NAAASP Data Linkage

HES Critical Care, Admitted Patient Care, Accident and Emergency and Outpatient data linked to Civil Registration data was supplied to the University of Leicester by the Health and Social Care Information Centre (which has since become NHS Digital) for the purpose of a research study referred to as NHS Abdominal Aortic Aneurysm Screening Programme (NAAASP) .

This Data Sharing Agreement permits the retention of the data for an interim period but no other processing of the data is permitted.

The University of Leicester is currently seeking publication of the research outcomes and wishes to retain data for the purposes of responding to any requests by journals to revise analyses. No further data will be required under this data sharing agreement.

Permission is to retain the data for the interim period agreed. Should the University of Leicester wish to process data or request further data in the future they will submit an application. This application must meet all the approved Standards.

The following information provides background information on the purpose of the original study. No new data will be released under this version of the agreement, and this agreement allows the applicant to hold and not otherwise process any further data that has already been disseminated.

BACKGROUND

This application is to continue to hold and process data held for the Abdominal Aortic Aneurysm Screening Programme (NAAASP). This processing is justified under Article 9(2)(j) (medical research). The risk of public harm and moral/ethical issues as a result of this dissemination is minimal. Public Health England (PHE) runs the NHS AAA Screening Programme (NAAASP) and is the data controller for the NAAASP data. The NAAASP is part of PHE. The clinical system commissioned and used by the NAAASP to deliver screening,

Community screening for Abdominal Aortic Aneurysm (AAA) by ultrasound has been proven to reduce AAA related deaths and has recently been adopted by the NHS with national coverage established in 2013 and from this year onwards, over 300,000 men will be screened for AAA every year with approximately 4000 AAA detected.

Community screening for AAA in England is carried out by the NHS AAA Screening Programme (NAAASP), part of Public Health England (PHE). NAAASP invites all men for AAA screening in the year of their 65th birthday. Screening is carried out by ultrasound and is both clinically effective and cost effective. NAAASP records the infra-renal aortic diameter for all men who attend for screening. Men found to have an AAA (aortic diameter >30mm) are either entered into a surveillance programme that is also run by NAAASP (AAA 30mm to 54mm) or referred to a vascular surgeon for consideration of surgical repair (AAA >54mm).

The incidence of AAA is falling in western populations and this raises the question of whether AAA screening will remain effective in the long-term. In addition, the NHS AAA Screening Programme (NAAASP), who has become part of Public Health England (PHE), will detect a large number of patients with small AAA that will require regular surveillance imaging.

In order to ensure cost-efficiency it is important to establish what happens to all men invited for screening over time and determine if the programme could be optimised. The aim of this research is to use available electronic health and mortality records linked to AAA screening outcomes to determine where the opportunities for programme optimisation lie.

The University of Leicester proposed to determine the outcomes of men being invited for screening by the NAAASP and investigate clinical factors associated with outcomes by linking screening outcome data for men invited for AAA screening by NAAASP with multiple years of Hospital Episode Statistics (HES) data via NHS Digital (formerly known as the Health and Social Care Information Centre (HSCIC)).

This work is the initial project in this programme of work. This project may lead on to several other important future projects.

This project has no relation to any other current work underway (but see outputs for work resulting from this project).

All men in England are invited for AAA screening in the year of their 65th birthday. The data subjects are all men invited for AAA screening in the 2013/14 screening year.

Public Health England (PHE) runs the NHS AAA Screening Programme (NAAASP) and is the data controller for the NAAASP data. The NAAASP is part of PHE. The clinical system commissioned and used by the NAAASP to deliver screening,

The University of Leicester received a dataset from NAAASP detailing the outcomes of screening. NAAASP sent NHS Digital a dataset comprising a list of NHS numbers and the study identifiers. NHS Digital used this dataset to identify the HES linked mortality records for the men in the dataset and provided this data to the University of Leicester with only the study identifiers, and no other identifiers.. The University of Leicester linked the NAAASP data and the HES linked mortality data and perform analysis.

