Establishing the impact of the national VTE prevention programme on post-operative VTE rates in England

The University of Surrey and the Royal College of General Practitioners (RCGP) are working together as joint data controllers to look at preventing Venous thromboembolism.

The Royal College of General Practitioners rely on Article 6(1)(f); processing is necessary for the purposes of the legitimate interests pursued by a controller, except where such interests are overridden by the interests or fundamental rights and freedoms of the data subject which require protection of personal data, in particular where the data subject is a child. This shall not apply to processing carried out by public authorities in the performance of their tasks. RCGP has conducted a Legitimate Interest Assessment and concluded that that their legitimate purpose through the use of the data is to provide information and analysis on general practice data – both disease and workload.

Project specific: To understand the impact of mandatory venous thromboembolism (VTE) risk assessment on the incidence and outcomes of VTE after surgery.

The benefits to patients is to improve the understand the impact of mandatory venous thromboembolism (VTE) risk assessment on the incidence and outcomes of VTE after surgery (Patient Care)

There are benefits to national bodies through:

• Provision of national surveillance information for Public Health England

• Provision of workload and workforce breakdown (influence policy) for NHS England

• Project specific: To understand the impact of mandatory venous thromboembolism (VTE) risk assessment on the incidence and outcomes of VTE after surgery (Policy implications)

Article 9(2)(j) (processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject)

- The data are required for research purposes in the public interest – meeting the conditions in the DPA 2018 Schedule 1 Part 1 (4).

This study is ongoing and University of Surrey and the Royal College of General Practitioners (RCGP) seek permission to retain the data sets provided under a previous version of this agreement: Hospital Episode Statistics - Admitted Patient Care, linked to Civil Registrations (Deaths).

The processing of the data will help the study to:

• Provide surveillance services based on general practice electronic healthcare records.

• Project specific: data from general practice is required to fill in the gaps in the current understanding of the incidence and outcomes of mandatory VTE risk assessment after surgery as currently much of the existing information is from secondary care. RSC data is required to identify where VTE has occurred once a patient has left hospital. No additional processing outside of what is required is approved by the RSC and all amendments have been made to ensure that data is processed in the least intrusive way possible while still enabling the purpose of the RCGP RSC

The University of Surrey are processing the data under General Data Protection Regulation Article 6(1)(e): (processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller), and, Article 9(2)(j) (processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject).

Data is pseudonymised as close to the source as possible, the RCGP RSC does not hold or process any identifiable personal data. There are no existing relationships that can be identified with the individuals whose data is processed. All data is pseudonymised,

In this application, the University of Surrey will be studying a comparable-sized population of about 3 million individuals from the Royal College of General Practitioners Research & Surveillance Centre (RCGP RSC) database. Over seven years the study team expect to see about 450 VTE events from approximately 78,000 surgical procedures (ie prior to mandatory screening). With 61,506 surgeries before, and 61,506 surgeries after the introduction of VTE guidelines, there is 80% power to detect a 10% reduction in VTE events from 23.7/1000 years to 21.3/1000 years.

The concept to use preventive measures to prevent Venous thromboembolism (VTE, also known as 'blood clots') for specific at-risk groups is well established (Haut et al, 2013). There are significant risks to medications that reduce clotting of the blood and so determining the risk-to-benefit ratio is essential to ensure that prevention is targeted appropriately. The National VTE prevention programme was launched in 2010 with the introduction of mandatory VTE risk assessment of all adults on admission to hospital. This was supported by NICE guidelines.

Where patients are at increased risk of VTE (ie the risk is NOT outweighed by risk factors for bleeding), then NICE recommend mobilisation of the patient as soon as possible, medicines to limit clotting and compression stockings. Patients are also given information of the risk of blood clots and discharge planning includes relaying this information to other care-givers.

