Outcomes of patients undergoing lower limb vascular procedures in the National Vascular Registry

The purpose of this service evaluation project is to investigate patient outcomes after lower limb vascular surgery in the United Kingdom. Peripheral arterial disease (PAD) affects 20% of the UK population aged 55-75, equating to a total of approximately 850,000 people, and the National Vascular Registry (NVR) includes data on approximately 41,000 patients who underwent lower limb vascular surgery from 2014 to 2016. The most severe manifestation, critical limb ischaemia, has an estimated annual cost to the National Health Service (NHS) of more than £200 million. This study will look specifically at the time 01/01/2014 to 31/12/2016, which covers a total of 40,890 patients who underwent lower limb vascular procedures recorded in the NVR, equivalent to 13,630 procedures per year. Unfortunately, little is known about the longer-term outcomes of this patient group in the 12 months after they leave hospital. Linking NVR data to HES-Civil Registration Mortality data held by NHS Digital would allow study of these longer-term outcomes and investigate any clinical or demographic risk factors for poor outcomes. The published findings would thus enable clinicians and patients to be better informed and facilitate future research on patient risk prediction for these procedures.

The data subjects are 40,890 patients who underwent lower limb vascular surgery procedures recorded in the NVR from 01/01/2014 to 31/12/2016.

The data request to NHS Digital will help achieve the aim identified above as they describe the longer-term patient outcomes after lower limb vascular surgery in the UK. This will provide valuable information for patients and clinicians with the ability to improve the provision of care and treatment.

North Bristol NHS Trust have experience of working with data from the NVR. However, this specific project does not relate to any other piece of completed work.

The proposed project seeks to undertake a descriptive analysis and basic statistical assessment of the linked dataset provided by NVR/NHS Digital. Any further work in the future will be covered by an appropriate application at that time.

The legal basis for processing falls under GDPR Article 6(1)(e) and GDPR Article 9(2)(i) . Processing and dissemination of these data meet the public interest criteria since it will grant clinicians, healthcare organisations and patients a greater understanding of the clinical outcomes patients experience after undergoing lower limb vascular surgery in the United Kingdom. As such, more will be known about current standards and outcomes of clinical practice and clinicians and patients will be better informed prior to surgery.

HES and Civil Registration Mortality data is required because the HES inpatient dataset contains inpatient hospital activity and as such will allow identification of re-admissions and re-interventions in the cohort of interest. The mortality data are necessary to calculate post-operative mortality up to 12 months after the date of procedure. Identifying data are to be provided by NVR to NHS Digital for the purpose of data linkage. Then data will be pseudonymised before it is shared with the project team at North Bristol NHS Trust. As such, no identifiable information will be shared outside of NVR/NHS Digital, as per CAG s251 approval.

The s251 support in place is for non-research purposes (i.e. service evaluation/improvement/audit etc).

The number of years requested matches the date range of the NVR dataset. The geographical spread is nationwide to cover all eligible patients and vascular surgical centres. There are no alternative ways of achieving the purpose, as the necessary outcome data is not available from any other source.

Only data for the 12-month period after each patient's date of surgery is requested. This epoch will be individually defined for each patient, such that the minimum data are required.

The joint Data Controllers are University of Bristol and North Bristol NHS Trust. The Data Processor is North Bristol NHS Trust.

NVR are flowing identifiable patient data to NHS Digital and pseudonymised patient data to University of Bristol but will not be processing NHS Digital data so as such are not a considered a data processor for this purpose.

The project is funded by a Research Fellowship from the Bristol Health Research Charity. The funder has no direct role in the project; played no part in the design of this project; nor will they be involved in the processing or dissemination of the data requested.

Data received from NHS Digital will be pseudonymised at source prior to being sent to North Bristol NHS Trust. Any results from subsequent analyses that are shared publicly will be aggregated.

No new data will be provided by NHS Digital under this Agreement.

Under the previous iteration of this Agreement, identifiable data was provided from NVR to NHS Digital to facilitate data linkage and the provision of pseudonymised data to the study team at North Bristol NHS Trust.

The data flowing into NHS Digital came from National Vascular Registry (commissioned by NHS England and part of the National Clinical Audit Programme managed by HQIP). This included identifying data - specifically the NHS number and date of birth of each patient. This was to facilitate data linkage with the HES and Civil registration Mortality data held by NHS Digital and this was covered by the CAG section 251 approval. The NVR study ID was also provided.

The data flow out of NHS Digital was to North Bristol NHS Trust. Personal identifiers were removed by NHS Digital prior to transfer; the Study_ID remained. Only pseudonymised data was shared with the project team.

There will be no subsequent flows of data.

All data flows are individual patient level.

