Licensing System is commissioned by the Department of Communities in fulfilment of a statutory requirement under Section 23 of the Licensing and Registration of Clubs (Amendment) Act 2021. Alcohol licensing is the statutory mechanism through which the number, location and operating hours of alcohol retail outlets are regulated. Alcohol licensing serves two key purposes: to ensure retail conditions that are both fair to operators and meet the demands of consumers, and to mitigate the health and social risks associated with the oversupply of alcohol and / or irresponsible retail practices. This review will provide a comprehensive stock take of key stakeholder and community views of the current system, review UK and global examples of licensing system design and related evidence, and provide robust estimates of the likely impacts of viable options for reform. This part of the review will explore the relationship between hospitalisations and deaths (specific to alcohol and related to alcohol) and alcohol availability. This review will explore three levels of alcohol availability – on sales premises, off sales premises and both off sales and on sales together. Under the legislation, the review is to be laid before the NI Assembly by August 2024 and the DfC is required to respond to its recommendations within six months.
E091: 5 year follow up - evaluation of the public health impact of the Connswater Community Greenway using administrative data; A natural experiment
The Connswater Community Greenway was a large urban regeneration project involving the development of a new urban greenway in east Belfast which was opened to the public in 2017. Access to green and blue space such as that provided by the greenway is hypothesised to improve the health and mental wellbeing of local residents.Researchers at the Centre for Public Health, Queen’s University Belfast have conducted research since 2010 evaluating the public health impact of the greenway. We have received funding as part of a large UK-wide research consortium to conduct a further investigation of the public health impact of the Connswater Community Greenway (CCG) on local residents five years after it has been opened. This is the first time that a 5-year follow-up study such as this has been undertaken. The proposed project will use administrative data (i.e. data that has been collected for the provision of services) to reduce the burden of local residents completing surveys. The study will link geographical data on proximity to the CCG to primary care registration data, prescription medication data, hospital admission and death data, and birth outcomes data to evaluate the impact of the CCG on a range of health outcomes (such as mental ill-health, asthma, heart disease, diabetes and dementia) over the last decade. Comparisons will be made between those who live in electoral ward areas within 1-mile of the greenway (an approximate 15 minute walk) and those who live further away or in other similar urban areas. The study will compare the difference in outcomes between those who live closest (within 400 metres of the greenway) to those who live further away (within 1200 metres). Finally, the study will compare the outcomes for different groups of the population such as males/females, different age groups and those living in deprived areas.
E092: MUM-PreDiCT2
Multimorbidity is when a person has two or more long-term health problems. Multimorbidity is becoming more common in pregnancy. However, we don’t understand why this is and what the consequences are for the pregnancy, the birth of the baby and the mother and baby’s health in the long-term.
The MuM-PreDiCT UK Consortium is a collaboration between all four UK nations and is investigating the prevalence and determinants of long-term conditions, how conditions cluster together and evolve over time, and how certain clusters affect the outcomes of the pregnancy, and future health and wellbeing of mothers and infants. The group is also investigating prevalence of medications, and combinations of medications (polypharmacy) around the time of pregnancy in women with and without multimorbidity, and will also study the impact that these medications have on outcomes of the pregnancy, and future health and wellbeing of mothers and infants.
This application is an extension to a previous MuM-PreDiCT study (E062) and seeks to expand on the outcomes under investigation. Since the initial application, the UK consortium has completed an extensive selection and prioritisation exercise to identify outcomes of pregnancy that are most important to other researchers, health professionals, parents and children. This involved conducting a series of literature reviews and patient and public engagement activities to identify potential outcomes of interest, followed by a Delphi process to prioritise the outcomes and select a “core set” for any research involving multimorbidity in pregnancy. The proposed application expands the datasets and variables requested to include those identified in the core outcome set.
These studies will give healthcare professionals and parents a much better understanding of having two or more health conditions in pregnancy. Through this enhanced understanding, they will be able to plan and design services that meet the needs of women and their families before, during and after pregnancy.
E093: Analysis of antibiotic prescribing through multi-level modelling: variability and drivers
Antimicrobial resistance (AMR) has become one of the most serious global public health threats. In 2019, AMR contributed to an estimated 1.27 million deaths worldwide. This number is predicted to rise to 10 million people per year by 2050, with a corresponding loss of $60-100 trillion in economic output. Increased AMR not only endangers the efficacy of antibiotics but also leads to infection-related complications, prolonged illness, higher hospitalisation rates, and increased mortality. Concerns about AMR resulted in increased monitoring and evaluation of antibiotic prescribing, with primary care responsible for over 80% of these prescriptions. Demographic, geographic, and socio-economic factors have been shown to have a significant effect on prescribing rates. Furthermore, higher antibiotic prescribing has been reported among patients who have high rates of comorbidity, suggesting the need for stewardship interventions that focus on comorbid patients with high-frequency antibiotic use. Although reducing antibiotic use is central to antimicrobial stewardship, it may have unintended consequences. In particular, there is currently uncertainty with regards to the relationship between antibiotic prescribing levels and complications that can arise after various common infections.