The risk of VTE persists for up to 12 months after surgery, and is particularly high in the first three months (Kearon, 2003; Sweetland et al, 2009). This risk was estimated before the era of Enhanced Recovery After Surgery (ERAS) which may change the natural history of the disease. ERAS is a package of care that mean patients stay in hospital is much shorter than it once was. In the modern era, in the UK the length of stay for bariatric surgery is less than three days (Awad et al, 2014) and for thyroidectomy is just two days (Perera et al, 2014), for example. Studies evaluating HES have estimated postoperative VTE rates in varicose vein (Sutton et al, 2012), urological (Dyer et al, 2013) and orthopaedic surgery (Jameson et al, 2010). However, data from hospitals (Hospital Episode Statistics, HES) by itself is limited to capturing in-hospital adverse events and those recorded during readmission. It is evident that the risk of VTE persists well beyond discharge from hospital. For instance, a study by Bouras and colleagues showed that a large proportion of postoperative VTE was detected in primary care (2015). Linkage to primary care electronic health records and mortality data will allow for a more accurate perspective of a patients’ entire postoperative course. Mortality is obviously a key clinical outcome after surgery and would be recorded in hospital-derived data but if it occurs in the community, has been shown to be not well recorded through clinical coding in primary care.

Linkage between NHS Digital data and the primary care record will be made via a pseudonymised NHS number. No patient identifiable information will be seen or used by Apollo Medical Software Solutions (the company that facilitate data extraction at the GP surgeries) or anyone at University of Surrey and the RCGP.

The aim of this study is to examine the impact of mandatory VTE risk assessment (introduced in 2010) on the incidence of VTE after general surgery and major orthopaedic surgery.

Patients undergoing one of twelve general surgical procedures will be chosen. Limiting to twelve operations allows the study to standardise for operative duration, likelihood of postoperative immobilisation etc. These procedures represent the majority of emergency and elective general surgical operations in UK hospitals. In terms of the number of years of data required, it is necessary to have data over such a long period because in the paper by Bouras et al (PLoS ONE 2015) there were 981 VTE events captured within 90 days of surgery, in 168005 procedures, from a background population of ~2.9 million people over 15 years (23.7/1000 patient-years). The period of time that the Bouras study relates to was 1997 to 2012. Importantly, this crosses the introduction of mandatory VTE risk assessment and so the paper cannot describe the effect of mandatory screening.

Orthopaedic sub study

It is hypothesized that an individual’s VTE risk after hip or knee surgery can be modelled with the use of a mathematical prediction model.

Study Objectives: To develop a model that predicts the risk of VTE in patients who undergo total hip arthoplasty (THA, 'hip replacement') or total knee arthroplasty (TKA, 'knee replacement') surgery. This will be based upon data of the clinical characteristics of the individual as well as data of the operation itself and routinely collected hospital biochemical (laboratory) data.

The Research question is therefore: What is the optimal prediction model for VTE risk following THA and TKA surgery?

Expected results and influence in society: Current strategies to prevent blood clots are a one-size fits all -ie for all patients who undergo THA and TKA - these are not optimal because patients vary hugely in their ability to form blood clots. Therefore a new strategy, i.e., advice on an individual basis, is necessary to reduce VTE, bleeding complications and costs. A prediction model should be able to reach a discriminative value (area under the curve) of at least 0.7 with a sensitivity of 75% (in other words, detects at least 75% of those with blood clots) and specificity of 50% (in other words, detect at least half of people who do not have blood clots). Ideally, three risk groups could be identified according to the prediction model; a low- (60% of the total), intermediate-(30%) and high-risk (10%) group. These risk groups could consequently be used to optimize strategies to prevent blood clots (thromboprophylaxis).

For patients in the low-risk group (VTE risk <0.5%), thromboprophylaxis could be limited to in-hospital preventative treatment only, resulting in less costs and less bleeding events. For patients in the intermediate group (0.5-1.0%), current thromboprophylaxis policies (lasing for 2 to 4 weeks) may be sufficient; while patients in the high risk group (>1.0%) could potentially benefit from an extended period (or higher dosage) of thromboprophylaxis.

