Vision / Medicine and wellbeing / Transforming lives by predicting and preventing disease

Medicine and wellbeing

Transforming lives by predicting and preventing disease

The UK’s population is ageing. According to the Office for National Statistics, in 2018 18.3% of people in the UK were 65 years old or over – by 2038, that figure is set to rise to approximately one in every four people.

From a healthcare point of view, this is particularly problematic. Leading causes of death in the UK, such as coronary heart disease, are more common among older people. Identifying those most at risk from such diseases as early as possible will help prevent the disease from manifesting, improving the quality of their lives and reducing the risk of premature death, while focusing healthcare resources.

Professor Carol Coupland uses statistical modelling to address this challenge, developing risk prediction algorithms for identifying individuals at increased risk from particular diseases. They can then be more closely monitored by healthcare professionals and considered for preventative treatments and interventions, leading to fewer diagnoses of disease and, ultimately, improved patient outcomes.

Carol’s work in this area began over a decade ago with Professor Julia Hippisley-Cox. The first algorithm, the QRisk tool, was developed in 2007 and was designed to estimate cardiovascular risk over a ten-year period. The QRisk tool has since been updated, and several new tools have been developed for conditions including diabetes, stroke and fracture. They have also developed a tool to estimate the risks of having different types of cancer based on symptoms and risk factors. This tool was developed to help GPs identify patients at increased risk of having cancer for referral for investigations, with the aim of aiding earlier diagnosis.

The algorithms use statistical modelling to create risk predictions, based on routinely recorded patient data on characteristics, symptoms and risk factors and also incorporate deprivation and ethnicity to help to address health inequalities. The data come from the QResearch primary care database, which contains anonymised health records of a large-scale, representative UK population. The algorithms have been updated over time to account for emerging trends and to incorporate new risk factors.

"It’s rewarding to know that these models are now being used both here and in other countries to help identify people who will benefit most from treatments and lifestyle intervention."
Professor Carol Coupland

Since its launch, QRisk has been incorporated into NHS Health Checks, a national programme for adults in England aged 40 to 74 that aims to prevent diseases such as heart disease, stroke and diabetes. It is also used by Public Health England on its NHS One You website to estimate heart age, with the goal of increasing people’s awareness of their heart health and encouraging them to make simple lifestyle changes to improve it.

In 2014, QRisk2, the updated algorithm, was recommended by the NICE Cardiovascular disease guideline. This means the tool is now widely used across England to guide treatment and lifestyle changes to prevent cardiovascular disease.

For Professor Coupland, it is this impact that means most to her. “It’s rewarding to know that these models are now being used both here and in other countries to help identify people who will benefit most from treatments and lifestyle interventions.”

She hopes the other risk prediction algorithms can make a similar impact. All of them are publicly available, and the majority have been integrated into EMIS, which supplies computer systems to over 55% of GP practices in England and covers over 30 million people across the country. Several of the algorithms have also received recommendations in national NICE guidelines. It is hoped this will help GPs make earlier diagnoses and work with high-risk individuals to reduce their chances of developing illnesses such as diabetes, stroke, kidney and heart disease.

Through the power of algorithms, health care professionals can make earlier interventions and their patients can better understand and reduce risks to their health, transforming lives and potentially saving many millions of pounds.

If you would like to learn more about Carol’s research, you can find out more via the links below:

Development and validation of risk prediction algorithms to estimate future risk of common cancers in men and women: prospective cohort study

Development and validation of QRISK3 risk prediction algorithms to estimate future risk of cardiovascular disease: prospective cohort study

Development and validation of QDiabetes-2018 risk prediction algorithm to estimate future risk of type 2 diabetes: cohort study

Carol Coupland

Carol Coupland is Professor of Medical Statistics in Primary Care in the Faculty of Medicine and Health Sciences.

More impact case studies

Creating spaces for young people’s voices in mental health research

Read More

Time to do things differently

Read More

Child and adolescent mental health in a post-lockdown world: a ticking time bomb?

Read More