This study aims to i) analyse temporal trends and variations in antibiotic prescribing at GP practice level; ii) investigate the association between antibiotic prescribing and demographic, clinical, geographic, and socio-economic characteristics, and iii) evaluate the association between antibiotic prescribing and infection-related hospitalisations. Investigating antibiotic prescribing can help public health authorities identify trends, outbreaks, and determinants of high antibiotic use, enabling them to take preventive measures.
E094: A comprehensive investigation into self-harm and ideation in Northern Ireland: a quantitative data linkage approach
Self-Harm and suicide ideation (i.e. thoughts about dying by suicide) are two of the most important known risk factors for death by suicide. Increasing suicide rates are a major public health concern and Northern Ireland consistently has the highest rate of suicide in the UK and Ireland. Recent policies to reduce the rates of suicide are now including a focus on reducing self-harm and suicide ideation as these are known precursors to suicide. However, little is known about the causes of self-harm or suicide ideation, how these two factors are related and what impact they have on mortality risk either individually or in combination. Understanding the individual level, household-level, area-level and health related predictors for self-harm, ideation and suicide is of vital public health importance so that intervention services can be targeted accurately.
Northern Ireland is unique in that it has a national Registry of Self-Harm that has collated information on all presentations to emergency departments for self-harm and self-harm / suicide ideation since 2008 (approximately 8,000 per/year) and the research team has secured access to these data from the data custodians in each of the five Health and Social Care Trusts. The proposed project will link these data to primary care registration data, prescription medication data, hospitalisation data, geographical and death data to better understand the antecedents to and outcomes of self-harm and self-harm / suicide ideation, including mortality risk. Trend analysis will explore trends in self-harm and ideation and associated risk over the last 15 years as well as model potential future trends. Lastly the project will attempt to evaluate the self-harm intervention programme (SHIP), making use of a natural experiment design as SHIP was initiated in the Western Trust before being rolled out in the other four Trusts in 2015. The findings from this project will help identify individuals most at risk and target effective interventions to improve outcomes post self-harm / ideation and reduce suicide risk.
E095: Update of NI Bowel Cancer Screening Programme Key Statistics
In Northern Ireland, from 2012 to 2016, there were 5,985 newly diagnosed cases of Colorectal Cancer (CRC), making it the second most common type of cancer found in men and women in Northern Ireland. Screening and removal of early precancerous lesions have been shown to reduce the mortality of CRC.
There has been recent interest for extending the age range of the screening programme to include 50–74-year-olds within Northern Ireland, in line with other UK nation. A health economic model is currently under construction to urgently provide the NI Bowel Cancer Screening Programme (BCSP) with evidence to support the decision to change the service. This data request will provide updated information which can be used within the cost-effectiveness model and give much needed answers as soon as possible.
E096: Data driven modelling, assessment and control of Gonorrhoea spread in NI
In recent years, the number of gonorrhoea cases in Northern Ireland has been growing rapidly. The purpose of our project is to use data-driven modelling to investigate the different factors which contribute to this growth. Recent modelling at QUB uses mathematics to estimate aspects of the disease which are important to the health authorities when making decisions to reduce the prevalence of gonorrhoea. These include but are not limited to, the transmission rate of gonorrhoea or the average amount of time that passes between becoming infected and getting a test for gonorrhoea. If we can lower the transmission rate or get people diagnosed quicker, the amount of people getting gonorrhoea will decrease. Another element of gonorrhoea that the model provides is the basic reproduction number, known as the R-number. If this R-number becomes larger than 1, there is an outbreak of gonorrhoea. The models let health officials investigate how different interventions will impact these factors and allow for the optimal decisions to be made to reduce an outbreak.
As mentioned, this modelling is data-driven. It uses monthly gonorrhoea cases in Northern Ireland to help estimate the contributing factors to gonorrhoea in Northern Ireland. This data request will provide updated information on the monthly cases of gonorrhoea, which in turn will provide updated estimates of these important factors. In addition, it will provide a consistent update on the R-number of gonorrhoea in Northern Ireland. Ensuring this information is relevant enables health authorities to implement the most appropriate interventions at a particular time. Finally, the request will be used to create more detailed models which can provide more key information on managing gonorrhoea, such as antibiotic resistance in gonorrhoea strains or how gonorrhoea affects different people based on sexuality or gender.