However, before such a tailored strategy can be implemented in clinical practice, an additional impact-analysis has to be performed that measures the validity of the prediction model, and the usefulness in clinical practice. This current study will form the basis for this approach.

For the orthopaedic sub study in this application, the two most common elective major orthopaedic surgeries (hip and knee replacement) have been chosen. In England and Wales (population 58 million) there are approximately 160,000 total hip and knee replacement procedures performed each year. From the population of 3 million in the RCGP RSC database, it would be expected that about 8300 surgeries occur per year ( in other words, about 17,000 over the two years requested). In patients who undergo total hip arthroplasty (THA) or total knee arthroplasty (TKA), 3.7% and 2.7% of patients will develop symptomatic VTE, respectively, despite use of preventative low-molecular-weight heparin (a drug used to thin the blood). This is considered the minimal data necessary upon which to build a predictive algorithm for postoperative VTE.

Sanofi provide a research grant for this study and have no obligation to provide any other support for the study. University of Surrey is responsible for the initiation, management and conduct of the study. The Parties acknowledge that nothing in the funding agreement is provided as or intended to be an inducement to prescribe, purchase, recommend, use, or dispense any of Sanofi’s or its Affiliates’ products. University of Surrey is performing the study independently of Sanofi. Sanofi will have no control of nor in any way contribute to the conduct of the Study.

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)”

The study will only use and store pseudonymised information extracted by an approved third party provider, Apollo Medical Software Solutions.

Each unique patient will be de-identified using a computer generated patient ID which could only be retraced by staff of the participating GP practices.

The research team at University of Surrey will not view patient identifiable information in any form.

Linkage between NHS Digital and the primary care data will be made via the pseudonymised NHS number.

Apollo generates the hash key which then de-identifies all the patients in the server. This is passed onto the University of Surrey.

University of Surrey will transfer the ‘hash’ algorithm to NHS Digital via Secure Electronic File Transfer (SEFT). The Hash algorithm is a one way encryption and can not be reversed so there is o ability for the pseudonymised data to be re-identified by University of Surrey.

Record level HES data pseudonymised at source using ‘hash’ algorithm downloaded to the Research Group at the University of Surrey for linkage via SEFT.

Pseudonymised record-level HES data will be processed and stored by the Research Group at the University of Surrey.

Patient level databases are held in the database server within the Research Group’s secure network. The Research Group is made up of staff substantially employed by University of Surrey. The Research Group’s dedicated secure network is sited behind a firewall within the University’s network. It is a standalone – independent network, all in-bounded connections are block, but out-bounded connections are allowed. All staff members of the research group working within the team base work from secure workstations or secure laptops with encrypted drive.

All staff members of the Research Group working within the team base work from secure workstations or secure laptops with encrypted drive within the Research Group’s secure network. The secure network is located behind a firewall within the University’s network, all in-bounded connections are blocked, but out-bounded connections are allowed.

The hashed data provided by NHS Digital for this study will be downloaded by the Research Group. The Research Group will not have access to the identifiable data or the SALT key used for encryption.

The University of Surrey will make no attempt to re-identify the data extract provided by NHS Digital under this agreement.

There will be no additional data linkage undertaken with NHS Digital data provided under this agreement that is not already noted in the purpose.

Data will only be accessed and processed by substantive employees of the University of Surrey who have been appropriately trained in data protection and confidentiality. Data provided by NHS Digital has not, and, will not be accessed or processed by any other third parties.

There are four key audiences for this research, these are:

A. patients, the public, and health care practitioners

B. commissioning organisations (such as Clinical Commissioning Groups and NHS England)

C. external statutory organisations (such as Department of Health, NHS Information Centre, NICE)

D. academia

The outputs will be in the form of aggregate data tables, graphs, reports and papers for publication, with any small numbers suppressed (in line with the HES Analysis Guide).

• The university of Surrey will work with the local Academic Health Science Network, who will advise and support routes for dissemination to the public.