The outcomes of patients attending the NAAASP are partially unknown. Patients with AAA are followed up by NAAASP through AAA surveillance and the outcomes of patients referred for surgery are recorded. The cause of death in patients with AAA who die whilst under surveillance is not automatically made available to NAAASP. In addition, those screened and found not to have AAA are discharged from NAAASP follow-up and some patients do not attend for screening.

There is some evidence that patients with a normal aortic diameter at age 65 may develop an AAA later in life and therefore be at risk of AAA related death. Also, NAAASP utilises a technique for the assessment of aortic diameter that results in a smaller measurement when compared to other methods and discharges patients if their aorta is below a 3.0cm threshold. This technique may therefore result in some patients being discharged by NAAASP who may be entered into surveillance in other screening programmes. It is not known whether this puts discharged patients at risk of aortic rupture.

Public Health England (PHE) runs the NHS AAA Screening Programme (NAAASP) and is the data controller for the NAAASP data. The NAAASP is part of PHE. The clinical system commissioned and used by the NAAASP to deliver screening, Screening Management and Referrals Tracking (SMaRT), is provided by Northgate Public Services Ltd (NPS). NPS provide this service for all NAAASP screening units in England.

In this research, under instruction from PHE, NPS uploaded confidential identifiers and NHS numbers for each man invited for screening in the 2013/14 NAAASP cohort year. NPS sent outcome data for AAA screening to the University of Leicester without any identifiable data, records only contained the confidential identifiers assigned by NPS. NPS did not receive any data back from NHS Digital or the University of Leicester.

The University of Leicester were approached by NAAASP at PHE to set up a process to find out what happens to men invited for AAA screening by using data linkage. The University of Leicester proposed to link all patients invited for screening by NAAASP in 2013/14 with NHS Digital to obtain HES data as outcomes, with yearly updates. This application seeks to merely retain the data obtained from the 2013/14 cohort.

The University of Leicester and Public Health England are joint data controllers. The data processor will be the University of Leicester.

This project is funded by the University Hospitals of Leicester Charity who have no role other than the provision of funding. The project has been approved by the NHS AAA Screening Programme.

All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract ie: employees, agents and contractors of the Data Recipient who may have access to that data).

No new or further data will be provided under this version of the agreement. A short term extension is in place as a pragmatic approach to enable legal retention of already disseminated data. This agreement allows retention of data, but not permission to otherwise process it.

Permission is to retain the data for the interim period agreed. Should the University of Leicester wish to process data or request further data in the future they will submit an application. This application must meet all the approved Standards.

Below is information provided by the applicant (The University of Leicester) on background to the study for details of previous disseminations, As stated above - this version of the agreement permits retention of data only - permission is not granted to otherwise process the data at this stage until relevant documentation can be provided.

Public Health England (PHE) runs the NHS AAA Screening Programme (NAAASP) and is the data controller for the NAAASP data. The NAAASP is part of PHE. The clinical system commissioned and used by the NAAASP to deliver screening, Screening Management and referrals Tracking (SMaRT), is provided by Northgate Public Services Ltd (NPS). NPS provide this service for all NAAASP screening units in England. In this research, under instruction from PHE, NPS uploaded confidential identifiers and NHS numbers for each man invited for screening in the 2013/14 NAAASP cohort year. NPS sent outcome data for AAA screening to the University of Leicester without any identifiable data, records only contained the confidential identifiers assigned by NPS. NPS did not receive any data back from NHS Digital or the University of Leicester.

NAAASP identified all men invited for screening in the 2013/14 English screening cohort and all men with small AAA already under NAAASP surveillance. NAAASP held and continues to hold this personal information for these men for the purposes of their clinical care.

NAAASP provided individual screening outcome data for the cohort to the University of Leicester. This data will contain a study ID for each individual. No personal data will be transferred to the University of Leicester.

NAAASP provided the NHS numbers of these men to the NHS Digital, together with a study ID.

NHS Digital linked the patients identified by NAAASP with HES linked mortality data using the NHS numbers and provided this linked data to the research team at the University of Leicester, using the same study ID as those used by NAAASP to transfer data to the University of Leicester.

