Skip to content
This website uses cookies to help us understand the way visitors use our website. We can't identify you with them and we don't share the data with anyone else. If you click Reject we will set a single cookie to remember your preference. Find out more in our privacy policy.

CVDPREVENT – Supporting primary care to prevent heart attacks and strokes at scale

21 September 2020 | Dr Matt Kearney
The national audit for cardiovascular disease will include six high risk conditions with extracts of routinely recorded general practice data being monitored and managed for reducing impact on patients. By Dr Matt Kearney and Lorraine Oldridge.

CVDprevent is the new national audit for primary care in England. First reporting is expected in spring next year and it promises to be a game changer for professionally led quality improvement in primary care.

Six high risk conditions for cardiovascular disease are included in the audit: atrial fibrillation, high blood pressure, high cholesterol, diabetes, pre-diabetes, and chronic kidney disease. These conditions are major causes of CVD.

For example, high blood pressure accounts for half of all heart attacks and strokes, having atrial fibrillation makes it five times more likely that you will have a stroke, and in diabetes CVD is the leading cause of morbidity and premature mortality.

However, although treatment in these conditions is highly effective at preventing cardiovascular disease, late diagnosis and suboptimal treatment are very common, with substantial variation across the country.

CVDprevent will utilise rolling three monthly extracts of routinely recorded general practice data, providing detailed insight into the diagnosis, investigation, and management of patients at risk of cardiovascular events.

The data will be extracted (with no additional burden for GPs) for three cohorts: patients who have one of the six high risk conditions, patients who have established cardiovascular disease, and patients not in the first two cohorts but whose records contain entries indicating that they may have an undiagnosed high-risk condition.

The extracts will include diagnostic codes, recording of risk factors such as smoking and alcohol, physical measurements such as blood pressure and body mass index, blood tests such as kidney function and cholesterol, as well as drug treatment and lifestyle interventions.

CVDprevent will utilise rolling three monthly extracts of routinely recorded general practice data, providing detailed insight into the diagnosis, investigation, and management of patients at risk of cardiovascular events

Why is the audit needed? Because without data we cannot know how well we are treating our patients and what we could be doing better. And what we know of current performance from QOF and other datasets shows that there is much room for improvement in CVD prevention.

This is partly because identification and management of these often asymptomatic conditions is difficult to do well in complex modern general practice where consultations are brief and multimorbidity and multiple priorities is the norm.

For example, around four in 10 people with high blood pressure are currently undiagnosed, and when diagnosed around four in 10 are not treated to target; in people with known atrial fibrillation who go on to suffer a stroke, almost 50 per cent have not been treated with anticoagulants prior to their stroke; and in patients who have already suffered a heart attack or stroke around half are on no treatment or suboptimal treatment to lower their cholesterol. And behind these averages, widespread local and national variation contributes significantly to health inequalities.

Reducing health inequalities

CVDprevent has been developed by a broad partnership that includes the Royal College of GPs, British Heart Foundation, National Institute for Health and Care Excellence, NHS Digital, Public Health England, NHS England and NHS Improvement.

Strong clinical and programme leadership along with active professional engagement has been key, along with initial investment from the British Heart Foundation to demonstrate that the audit was feasible and would add real value.

Funding has now been secured from NHSE/I via the NHS long-term plan with delivery by a partnership: NHS Digital responsible for the data extraction via their GP Extraction Service; a provider to be appointed by the Healthcare Quality Improvement Partnership to provide the strategic oversight and clinical leadership, and the National Cardiovascular Intelligence Network in PHE working with the provider to deliver the analysis and reporting.

CVDprevent will provide timely aggregate data at practice, primary care network, clinical commissioning groups and integrated care system level. The data will be stratified by ethnicity, deprivation, severe mental illness and learning disability. It will allow local systems to clearly identify the gaps, inequalities and opportunities for improvement. It will show networks and CCGs where to focus their energies to prevent heart attacks and strokes at scale in their populations and to reduce health inequalities.

For example in general practice we will be able to see how many people in our local communities have high blood pressure that is not controlled to target, atrial fibrillation that is not treated with anticoagulants, raised cholesterol with high 10 year risk score that is not being reduced by statins, or chronic kidney disease that has not been coded and managed appropriately.

The impact of this on population health could be substantial. Based on modelling of treatment effect on outcomes and realistic treatment aspirations for atrial fibrillation, high blood pressure and high cholesterol, the NHS long-term plan has set a national ambition to prevent 15,000 heart attacks, strokes and cases of dementia every year on average for the next 10 years. This translates into substantial numbers of cardiovascular events prevented in every ICS, CCG and Primary Care Network.

Audit can also alert us when preventive medications may be causing harm or questionable benefit. So in addition to identifying under diagnosis and under treatment of the high risk conditions, CVDprevent will also show where people are at risk of over treatment – for example through over rigorous blood pressure control in people who are frail or on multiple drugs, or through use of preventive therapies near the end of life.

Of course, good quality data is essential but not sufficient to drive improvement. Other critical elements are clinical leadership and system support to do things differently. CVDprevent has been developed and will be delivered with strong primary care leadership and close engagement with the profession.

Long-term condition frameworks

The NHS long-term plan and the new GP contract will bring the structural and financial support for new models of care, for example with the new directed enhanced service contract for primary care networks expected in April 2021. This will specifically resource practices working together in PCNs and using the newly employed clinical pharmacists and other staff to target improvement in the high-risk conditions for CVD particularly high blood pressure, atrial fibrillation and high cholesterol.

Recognising the need for very practical support to embed innovation and quality improvement, UCLPartners has developed a series of long-term condition frameworks for cardiovascular and other conditions that will help practices deliver high quality proactive care for patients in the very changed world of primary care post covid.

The frameworks include search and stratification tools, pathways that support the shift to remote care; training and resources for wider workforce (such as pharmacists and health care assistants) to optimise monitoring at home and self-management; and digital technologies to support the transformation.

These frameworks are now being implemented in the North East London and North Central London ICSs, and uptake is also spreading more widely across the country. The frameworks with search tools and other free resources are available on here.

This blog was first published in the HSJ