• Outputs to the public will be made via University of Surrey, University of Leiden and King’s College Hospital twitter feeds, Facebook and the media offices. Results of the study will be posted on www.clininfo.eu and University of Surrey webpages.

• Publications including Full, Executive Summary and Plain English summary reports of the research will be made in peer review journals and local NHS newsletters. Journals may include, but not limited to: JAMA surgery, BMJ, British Journal of Haematology, Journal of Thrombosis and Haemostasis, Thrombosis research.

• Wherever possible, publication will be made using a Creative Commons Licence. This will allow downloading the report, free of charge. Publications are made available on the University of Surrey library page, academics webpage and Researchgate.net.

• Presentations at national and international haemostasis and perioperative conferences.

• A Report of the study will be written for Sanofi (funding body)

• There is a website for the National VTE prevention programme: vteengland.org.uk where the study will promote the findings.

• Outcomes from the research will be included in future iterations of the guide to achieving CQUIN targets by King’s Thrombosis Centre, in conjunction with VTE Exemplar Centres. http://www.kingsthrombosiscentre.org.uk/kings/Delivering%20the%20CQUIN%20Goal_2ndEdition_LR.pdf

• A co-applicant is the Director of the King's Thrombosis Centre and a Senior Medical Advisor to the National VTE Prevention Programme in England. Through this channel, the research outcomes will influence Department of Health, NICE guidance for thrmboprophylaxis.

Expected Output of Research/Impact

OUTPUTS

1. An understanding of the effect of mandatory VTE risk assessment, introduced in 2010

2. A risk prediction tool for VTE after orthopaedic surgery.

IMPACT

The approach to research and dissemination will:

• Potentially reduce NHS costs through better assessment of VTE risk and through more accurate understanding of thrombosis risk after hospitalisation.

• Provide findings to enhance the current evidence base for quality indicators and commissioning practices enabling commissioners and providers to make evidence based decisions to ensure maximum benefit to patients and the NHS

• Contribute to national debates on the role of VTE thrombo-prophylaxis in driving forward improvements in patient care.

Submission of manuscripts will be targeted for the end of 2022 / Summer 2023

The results are expected to inform the evaluation of the NHS policy on VTE risk assessment and thromboprophylaxis in the surgical populations studied. Understanding the impact of the VTE prevention programme and consequent VTE rates following surgical procedures will identify areas with scope for further improvement. Expected benefits will include length of stay, costs, complication rate after surgery and patient satisfaction. As an example, the orthopaedic substudy will enable VTE risk stratification of patients undergoing joint replacement (currently all are considered high risk). This will enable delivery of thromboprophylaxis only to those at very high risk (anticipated to be 50% of patients). Avoiding thromboprophylaxis in those at low risk will minimise adverse effects such as surgical site bleeding/ooze which predisposes to infection, which can adversely impact patient quality of life immediately post operatively and in the long term. Additionally, this represents a significant cost saving to the NHS. This will help inform best practice and guideline development in the continuum of care for joint replacement, as well as in general surgery.

The data will be made immediately available to the National VTE Programme Board at NHS England (via the co-applicant of this study from the Royal College of General Practitioners) - who is also a Director of the National VTE Exemplar Centres Network.

The study team have developed a predictive equation for postoperative VTE after hip and knee surgery (The orthopaedic substudy). An abstract of this work was accepted for presentation at the International Society on Thrombosis and Haemostasis Annual Conference 2022 'Development and validation of a clinical prediction model to predict 90-day venous thromboembolism risk following total hip and total knee arthroplasty', Banne et al.

Currently, anticoagulation prophylaxis is given to all patients undergoing hip and knee surgery (unless very clear, individualised, medical reasons not to do so). However, anticoagulation can lead to bleeding – both minor and major. It would be preferable therefore to not to have to use prophylactic anticoagulation in low-risk individuals. The risk prediction model could be used to estimate an individual’s risk for VTE following THA/TKA and guide thromboprophylaxis therapy. This will ultimately benefit patients as it would allow anticoagulation prophylaxis to be directed only to those more likely to benefit from it.