Back to Projects Approved 2023/24
Appendix 5
List of Journal Articles Facilitated by HBS, 2023/24
| Year | Topic | Title | Authors | Journal | Link |
|---|---|---|---|---|---|
| 2023 | Child Welfare | The mental health of all children in contact with social services: A population-wide record-linkage study in Northern Ireland. | McKenna, S., O’Reilly, D., & Maguire, A. | Epidemiology and Psychiatric Sciences, 32, E35. | https://pubmed.ncbi.nlm.nih.gov/37190768/ |
| 2023 | COVID-19 | Deaths at home, area-based deprivation and the effect of the Covid-19 pandemic: An analysis of mortality data across four nations. | J Leniz, JM Davies, A E Bone, M Hocaoglu, J Verne, S Barclay, FEM Murtagh, LK Fraser, IJ Higginson, and KE Sleeman | Palliative Medicine 1–6. | https://doi.org/10.1177/02692163231167212 |
| 2023 | Dental | Fiscal and pricing policies related to food and non-alcoholic drinks A review of the evidence. | C O’Neill, J Woodside, F Kee, M Clarke, L Barry, J Cawley, E Doherty, G Crealey, J Duggan. | SafeFood Ireland | https://www.safefood.net/Professional/Research/Research-Reports/fiscal-food-drink |
| 2023 | Mental Health | Severe mental illness and ophthalmic health: A linked administrative data study. | Finola Ferry, Michael Rosato, Gerard Leavey | PLOS ONE, June 7, 2023. | https://doi.org/10.1371/journal.pone.0286860 |
| 2023 | Mental Health | Mortality risk following self-harm in young people: a population cohort study using the Northern Ireland Registry of Self-Harm. | Emma Ross, Dermot O’Reilly, Denise O’Hagan, Aideen Maguire. |
The Journal of Child Psychology & Psychiatry. First published: 16 March 2023 |
https://doi.org/10.1111/jcpp.13784 |
| 2023 | Neonatal | National Vulnerable Newborn Mortality Collaborative Group and Vulnerable Newborn Measurement Core Group. Neonatal mortality risk for vulnerable newborn types in 15 countries using 125.5 million nationwide birth outcome records, 2000-2020 | Suárez-Idueta L, Blencowe H, et al. | BJOG: An International Journal of Obstetrics and Gynaecology | https://doi.org/10.1111/1471-0528.17506 |
| 2023 | Neonatal | National Vulnerable Newborn Prevalence Collaborative Group and Vulnerable Newborn Measurement Core Group. Vulnerable newborn types: Analysis of population-based registries for 165 million births in 23 countries, 2000-2021. | Suárez-Idueta L, Yargawa J, Blencowe H, et al. | BJOG: An International Journal of Obstetrics and Gynaecology | https://doi.org/10.1111/1471-0528.17505 |
| 2023 | Mental Health | Childhood contact with social services and risk of suicide or sudden death in young adulthood: identifying hidden risk in a population-wide cohort study | Sarah McKenna, Dermot O’Reilly, Aideen Maguire | Journal of Epidemiology & Community Health | http://dx.doi.org/10.1136/jech-2023-220975 |
| 2023 | Mental Health, prescribing | Measuring the impact of therapy on medication use: data-linkage study | Julie-Ann Jordan, Adam Elliott, David Mongan and Kevin F. W. Dyer | The British Journal of Psychiatry | https://doi.org/10.1192/bjp.2023.130 |
| 2023 | Suicide, Mental Health, Self Harm | Suicide risk following ED presentation with self‑harm varies by hospital | Siobhan Murphy, Emma Ross, Denise O’Hagan, Aideen Maguire, Dermot O’Reilly | Social Psychiatry and Psychiatric Epidemiology | https://doi.org/10.1007/s00127-023-02561-6 |
| 2024 | Obesity, Maternity | S05-02 Preconception health behaviours among women with obesity in Northern Ireland: an analysis of a national maternity dataset 2011-2021 | Emma H Cassinelli, Michelle C McKinley, Lisa Kent, Kelly-Ann Eastwood, Danielle AJM Schoenaker, Laura McGowan | The International Journal of Obesity | https://www.nature.com/articles/s41366-023-01431-0 |
| 2024 | COVID-19 | Undervaccination and severe COVID-19 outcomes: meta-analysis of national cohort studies in England, Northern Ireland, Scotland, and Wales | The HDR UK COALESCE Consortium | The Lancet | https://doi.org/10.1016/S0140-6736(23)02467-4 |
Please rotate your screen for optimal experience