NHS Digital then supplied the University of Leicester with HES/HES linked mortality data stripped of identifiers other than the study ID supplied by NAAASP.

The University of Leicester received data from both NAAASP and NHS Digital. This data will be linked using the study ID and analysed.

The University of Leicester provided NAAASP with annual reports based upon the data, the content of which was determined by NAAASP but will primarily consist of all-cause and aneurysm-specific mortality and aneurysm-related morbidity.

The University of Leicester analysed the data for the purposes of producing research papers focused on the description of mid- to long-term outcomes of contemporary AAA screening. The University of Leicester is currently seeking publication of the research outcomes and wishes to retain data for the purposes of responding to any requests by journals to revise analyses. No further data will be required under this data sharing agreement.

The data will be exclusively stored and processed at University of Leicester and not shared with any third parties except in the form of aggregated outputs with small numbers suppressed in line with the HES Analysis Guide.

The data received from NHS Digital will not be linked with any other data not mentioned in this agreement.

The above data processing included the following specific data flows:

Flow of data into NHS Digital:

Data flowed from the NHS AAA Screening Programme to NHS Digital. This was a list of NHS numbers for the men invited for AAA screening in 2013/14 together with an anonymous study identifier.

Flow of data out of NHS Digital:

NHS Digital flowed pseudonymised health and mortality data to the University of Leicester. This included the study ID provided by the NHS AAA Screening Programme but no other identifiable data.

Subsequent flows of data:

The NHS AAA Screening Programme flowed pseudonymised AAA screening records (health data) to the University of Leicester. These were identified by the study ID only.

Organisation processing data:

All data will be processed by the University of Leicester after data flows are completed.

The data received at the University of Leicester from both NHS Digital and the NHS AAA Screening Programme will be linked using the anonymous study IDs. The combined dataset will then be processed into a research ready dataset for analysis (e.g. coded HES data processed into study outcomes such as ‘cardiovascular admission’).

Data from NHS Digital and the NHS AAA Screening Programme will be linked.

The risk of re-identification will be mitigated through staff training, restricted data access and information security procedures/policies. The dataset will not be used for other purposes or shared with any other parties not mentioned in this agreement.

The dataset will not be matched to any publicly available data.

There will be no requirement or attempt to re-identify individuals

This Data Sharing Agreement permits the retention of the data for an interim period but no other processing of the data is permitted. No further outputs of the data are permitted to be created under this version of the agreement. Permission is to retain the data for the interim period agreed.

The information below provides background on to what has already been produced. Should the University of Leicester wish to process data or request further data in the future they will submit an application. This application must meet all the approved Standards.

The University of Leicester will use this data for the research purposes specified in this agreement and to feedback to NAAASP outcomes for service evaluation when requested. Only pseudonymised data will be supplied to the University of Leicester. Due to the long-term nature of AAA related outcomes after screening the University of Leicester expect that the main research outputs of interest to NAAASP (AAA related mortality and morbidity) will not occur for at least 5 years and will continue to be produced at such intervals for at least 15 years. The service evaluation aspects of this work will be produced on a yearly basis. The University of Leicester will produce a report for the NHS AAA Screening Programme based on the linked cohort as requested. The University of Leicester will produce peer reviewed research publications from the data.

NAAASP reports will be sent directly to NAAASP. These will consist of aggregated data. NAAASP may include summary data in their publicly available national programme reports. The University of Leicester’s data table suppression rules will be adhered to. These suppression rules are aligned with the NHS Digital HES analysis guide.

Research publications will be open-access and available to the public. Research presentations may consist of verbal presentations, posters and published abstracts. Only aggregated data will be included in publications. The University of Leicester will again ensure that the University’s table suppression rules are adhered to in the preparation of these publications.

Dissemination to clinical stakeholders will be via the NAAASP reports and the links that the NAAASP have with other UK AAA screening programmes, the National Screening Committee and Public Health England.

Academic beneficiaries will be reached through publications and presentations.

