Utilising the extensive range of data assets exclusively available to IQVIA, combined with our market leading expertise, we are able to support the NHS meet the challenge of delivering high quality, efficient and cost effective healthcare by identifying actionable insight and realising the value in data.
Learn MoreWe believe in the UK’s potential to become a global leader in the life sciences sector, and our investments announced in the 2018 Second Life Sciences Sector Deal further support the UK Government and the industry’s aim to make that potential a reality.
Learn MoreWe are committed to providing solutions that inform and enable all stakeholders across UK healthcare, to innovate with confidence, maximise opportunities, and, ultimately, drive human health outcomes forward.
Learn More
IQVIA’s Pharmacy Communications offers pharma companies a simple and cost-effective channel to communicate product changes, updates and patient advice directly to patients via pharmacists.
Learn MoreExplore our library of resources spanning the healthcare & life science ecosystem on the latest topics, trends, and market developments that matter to you.
Learn MoreAuthor Affiliations:
IQVIA, The Point, 37 N Wharf Rd, London W2 1AF
To address the urgent need for early identification of Atrial Fibrillation (AF), the SCOPE-AF model was developed by IQVIA using nearly one million non-identified patient records from the UK IQVIA Medical Research Data (IMRD) primary care database. This Machine Learning (ML) based Artificial Intelligence (AI) model was trained to predict the 6-month incidence of AF in individuals aged 50 and over, leveraging routinely collected electronic health record (EHR) data. Key predictors included blood pressure and BMI measurements, prescriptions for antihypertensives and loop diuretics, and records of echocardiograms and electrocardiograms. As a disease detection tool, SCOPE-AF enables proactive profiling of individuals at elevated risk of developing AF, supporting crucial, timely clinical intervention and stroke prevention. The model was prospectively deployed within the IQVIA Healthcare Analytics Network (IHAN), a secure network of GP practices across England. Through IHAN, patients from selected practices were scored using their EHR data, successfully identifying individuals with undiagnosed AF and other arrhythmias who are now receiving appropriate care to reduce their stroke risk. This real-world implementation demonstrates both the scalability and clinical relevance of the approach. Importantly, it highlights how predictive models can be integrated into routine care pathways to support earlier diagnosis and improve patient outcomes. Building on this success, development is already underway for a similar model targeting heart failure, further expanding the potential of machine learning in cardiovascular disease management.
The synergistic use of IMRD and IHAN demonstrates how real-world data can be leveraged to enhance patient outcomes. This enables the deployment of disease detection models like SCOPE-AF, effectively bridging the gap between data-driven research and real-world clinical practice. The result is earlier diagnosis, more targeted interventions, and improved population health management. By helping clinicians identify at-risk patients sooner, these tools support timely, preventative care, particularly for chronic conditions such as AF where earlier intervention make a dramatic difference to patients. For the NHS, it also translates into reduced pressure on services, fewer emergency admissions, and more efficient resource allocation (1). This approach aligns with the NHS’s strategic focus on prevention, personalised care, and value-based healthcare, where outcomes and cost-effectiveness take precedence over volume. IQVIA’s data-driven methodology supports this shift by enabling proactive, evidence-based decision-making at scale (2).
IQVIA Medical Research Data (IMRD) was used to support this innovative AI model. It offers a rich source of longitudinal patient-level data, capturing a wide array of information including demographics, clinical diagnoses, prescriptions, consultations, and socioeconomic and lifestyle factors. The data are collected from routine general practice clinical systems and include details such as year of birth, sex and registration history. Clinical information includes diagnoses, symptoms, referrals, tests and immunisations, alongside other heath-related details such as behavioural factors (e.g. smoking status) and biometric measures such as body mass index (BMI). Prescription records are detailed, covering prescribed medication, dosage, quantity, strength and formulation. Consultations are timestamped and linked to the roles of healthcare professionals involved. This depth and breadth of data make IMRD an invaluable tool for epidemiological research, pharmacovigilance, and health outcomes analysis (3). In the context of AF, it enables researchers and policymakers to track trends in incidence and prevalence, evaluate treatment patterns, assess the effectiveness of interventions, and identify gaps in care, particularly among undiagnosed or high-risk populations (3). By leveraging datasets like IMRD, researchers can generate robust, evidence-based insights to support early detection, and to optimise management pathways with the aim of reducing both the clinical and economic burden of AF across the UK.