North Bristol NHS Trust received the pseudonymised linked dataset from NHS Digital and pseudonymised data from NVR via University of Bristol, who are joint data controllers. North Bristol NHS Trust will not hold any identifiers and therefore cannot re-identify individuals. They use the study id to link the data from NHS Digital and NVR. North Bristol NHS Trust perform the descriptive analyses on the linked dataset and prepare outputs for dissemination as outlined in the ‘Outputs’ section (below). These analyses will include an aggregate description of the patient group undergoing these procedures, by producing summaries of demographic characteristics such as average age, male/female and medical comorbidities (descriptive statistics). The Trust will also be able to describe the number of patients having each type of surgery and produce a summary of their postoperative outcomes. For example, the percentage of patients needing to be readmitted to hospital, or the percentage who survive beyond 1 year from their operation. All data will be presented in aggregate form, for example in tables and graphs, with small numbers suppressed.

No attempt will be made by the team at North Bristol NHS Trust to re-identify any individual. The steps to mitigate the risk of re-identification include the removal of all identifiers by NVR/NHS Digital before any data are shared with the project team. Some data fields will be reduced in granularity to reduce risk of re-identification: for example, date of birth will be converted to age; date/time of procedure in NVR will be reduced to Month/Year; date of hospital discharge and/or date of death will be converted to length of stay in days.

Data processing will only be carried out by appropriately trained employees and in accordance with all Organisation information governance and data protection policies. All data access will be done in secure environments with password-protected system access.

All data access will be done in secure environments which are access-controlled by ID badge. System access will be password-protected, as will individual files. There are no plans for remote access.

The aim is to disseminate the results of the project to researchers, scientists, clinicians, and patients involved in vascular surgery and anaesthesia. All data contained in outputs will be presented in aggregate form with small numbers suppressed as per the HES analysis guide. No personal identifiable data will be released or published.

The project is supported by the Bristol Hospitals Research Charity. The Charity will be acknowledged in any outputs but will have no input into the design, analysis, or write-up of the project.

The expected outputs of the data processing are submissions to peer-reviewed journals such as the British Journal of Surgery or the Journal of Vascular Surgery. The published outputs of this work will then be referenced in any future grant applications or proposals to study the subject further.

Additionally, conference presentations are also expected at The Vascular Society Annual Scientific Meeting or European Society for Vascular Surgery Annual Meeting, and a report for the local vascular patient group.

The aim is also to disseminate the results to patients: specifically a presentation will be given to the Bristol Vascular Surgery Patient and Public Focus Group.

The Bristol Vascular Patient and Public Involvement Group has 11 members who have previously undergone vascular surgery. The Group meet twice a year with clinicians from both vascular surgery and vascular anaesthesia. The Group have made significant contributions to both local and national vascular surgery research projects and other initiatives. For example, they are acknowledged as contributors in both the National Vascular Registry 2019 report (https://www.vsqip.org.uk/content/uploads/2019/12/NVR-2019-Annual-Report.pdf) and the Vascular Society Best Practice Clinical Care Pathway for peripheral arterial disease (https://www.vascularsociety.org.uk/_userfiles/pages/files/Resources/PAD%20QIF%20March%202019%20v2.pdf), having informed the infographics used in these documents.

Outputs will be communicated to interested parties using social media, namely Twitter to reach interested clinicians, researchers and patients across a range of disciplines and areas

The aim is to inform the relevant medical professionals, in order to inform their practice going forwards and to make patients aware of the findings, such that they are as well-informed as possible.

These outputs should be achieved within 12 months of data receipt, pending acceptance to relevant journals and conferences. The measurement of dissemination of the published work will be possible through monitoring citations and online discussion, for example using Altmetrics.

The target dates are as follows (from the receipt of data):

Submission for peer-reviewed publication within 12 months

Submission for conference presentation within 12 months

Presentation to Patient Group within 12 months according to scheduled meetings.

The expected measurable benefits are through the increase in knowledge which will facilitate better provision of health care to this patient population. Patients will be able to make better informed decisions about their choice of treatment.

This project will help address the current uncertainty and variation in clinical practice, due in part to an absence of published data on the longer-term clinical outcomes of patients after these procedures. Patients and clinicians will therefore have a new source of information to help guide their treatment decisions going forwards.

Dissemination will benefit the provision of health care to vascular surgical patients in the United Kingdom, both those who are considered for lower limb vascular surgery and those who ultimately undergo a procedure.

The outputs will raise awareness of the health outcomes of the patients involved.

Patient presentation and newsletters will ensure the results of the study are communicated to the patient group it relates to. This will enable patients to be better informed about vascular surgery, but also enable them to ask questions and find out more about the condition they have and the treatment they might undergo. It may also serve as a reference to inform discussions with patients around planning future studies.

The benefits to patient care and patient outcomes will be reflected in the annual reports released by the National Vascular Registry. The aggregate data reported will allow measurement of the changes in surgical care for these patients after the dissemination of the outputs and demonstrate how their care and outcomes have improved. it will also be possible to measure the interest and reach of the published outcomes, by tracking metrics such as article downloads and citations.

Finally, this initial analysis may form the foundation of future projects and grant applications to examine this patient group further and potentially develop risk prediction models for these surgeries, as has been done for other vascular surgery (e.g. the British Aneurysm Repair Score, which was developed using the NVR). The timescale for any such work would be in the order of several years as grant applications, research ethics approval and appropriate time for analysis would be required.

Not stated in the register.