Short-term (up to 5 years after screening):

Cardiovascular risk in men screened for AAA, including the assessment of aortic diameter as a risk for cardiovascular events.

Cardiovascular risk in men according to attendance for screening

Medium-term (5-10 years after screening):

AAA related mortality and morbidity in men invited for AAA screening

Long-term (10 years+ after screening):

AAA related mortality and morbidity in men invited for AAA screening

AAA related mortality and morbidity in men close to diagnostic thresholds for AAA and non-attenders

One publication relating to cardiovascular risk in men screened for AAA has been prepared and will be submitted between June 2019 and August 2019.

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide

HES Critical Care, Admitted Patient Care, Accident and Emergency and Outpatient data linked to Civil Registration data was supplied to the University of Leicester by the Health and Social Care Information Centre (which has since become NHS Digital) for the purpose of a research study referred to as NHS Abdominal Aortic Aneurysm Screening Programme (NAAASP) .

This Data Sharing Agreement permits the retention of the data for an interim period but no other processing of the data is permitted.

Should the University of Leicester wish to process data or request further data in the future they will submit an application. This application must meet all the approved Standards.

The following information provides background information on the purpose of the original study. No new data will be released under this version of the agreement, and this agreement allows the applicant to hold and not otherwise process any further data that has already been disseminated.

In the research proposed the University of Leicester wish to determine whether there are opportunities to improve the health of men attending for AAA screening beyond simply the detection and treatment of AAA and determine if the current screening policy prevents AAA related mortality and morbidity over time. Since NAAASP has already been set up and is measuring aortic diameters in all men attending for screening, if the University of Leicester can identify those men at high risk of cardiovascular events and flag these men for the institution of secondary prevention in primary care, the University of Leicester can add significant value to the process of AAA screening. Secondarily, the University of Leicester wish to identify whether screening non-attenders are at high-risk or low-risk of AAA-related or cardiovascular events to determine whether additional effort in re-inviting these non-attenders would be worthwhile or not.

This dissemination is important since prevention of cardiovascular events (including AAA rupture/mortality) is likely to represent a signification cost saving for the NHS and clinical benefit for the individual. The demonstration of these benefits will require high quality epidemiological data that this research will provide.

The outputs themselves may contribute to improved clinical care through raising awareness of cardiovascular risk in men with AAA, or screened for AAA. The main route to benefit will be as strong epidemiological evidence to support future clinical research studies.

The benefits of this particular dissemination are in quantifying cardiovascular risk in men invited for AAA screening and the determination of groups at high risk of future cardiovascular disease.

Dissemination will be via publications and presentations (academic beneficiaries) and reports to the NHS AAA Screening Programme (clinical beneficiary).

This research will quantify cardiovascular risk in relation to AAA screening outcomes. This will lead to improved cardiovascular risk management in high risk populations.

The likely change that this will effect is a quality improvement programme or clinical research study to improve cardiovascular risk management as part of AAA screening.

This research is likely to affect around 2-3% of men screened for AAA (approximately 10,000 men per year).

Benefit will be achieved by the NHS. This (improvement in cardiovascular health) will be measured by national mechanisms such as Public Health Observatories. This improvement will take between 10 and 20 years to achieve.

This study is not supporting a PhD or other post-graduate study.

HES Critical Care, Admitted Patient Care, Accident and Emergency and Outpatient data linked to Civil Registration data was supplied to the University of Leicester by the Health and Social Care Information Centre (which has since become NHS Digital) for the purpose of a research study referred to as NHS Abdominal Aortic Aneurysm Screening Programme (NAAASP) .

This Data Sharing Agreement permits the retention of the data for an interim period but no other processing of the data is permitted.

Should the University of Leicester wish to process data or request further data in the future they will submit an application. This application must meet all the approved Standards.

The following information provides background information on the purpose of the original study. No new data will be released under this version of the agreement, and this agreement allows the applicant to hold and not otherwise process any further data that has already been disseminated.