Early identification and risk stratification of AF are essential for preventing serious complications such as stroke, heart failure, and other cardiovascular events (4). However, AF often presents without symptoms, making early detection particularly challenging. Timely diagnosis is critical, as it enables clinicians to initiate appropriate interventions, such as anticoagulation or rhythm control, based on an individual’s risk profile (4). Despite its importance, current diagnostic and predictive pathways remain limited. Traditional screening methods and clinical risk scores, while widely used, may lack precision, particularly in diverse or underserved populations (5). When diagnostic precision is low, healthcare providers must screen large numbers of individuals to detect a single case of AF, placing additional strain on already stretched healthcare resources (5). Emerging tools such as biomarkers and advanced imaging techniques offer promise for improving diagnostic accuracy, but their integration into routine clinical practice remains limited due to cost, complexity, and accessibility (5).
Data-driven approaches, particularly those leveraging AI and ML, have the potential to transform AF detection and risk assessment. By analysing large-scale health datasets, these technologies can identify subtle patterns and risk indicators that may be missed by conventional methods (6). Integrating AI-driven tools into clinical workflows could enable more proactive, personalised, and efficient AF management, ultimately improving outcomes while reducing the burden on healthcare systems (6). Given the rising prevalence, clinical burden, and economic impact of AF, access to high-quality, real-world data, such as IMRD, are essential for informing effective healthcare strategies.
Atrial Fibrillation (AF) is the most prevalent cardiac arrhythmia, marked by irregular and often rapid electrical activity in the atria that disrupts coordinated heart contractions and increases the risk of clot formation (7). Clinically, AF is significant due to its strong association with stroke, patients are five times more likely to suffer one, and other cardiovascular complications. Symptoms may include palpitations, fatigue, and dizziness, though many cases are asymptomatic and go undetected until a serious event occurs. Early diagnosis and management, particularly with anticoagulation therapy, are crucial to reducing these risks (7).
In the UK, AF has become a growing public health issue, with over 1.5 million diagnosed cases as of 2023, a 50% rise since 2013, and an estimated 270,000 undiagnosed individuals. This increase is driven by improved awareness, diagnostics, and an ageing population (8). Economically, AF places a heavy burden on the NHS, with hospitalisations as a major cost driver. In 2020, AF-related costs ranged from £1.44 to £2.55 billion, and projections suggest this could rise to between £3.85 and £12.14 billion by 2040 (8). Stroke, a common AF complication, significantly contributes to these costs due to long-term care needs. As emphasised by Health Innovation North West Coast, proactive strategies, such as ML models for early detection, are vital to improving outcomes and reducing financial strain (8).
If you're interested in learning more about our work, exploring collaboration opportunities, or discussing how we’ve used datasets like IMRD to develop disease detection models such as SCOPE-AF, we’d be delighted to connect and explore how this could benefit your organisation. Collaboration across clinical, academic, and industry settings is key to unlocking the full potential of real-world data and driving meaningful improvements in patient care. Please contact us for more information.
Bibliography
1. NHS England Digital [Internet]. [cited 2025 Aug 28]. How data is used to improve health and care. Available from: https://digital.nhs.uk/your-data/how-health-and-care-data-is-used
2. NHS England | Workforce, training and education [Internet]. 2021 [cited 2025 Aug 28]. Data Driven Healthcare in 2030: Transformation Requirements of the NHS Digital Technology and Health Informatics Workforce. Available from: https://www.hee.nhs.uk/our-work/building-our-future-digital-workforce/data-driven-healthcare-2030
3. IQVIA Medical Research Data (IMRD) [Internet]. [cited 2025 Aug 28]. Available from: https://www.iqvia.com/locations/united-kingdom/solutions/life-sciences-industry-solutions/real-world-solutions/iqvia-medical-research-data
4. GOV.UK [Internet]. [cited 2025 Aug 28]. Health matters: preventing cardiovascular disease. Available from: https://www.gov.uk/government/publications/health-matters-preventing-cardiovascular-disease/health-matters-preventing-cardiovascular-disease
5. Atrial fibrillation - UK National Screening Committee (UK NSC) - GOV.UK [Internet]. [cited 2025 Aug 28]. Available from: https://view-health-screening-recommendations.service.gov.uk/atrial-fibrillation
6. Wegner FK, Plagwitz L, Doldi F, Ellermann C, Willy K, Wolfes J, et al. Machine learning in the detection and management of atrial fibrillation. Clin Res Cardiol. 2022 Sep 1;111(9):1010–7.
7. Ko D, Chung MK, Evans PT, Benjamin EJ, Helm RH. Atrial Fibrillation: A Review. JAMA. 2025 Jan 28;333(4):329–42.
8. Burdett P, Lip GYH. Atrial fibrillation in the UK: predicting costs of an emerging epidemic recognizing and forecasting the cost drivers of atrial fibrillation-related costs. Eur Heart J Qual Care Clin Outcomes. 2022 Mar 1;8(2):187–94.
IQVIA Real World Solutions is the proven partner life science companies trust to take on their complex evidence challenges. We enable our customers to generate and disseminate real world evidence to answer crucial questions and meet the needs of their stakeholders.