The main output for the short-term aims of this research is a paper that is currently in preparation will be submitted for publication in January 2020. In this paper University of Leicester have demonstrated a strong association of screening attendance and aortic diameter with future cardiovascular risk. Using additional data from other cohorts University of Leicester have shown that this association is independent of cardiovascular risk scoring systems. The main impacts of this paper will be to highlight the potential use of screening data that is currently not clinically used (aortic diameter) and the high cardiovascular risk of men with AAA. The latter impact strengthens evidence for a cardiovascular risk reduction programme to be built into the existing AAA surveillance programme.

This extension will run until 28/08/2020 during which time the applicant will publish finalised research papers and retain data for the purposes of responding to any requests by journals to revise analyses.

Other yielded benefits of this dataset relate to the planning of future research.

i) This work in cardiovascular risk has provided background evidence for a grant application to the NIHR to co-develop a patient-centred cardiovascular risk reduction programme for the NHS AAA Screening Programme. This was submitted in November 2018 and was funded by the NIHR in August 2019.

ii) The NHS AAA Screening Programme also has an interest in a trial of triple cardiovascular screening (AAA + hypertension + peripheral arterial disease) since this has been shown to reduce all-cause mortality in a trial in Denmark. The logical and cost-efficient research design for this is a cluster randomized trail but to design this it is essential to be able to estimate variations in mortality by region. The linked NAAASP-HES-Mortality that this DSA has enabled allows this information to be easily calculated. Without this DSA, planning this future trial would be very difficult. Data received under this DSA has directly supported feasibility work for an NIHR Programme Grant to undertake this clinical trial. This Programme has now been conditionally funded.

Healthcare benefits:

In work to date using this dataset - University of Leicester have demonstrated that there is a strong association between the outcomes of AAA screening and future cardiovascular risk. University of Leicester have shown that men who do not attend for AAA screening, and men who are screened but found to have abnormal aortic diameters are at particular high risk of adverse cardiovascular events. Of particular novelty, the University of Leicester have shown than small aortic diameters are associated with high cardiovascular risk as well as large aortic diameters.

To disseminate these benefits University of Leicester are in the process of writing a scientific paper that will be submitted for publication in January 2020. As well as using the data derived from this data sharing agreement University of Leicester have used additional data from other cohorts and shown that these associations are independent of cardiovascular risk scoring systems. The main impacts of this paper will be to highlight the potential use of screening data that is currently not clinically used (aortic diameter) and the high cardiovascular risk of men with AAA. The latter impact strengthens evidence for a cardiovascular risk reduction programme to be built into the existing AAA surveillance programme.

The data has been presented at national conferences as part of the University of Leicester's dissemination strategy and this has resulted in the following published abstracts:

1. Cardiovascular risk in patients with small abdominal aortic aneurysms. A Saratzis, D Sidloff, M Bown. Lancet 2017;389(S1):89. Spring Meeting on Clinician Scientists in Training, London 2017.

2. A national analysis of short-term AAA screening outcomes: a reduction in operative mortality but not in rupture related mortality. D Sidloff. British Journal of Surgery 2017;104(S5):9. Vascular-Societies Annual Scientific Meeting, Manchester 2016.

3. Social Deprivation, screening attendance and mortality. D Sidloff. British Journal of Surgery 2017;104(S5):6. Vascular-Societies Annual Scientific Meeting, Manchester 2016.

Academic benefits:

Other important benefits of this research have been to future research.

Work in cardiovascular risk has provided background evidence for a grant application to the NIHR to co-develop a patient-centred cardiovascular risk reduction programme for the NHS AAA Screening Programme. This was submitted in November 2018.

The NHS AAA Screening Programme has an interest in extending the scope of the AAA screening programme to a multi-component screening programme. Triple cardiovascular screening for AAA, hypertension and peripheral arterial disease has been shown to reduce all-cause mortality in a trial in Denmark. There is an interest in running a trial in the UK to determine if such multi-component screening would be clinically effective in the UK population. The logical and cost-efficient research design for this is a stepped-wedge cluster randomized trial. To design this it is essential to be able to estimate variations in mortality by region screening region. The linked NAAASP-HES-Civil Registration data received has enabled and will continue to allow these data to be easily calculated.